Tutorial 3: Training deeptb-sk model for Silicon#
Introduction#
DeePTB is a method that uses deep learning to accelerate first-principles electronic structure simulations.
Version Features#
v1: Constructed tight-binding (TB) models with first-principles accuracy (DeePTB-SK)
v2.0-2.1: Added E3 equivariant networks to represent single-electron operators (Hamiltonian, density matrix, and overlap matrix) (DeePTB-E3)
v2.2: Incorporated built-in SK empirical parameters covering commonly used elements across the periodic table
Through these capabilities, DeePTB provides multiple approaches to accelerate electronic structure simulations of materials.
Learning Objectives#
This tutorial mainly introduces the basic operations of constructing TB models using the DeePTB-SK module.
Reading this tutorial will help you:
Familiarize yourself with the training process of DeePTB models
Obtain a complete DeePTB model for silicon crystal with high accuracy
Familiarize yourself with the usage of DeePTB property calculation module
Method Practice #
import os
os.chdir("/root/soft/DeePTB/examples/silicon/tutorial_v2.2")
---------------------------------------------------------------------------
PermissionError Traceback (most recent call last)
Cell In[1], line 2
1 import os
----> 2 os.chdir("/root/soft/DeePTB/examples/silicon/tutorial_v2.2")
PermissionError: [Errno 13] Permission denied: '/root/soft/DeePTB/examples/silicon/tutorial_v2.2'
1. data preparation #
The data used to train the model and plot the verification data is in the data folder:
deeptb/examples/silicon/data/
|-- kpath.0 # train data of primary cell. (k-path bands)
|-- kpathmd25.0 # train data of 10 MD snapshots at T=25K (k-path bands)
|-- kpathmd100.0 # train data of 10 MD snapshots at T=100K (k-path bands)
|-- kpathmd300.0 # train data of 10 MD snapshots at T=300K (k-path bands)
|-- kpt.0 # kmesh samples of primary cell (k-mesh bands)
|-- kpath_spk.0
|-- silicon.vasp # structure of primary cell
The meaning of the datasets in this folder is as follows:
kpath.0: Band data of the primitive cellkpathmd25.0: Band data of 10 MD snapshots at 25Kkpathmd100.0: Band data of 10 MD snapshots at 100Kkpathmd300.0: Band data of 10 MD snapshots at 300Kkpt.0: K-point mesh sampling data of the primitive cellsilicon.vasp: Structure data of the primitive cellkpath_spk.0: Band data of the primitive cell, spare k points.
Each dataset contains DeePTB data files, such as kpath.0:
deeptb/examples/silicon/data/kpath.0/
-- info.json # defining the training objective and edge cutoff of atomic data
-- eigenvalues.npy # numpy array of shape [num_frame, num_kpoint, num_band]
-- kpoints.npy # numpy array of shape [num_kpoint, 3]
-- xdat.traj # ase trajectory file with num_frame
Where:
info.json: The filename of this file is fixed and provides information about the dataset loaded in the DeePTB model.
{
"nframes": 1,
"natoms": 2,
"pos_type": "ase",
"pbc": true,
"bandinfo": {
"band_min": 0,
"band_max": 6,
"emin": null,
"emax": null
}
}
nframesmarks the number of trajectory snapshots,natomsmarks the number of atoms in each snapshot,pos_typemarks the coordinate type, andpbcmarks whether periodic boundary conditions are applied. Thebandinfocontains information about the band window, which can be set according to the needs of the user. The band window information can be sorted by band index or divided according to energy size. Note that the value of emin is relative to min(eig[band_min]). Taking min(eig[band_min]) as 0 point.
eigenvalues.npy: This file has a fixed name and contains the original band data, with shape[n_frames, nkpoints, nbands]kpoints.npy: This file has a fixed name and contains the original k-point data, with shape[nkpoints, 3]xdat.traj: This file can have any prefix, but must have the fixed suffix “.traj”, and contains trajectory structure data that can be read using ase.
In addition to providing ase trajectory data with the
.trajsuffix, you can also choose to provide three text files:positions.dat,cell.dat, andatomic_numbers.datto load the trajectory. The coordinate type provided by the user is specified ininfo.json: it can be fractional coordinatesfrac, actual coordinatescart, or ase trajectory filease.
**2. Model Training ** #
2.1 Extract Initial Experience sktb Model#
Extract the initial experience parameters from the built-in experience parameters. Here is the initial experience parameter model for Si. For details, please refer to the tutorial-1.
Prepare an input file sk_in.json for parameter extraction, as follows:
{
"common_options": {
"basis": {
"Si": ["3s","3p","d*"]
}
}
}
note: the basis can also be [‘s’,‘p’,‘d’].
First, run the command to generate an initial sk model:
dptb esk sk_in.json -m poly4
After running, you can see a sktb.json model file.
!dptb esk sk_in.json -m poly4
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO Extracting empirical SK parameters for Si
DEEPTB INFO dtype is not provided in the input json, set to the value torch.float32 in model ckpt.
DEEPTB INFO device is not provided in the input json, set to the value cpu in model ckpt.
DEEPTB INFO overlap is not provided in the input json, set to the value True in model ckpt.
DEEPTB INFO Empirical SK parameters are saved in ./sktb.json
DEEPTB INFO If you want to further train the model, please use `dptb config` command to generate input template.
We can compare the band structure of the initial model with DFT results.
!dptb run band.json -i sktb.json -o band -stu ../data/silicon.vasp
# display the band plot:
from IPython.display import Image, display
import os
image_path = f'./band/results/band.png'
display(Image(filename=image_path,width=400))
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
DEEPTB INFO KPOINTS klist: 302 kpoints
DEEPTB INFO The eigenvalues are already in data. will use them.
DEEPTB INFO Calculating Fermi energy in the case of spin-degeneracy.
DEEPTB INFO Fermi energy converged after 19 iterations.
DEEPTB INFO q_cal: 8.000000000143103, total_electrons: 8.0, diff q: 1.4310330698208418e-10
DEEPTB INFO Estimated E_fermi: -3.666769775062834 based on the valence electrons setting nel_atom : {'Si': 4} .
DEEPTB INFO Using input Fermi energy: -4.7220 eV (estimated: -3.6668 eV)
Figure(640x560)
DEEPTB INFO band calculation successfully completed.
2.2 Generate Training Input Control Parameters#
The developers have provided a template for generating training sk models to facilitate user use. The command to obtain the template is as follows:
dptb config ./ -tr -sk -m ./sktb.json
Note: Here I loaded the sktb.json model file generated in the previous step, so some parameters will be set according to the model. After running the above command, you will get ./input_templete.json.
Note: the template cannot be used directly and needs to be modified according to the situation. For example, the parameters in train_options and data_options. You should also ensure that the paths and options are correctly set for your specific use case.
!dptb config ./ -tr -sk -m ./sktb.json
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO Writing full config for train_SK to ./input_templete.json
Donot forget to modify the data_options and train_options parameters in the input file.
We suggest copying the template and modifying the copied input parameter file. For example, we have already prepared the first training parameter file input_1.json in the case folder, which can be used for training the model for perfect crystal Si.
2.3.1 Training the DeePTB-SK model for perfect lattice#
# v100 1m45s
!dptb train input_1.json -i sktb.json -o nnsk1 lattice
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO ------------------------------------------------------------------
DEEPTB INFO Cutoff options:
DEEPTB INFO
DEEPTB INFO r_max : {'Si-Si': 6.24}
DEEPTB INFO er_max : None
DEEPTB INFO oer_max : None
DEEPTB INFO ------------------------------------------------------------------
Processing dataset...
Loading data: 0%| | 0/1 [00:00<?, ?it/s]/root/dptb_venv/lib/python3.10/site-packages/dptb/data/dataset/_default_dataset.py:254: UserWarning: The given NumPy array is not writable, and PyTorch does not support non-writable tensors. This means writing to this tensor will result in undefined behavior. You may want to copy the array to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at ../torch/csrc/utils/tensor_numpy.cpp:206.)
kwargs[AtomicDataDict.KPOINT_KEY] = torch.as_tensor(self.data[AtomicDataDict.KPOINT_KEY][frame], dtype=torch.get_default_dtype())
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
Loading data: 100%|███████████████████████████████| 1/1 [00:00<00:00, 70.45it/s]
DEEPTB INFO Loaded data: Batch(atomic_numbers=[2, 1], batch=[2], bwindow=[1, 2], cell=[1, 3, 3], edge_cell_shift=[92, 3], edge_features=[92, 1], edge_index=[2, 92], edge_overlap=[92, 1], eigenvalue=nested, kpoint=nested, node_features=[2, 1], node_overlap=[2, 1], node_soc=[2, 1], node_soc_switch=[1, 1], pbc=[1, 3], pos=[2, 3], ptr=[2])
processed data size: ~0.01 MB
DEEPTB INFO Cached processed data to disk
Done!
DEEPTB WARNING The cutoffs in data and model are not checked. be careful!
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO iteration:1 train_loss: 1.713102 (0.513931) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter1
DEEPTB INFO Epoch 1 summary: train_loss: 1.713102
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep1
DEEPTB INFO iteration:2 train_loss: 0.789605 (0.596633) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter2
DEEPTB INFO Epoch 2 summary: train_loss: 0.789605
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep2
DEEPTB INFO iteration:3 train_loss: 0.344747 (0.521067) lr: 0.00998
DEEPTB INFO checkpoint saved as nnsk.iter3
DEEPTB INFO Epoch 3 summary: train_loss: 0.344747
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep3
DEEPTB INFO iteration:4 train_loss: 0.299569 (0.454618) lr: 0.00997
DEEPTB INFO checkpoint saved as nnsk.iter4
DEEPTB INFO Epoch 4 summary: train_loss: 0.299569
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep4
DEEPTB INFO iteration:5 train_loss: 0.454804 (0.454674) lr: 0.00996
DEEPTB INFO checkpoint saved as nnsk.iter5
DEEPTB INFO Epoch 5 summary: train_loss: 0.454804
---------------------------------------------------------------------------------
DEEPTB INFO iteration:6 train_loss: 0.573384 (0.490287) lr: 0.00995
DEEPTB INFO checkpoint saved as nnsk.iter6
DEEPTB INFO Epoch 6 summary: train_loss: 0.573384
---------------------------------------------------------------------------------
DEEPTB INFO iteration:7 train_loss: 0.560850 (0.511456) lr: 0.00994
DEEPTB INFO checkpoint saved as nnsk.iter7
DEEPTB INFO Epoch 7 summary: train_loss: 0.560850
---------------------------------------------------------------------------------
DEEPTB INFO iteration:8 train_loss: 0.454255 (0.494295) lr: 0.00993
DEEPTB INFO checkpoint saved as nnsk.iter8
DEEPTB INFO Epoch 8 summary: train_loss: 0.454255
---------------------------------------------------------------------------------
DEEPTB INFO iteration:9 train_loss: 0.327400 (0.444227) lr: 0.00992
DEEPTB INFO checkpoint saved as nnsk.iter9
DEEPTB INFO Epoch 9 summary: train_loss: 0.327400
---------------------------------------------------------------------------------
DEEPTB INFO iteration:10 train_loss: 0.235821 (0.381705) lr: 0.00991
DEEPTB INFO checkpoint saved as nnsk.iter10
DEEPTB INFO Epoch 10 summary: train_loss: 0.235821
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep10
DEEPTB INFO iteration:11 train_loss: 0.203891 (0.328361) lr: 0.0099
DEEPTB INFO checkpoint saved as nnsk.iter11
DEEPTB INFO Epoch 11 summary: train_loss: 0.203891
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep11
DEEPTB INFO iteration:12 train_loss: 0.223004 (0.296754) lr: 0.009891
DEEPTB INFO checkpoint saved as nnsk.iter12
DEEPTB INFO Epoch 12 summary: train_loss: 0.223004
---------------------------------------------------------------------------------
DEEPTB INFO iteration:13 train_loss: 0.262044 (0.286341) lr: 0.009881
DEEPTB INFO checkpoint saved as nnsk.iter13
DEEPTB INFO Epoch 13 summary: train_loss: 0.262044
---------------------------------------------------------------------------------
DEEPTB INFO iteration:14 train_loss: 0.288178 (0.286892) lr: 0.009871
DEEPTB INFO checkpoint saved as nnsk.iter14
DEEPTB INFO Epoch 14 summary: train_loss: 0.288178
---------------------------------------------------------------------------------
DEEPTB INFO iteration:15 train_loss: 0.282941 (0.285707) lr: 0.009861
DEEPTB INFO checkpoint saved as nnsk.iter15
DEEPTB INFO Epoch 15 summary: train_loss: 0.282941
---------------------------------------------------------------------------------
DEEPTB INFO iteration:16 train_loss: 0.247868 (0.274355) lr: 0.009851
DEEPTB INFO checkpoint saved as nnsk.iter16
DEEPTB INFO Epoch 16 summary: train_loss: 0.247868
---------------------------------------------------------------------------------
DEEPTB INFO iteration:17 train_loss: 0.196791 (0.251086) lr: 0.009841
DEEPTB INFO checkpoint saved as nnsk.iter17
DEEPTB INFO Epoch 17 summary: train_loss: 0.196791
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep17
DEEPTB INFO iteration:18 train_loss: 0.149259 (0.220538) lr: 0.009831
DEEPTB INFO checkpoint saved as nnsk.iter18
DEEPTB INFO Epoch 18 summary: train_loss: 0.149259
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep18
DEEPTB INFO iteration:19 train_loss: 0.119433 (0.190206) lr: 0.009822
DEEPTB INFO checkpoint saved as nnsk.iter19
DEEPTB INFO Epoch 19 summary: train_loss: 0.119433
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep19
DEEPTB INFO iteration:20 train_loss: 0.112715 (0.166959) lr: 0.009812
DEEPTB INFO checkpoint saved as nnsk.iter20
DEEPTB INFO Epoch 20 summary: train_loss: 0.112715
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep20
DEEPTB INFO iteration:21 train_loss: 0.123056 (0.153788) lr: 0.009802
DEEPTB INFO checkpoint saved as nnsk.iter21
DEEPTB INFO Epoch 21 summary: train_loss: 0.123056
---------------------------------------------------------------------------------
DEEPTB INFO iteration:22 train_loss: 0.138459 (0.149189) lr: 0.009792
DEEPTB INFO checkpoint saved as nnsk.iter22
DEEPTB INFO Epoch 22 summary: train_loss: 0.138459
---------------------------------------------------------------------------------
DEEPTB INFO iteration:23 train_loss: 0.147752 (0.148758) lr: 0.009782
DEEPTB INFO checkpoint saved as nnsk.iter23
DEEPTB INFO Epoch 23 summary: train_loss: 0.147752
---------------------------------------------------------------------------------
DEEPTB INFO iteration:24 train_loss: 0.145200 (0.147691) lr: 0.009773
DEEPTB INFO checkpoint saved as nnsk.iter24
DEEPTB INFO Epoch 24 summary: train_loss: 0.145200
---------------------------------------------------------------------------------
DEEPTB INFO iteration:25 train_loss: 0.131093 (0.142711) lr: 0.009763
DEEPTB INFO checkpoint saved as nnsk.iter25
DEEPTB INFO Epoch 25 summary: train_loss: 0.131093
---------------------------------------------------------------------------------
DEEPTB INFO iteration:26 train_loss: 0.111269 (0.133279) lr: 0.009753
DEEPTB INFO checkpoint saved as nnsk.iter26
DEEPTB INFO Epoch 26 summary: train_loss: 0.111269
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep26
DEEPTB INFO iteration:27 train_loss: 0.092943 (0.121178) lr: 0.009743
DEEPTB INFO checkpoint saved as nnsk.iter27
DEEPTB INFO Epoch 27 summary: train_loss: 0.092943
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep27
DEEPTB INFO iteration:28 train_loss: 0.082383 (0.109540) lr: 0.009733
DEEPTB INFO checkpoint saved as nnsk.iter28
DEEPTB INFO Epoch 28 summary: train_loss: 0.082383
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep28
DEEPTB INFO iteration:29 train_loss: 0.080076 (0.100700) lr: 0.009724
DEEPTB INFO checkpoint saved as nnsk.iter29
DEEPTB INFO Epoch 29 summary: train_loss: 0.080076
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep29
DEEPTB INFO iteration:30 train_loss: 0.082685 (0.095296) lr: 0.009714
DEEPTB INFO checkpoint saved as nnsk.iter30
DEEPTB INFO Epoch 30 summary: train_loss: 0.082685
---------------------------------------------------------------------------------
DEEPTB INFO iteration:31 train_loss: 0.085412 (0.092331) lr: 0.009704
DEEPTB INFO checkpoint saved as nnsk.iter31
DEEPTB INFO Epoch 31 summary: train_loss: 0.085412
---------------------------------------------------------------------------------
DEEPTB INFO iteration:32 train_loss: 0.084484 (0.089977) lr: 0.009695
DEEPTB INFO checkpoint saved as nnsk.iter32
DEEPTB INFO Epoch 32 summary: train_loss: 0.084484
---------------------------------------------------------------------------------
DEEPTB INFO iteration:33 train_loss: 0.078902 (0.086654) lr: 0.009685
DEEPTB INFO checkpoint saved as nnsk.iter33
DEEPTB INFO Epoch 33 summary: train_loss: 0.078902
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep33
DEEPTB INFO iteration:34 train_loss: 0.070558 (0.081826) lr: 0.009675
DEEPTB INFO checkpoint saved as nnsk.iter34
DEEPTB INFO Epoch 34 summary: train_loss: 0.070558
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep34
DEEPTB INFO iteration:35 train_loss: 0.062745 (0.076101) lr: 0.009666
DEEPTB INFO checkpoint saved as nnsk.iter35
DEEPTB INFO Epoch 35 summary: train_loss: 0.062745
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep35
DEEPTB INFO iteration:36 train_loss: 0.058193 (0.070729) lr: 0.009656
DEEPTB INFO checkpoint saved as nnsk.iter36
DEEPTB INFO Epoch 36 summary: train_loss: 0.058193
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep36
DEEPTB INFO iteration:37 train_loss: 0.057664 (0.066809) lr: 0.009646
DEEPTB INFO checkpoint saved as nnsk.iter37
DEEPTB INFO Epoch 37 summary: train_loss: 0.057664
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep37
DEEPTB INFO iteration:38 train_loss: 0.059696 (0.064675) lr: 0.009637
DEEPTB INFO checkpoint saved as nnsk.iter38
DEEPTB INFO Epoch 38 summary: train_loss: 0.059696
---------------------------------------------------------------------------------
DEEPTB INFO iteration:39 train_loss: 0.061717 (0.063788) lr: 0.009627
DEEPTB INFO checkpoint saved as nnsk.iter39
DEEPTB INFO Epoch 39 summary: train_loss: 0.061717
---------------------------------------------------------------------------------
DEEPTB INFO iteration:40 train_loss: 0.061624 (0.063139) lr: 0.009617
DEEPTB INFO checkpoint saved as nnsk.iter40
DEEPTB INFO Epoch 40 summary: train_loss: 0.061624
---------------------------------------------------------------------------------
DEEPTB INFO iteration:41 train_loss: 0.058877 (0.061860) lr: 0.009608
DEEPTB INFO checkpoint saved as nnsk.iter41
DEEPTB INFO Epoch 41 summary: train_loss: 0.058877
---------------------------------------------------------------------------------
DEEPTB INFO iteration:42 train_loss: 0.054545 (0.059666) lr: 0.009598
DEEPTB INFO checkpoint saved as nnsk.iter42
DEEPTB INFO Epoch 42 summary: train_loss: 0.054545
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep42
DEEPTB INFO iteration:43 train_loss: 0.050468 (0.056906) lr: 0.009588
DEEPTB INFO checkpoint saved as nnsk.iter43
DEEPTB INFO Epoch 43 summary: train_loss: 0.050468
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep43
DEEPTB INFO iteration:44 train_loss: 0.048096 (0.054263) lr: 0.009579
DEEPTB INFO checkpoint saved as nnsk.iter44
DEEPTB INFO Epoch 44 summary: train_loss: 0.048096
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep44
DEEPTB INFO iteration:45 train_loss: 0.047749 (0.052309) lr: 0.009569
DEEPTB INFO checkpoint saved as nnsk.iter45
DEEPTB INFO Epoch 45 summary: train_loss: 0.047749
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep45
DEEPTB INFO iteration:46 train_loss: 0.048553 (0.051182) lr: 0.00956
DEEPTB INFO checkpoint saved as nnsk.iter46
DEEPTB INFO Epoch 46 summary: train_loss: 0.048553
---------------------------------------------------------------------------------
DEEPTB INFO iteration:47 train_loss: 0.049181 (0.050582) lr: 0.00955
DEEPTB INFO checkpoint saved as nnsk.iter47
DEEPTB INFO Epoch 47 summary: train_loss: 0.049181
---------------------------------------------------------------------------------
DEEPTB INFO iteration:48 train_loss: 0.048701 (0.050018) lr: 0.009541
DEEPTB INFO checkpoint saved as nnsk.iter48
DEEPTB INFO Epoch 48 summary: train_loss: 0.048701
---------------------------------------------------------------------------------
DEEPTB INFO iteration:49 train_loss: 0.047034 (0.049122) lr: 0.009531
DEEPTB INFO checkpoint saved as nnsk.iter49
DEEPTB INFO Epoch 49 summary: train_loss: 0.047034
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep49
DEEPTB INFO iteration:50 train_loss: 0.044856 (0.047842) lr: 0.009522
DEEPTB INFO checkpoint saved as nnsk.iter50
DEEPTB INFO Epoch 50 summary: train_loss: 0.044856
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep50
DEEPTB INFO iteration:51 train_loss: 0.043061 (0.046408) lr: 0.009512
DEEPTB INFO checkpoint saved as nnsk.iter51
DEEPTB INFO Epoch 51 summary: train_loss: 0.043061
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep51
DEEPTB INFO iteration:52 train_loss: 0.042162 (0.045134) lr: 0.009503
DEEPTB INFO checkpoint saved as nnsk.iter52
DEEPTB INFO Epoch 52 summary: train_loss: 0.042162
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep52
DEEPTB INFO iteration:53 train_loss: 0.042025 (0.044202) lr: 0.009493
DEEPTB INFO checkpoint saved as nnsk.iter53
DEEPTB INFO Epoch 53 summary: train_loss: 0.042025
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep53
DEEPTB INFO iteration:54 train_loss: 0.042054 (0.043557) lr: 0.009484
DEEPTB INFO checkpoint saved as nnsk.iter54
DEEPTB INFO Epoch 54 summary: train_loss: 0.042054
---------------------------------------------------------------------------------
DEEPTB INFO iteration:55 train_loss: 0.041654 (0.042986) lr: 0.009474
DEEPTB INFO checkpoint saved as nnsk.iter55
DEEPTB INFO Epoch 55 summary: train_loss: 0.041654
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep55
DEEPTB INFO iteration:56 train_loss: 0.040620 (0.042276) lr: 0.009465
DEEPTB INFO checkpoint saved as nnsk.iter56
DEEPTB INFO Epoch 56 summary: train_loss: 0.040620
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep56
DEEPTB INFO iteration:57 train_loss: 0.039222 (0.041360) lr: 0.009455
DEEPTB INFO checkpoint saved as nnsk.iter57
DEEPTB INFO Epoch 57 summary: train_loss: 0.039222
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep57
DEEPTB INFO iteration:58 train_loss: 0.037962 (0.040341) lr: 0.009446
DEEPTB INFO checkpoint saved as nnsk.iter58
DEEPTB INFO Epoch 58 summary: train_loss: 0.037962
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep58
DEEPTB INFO iteration:59 train_loss: 0.037204 (0.039400) lr: 0.009436
DEEPTB INFO checkpoint saved as nnsk.iter59
DEEPTB INFO Epoch 59 summary: train_loss: 0.037204
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep59
DEEPTB INFO iteration:60 train_loss: 0.036955 (0.038666) lr: 0.009427
DEEPTB INFO checkpoint saved as nnsk.iter60
DEEPTB INFO Epoch 60 summary: train_loss: 0.036955
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep60
DEEPTB INFO iteration:61 train_loss: 0.036909 (0.038139) lr: 0.009417
DEEPTB INFO checkpoint saved as nnsk.iter61
DEEPTB INFO Epoch 61 summary: train_loss: 0.036909
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep61
DEEPTB INFO iteration:62 train_loss: 0.036700 (0.037707) lr: 0.009408
DEEPTB INFO checkpoint saved as nnsk.iter62
DEEPTB INFO Epoch 62 summary: train_loss: 0.036700
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep62
DEEPTB INFO iteration:63 train_loss: 0.036143 (0.037238) lr: 0.009399
DEEPTB INFO checkpoint saved as nnsk.iter63
DEEPTB INFO Epoch 63 summary: train_loss: 0.036143
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep63
DEEPTB INFO iteration:64 train_loss: 0.035332 (0.036666) lr: 0.009389
DEEPTB INFO checkpoint saved as nnsk.iter64
DEEPTB INFO Epoch 64 summary: train_loss: 0.035332
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep64
DEEPTB INFO iteration:65 train_loss: 0.034522 (0.036023) lr: 0.00938
DEEPTB INFO checkpoint saved as nnsk.iter65
DEEPTB INFO Epoch 65 summary: train_loss: 0.034522
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep65
DEEPTB INFO iteration:66 train_loss: 0.033935 (0.035397) lr: 0.00937
DEEPTB INFO checkpoint saved as nnsk.iter66
DEEPTB INFO Epoch 66 summary: train_loss: 0.033935
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep66
DEEPTB INFO iteration:67 train_loss: 0.033608 (0.034860) lr: 0.009361
DEEPTB INFO checkpoint saved as nnsk.iter67
DEEPTB INFO Epoch 67 summary: train_loss: 0.033608
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep67
DEEPTB INFO iteration:68 train_loss: 0.033400 (0.034422) lr: 0.009352
DEEPTB INFO checkpoint saved as nnsk.iter68
DEEPTB INFO Epoch 68 summary: train_loss: 0.033400
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep68
DEEPTB INFO iteration:69 train_loss: 0.033135 (0.034036) lr: 0.009342
DEEPTB INFO checkpoint saved as nnsk.iter69
DEEPTB INFO Epoch 69 summary: train_loss: 0.033135
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep69
DEEPTB INFO iteration:70 train_loss: 0.032724 (0.033642) lr: 0.009333
DEEPTB INFO checkpoint saved as nnsk.iter70
DEEPTB INFO Epoch 70 summary: train_loss: 0.032724
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep70
DEEPTB INFO iteration:71 train_loss: 0.032213 (0.033214) lr: 0.009324
DEEPTB INFO checkpoint saved as nnsk.iter71
DEEPTB INFO Epoch 71 summary: train_loss: 0.032213
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep71
DEEPTB INFO iteration:72 train_loss: 0.031730 (0.032768) lr: 0.009314
DEEPTB INFO checkpoint saved as nnsk.iter72
DEEPTB INFO Epoch 72 summary: train_loss: 0.031730
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep72
DEEPTB INFO iteration:73 train_loss: 0.031372 (0.032349) lr: 0.009305
DEEPTB INFO checkpoint saved as nnsk.iter73
DEEPTB INFO Epoch 73 summary: train_loss: 0.031372
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep73
DEEPTB INFO iteration:74 train_loss: 0.031137 (0.031986) lr: 0.009296
DEEPTB INFO checkpoint saved as nnsk.iter74
DEEPTB INFO Epoch 74 summary: train_loss: 0.031137
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep74
DEEPTB INFO iteration:75 train_loss: 0.030939 (0.031672) lr: 0.009286
DEEPTB INFO checkpoint saved as nnsk.iter75
DEEPTB INFO Epoch 75 summary: train_loss: 0.030939
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep75
DEEPTB INFO iteration:76 train_loss: 0.030683 (0.031375) lr: 0.009277
DEEPTB INFO checkpoint saved as nnsk.iter76
DEEPTB INFO Epoch 76 summary: train_loss: 0.030683
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep76
DEEPTB INFO iteration:77 train_loss: 0.030332 (0.031062) lr: 0.009268
DEEPTB INFO checkpoint saved as nnsk.iter77
DEEPTB INFO Epoch 77 summary: train_loss: 0.030332
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep77
DEEPTB INFO iteration:78 train_loss: 0.029930 (0.030722) lr: 0.009259
DEEPTB INFO checkpoint saved as nnsk.iter78
DEEPTB INFO Epoch 78 summary: train_loss: 0.029930
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep78
DEEPTB INFO iteration:79 train_loss: 0.029552 (0.030371) lr: 0.009249
DEEPTB INFO checkpoint saved as nnsk.iter79
DEEPTB INFO Epoch 79 summary: train_loss: 0.029552
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep79
DEEPTB INFO iteration:80 train_loss: 0.029249 (0.030035) lr: 0.00924
DEEPTB INFO checkpoint saved as nnsk.iter80
DEEPTB INFO Epoch 80 summary: train_loss: 0.029249
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep80
DEEPTB INFO iteration:81 train_loss: 0.029012 (0.029728) lr: 0.009231
DEEPTB INFO checkpoint saved as nnsk.iter81
DEEPTB INFO Epoch 81 summary: train_loss: 0.029012
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep81
DEEPTB INFO iteration:82 train_loss: 0.028791 (0.029447) lr: 0.009222
DEEPTB INFO checkpoint saved as nnsk.iter82
DEEPTB INFO Epoch 82 summary: train_loss: 0.028791
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep82
DEEPTB INFO iteration:83 train_loss: 0.028539 (0.029174) lr: 0.009212
DEEPTB INFO checkpoint saved as nnsk.iter83
DEEPTB INFO Epoch 83 summary: train_loss: 0.028539
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep83
DEEPTB INFO iteration:84 train_loss: 0.028245 (0.028896) lr: 0.009203
DEEPTB INFO checkpoint saved as nnsk.iter84
DEEPTB INFO Epoch 84 summary: train_loss: 0.028245
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep84
DEEPTB INFO iteration:85 train_loss: 0.027937 (0.028608) lr: 0.009194
DEEPTB INFO checkpoint saved as nnsk.iter85
DEEPTB INFO Epoch 85 summary: train_loss: 0.027937
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep85
DEEPTB INFO iteration:86 train_loss: 0.027651 (0.028321) lr: 0.009185
DEEPTB INFO checkpoint saved as nnsk.iter86
DEEPTB INFO Epoch 86 summary: train_loss: 0.027651
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep86
DEEPTB INFO iteration:87 train_loss: 0.027403 (0.028046) lr: 0.009176
DEEPTB INFO checkpoint saved as nnsk.iter87
DEEPTB INFO Epoch 87 summary: train_loss: 0.027403
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep87
DEEPTB INFO iteration:88 train_loss: 0.027178 (0.027785) lr: 0.009166
DEEPTB INFO checkpoint saved as nnsk.iter88
DEEPTB INFO Epoch 88 summary: train_loss: 0.027178
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep88
DEEPTB INFO iteration:89 train_loss: 0.026951 (0.027535) lr: 0.009157
DEEPTB INFO checkpoint saved as nnsk.iter89
DEEPTB INFO Epoch 89 summary: train_loss: 0.026951
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep89
DEEPTB INFO iteration:90 train_loss: 0.026704 (0.027286) lr: 0.009148
DEEPTB INFO checkpoint saved as nnsk.iter90
DEEPTB INFO Epoch 90 summary: train_loss: 0.026704
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep90
DEEPTB INFO iteration:91 train_loss: 0.026442 (0.027033) lr: 0.009139
DEEPTB INFO checkpoint saved as nnsk.iter91
DEEPTB INFO Epoch 91 summary: train_loss: 0.026442
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep91
DEEPTB INFO iteration:92 train_loss: 0.026188 (0.026779) lr: 0.00913
DEEPTB INFO checkpoint saved as nnsk.iter92
DEEPTB INFO Epoch 92 summary: train_loss: 0.026188
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep92
DEEPTB INFO iteration:93 train_loss: 0.025954 (0.026532) lr: 0.009121
DEEPTB INFO checkpoint saved as nnsk.iter93
DEEPTB INFO Epoch 93 summary: train_loss: 0.025954
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep93
DEEPTB INFO iteration:94 train_loss: 0.025741 (0.026295) lr: 0.009112
DEEPTB INFO checkpoint saved as nnsk.iter94
DEEPTB INFO Epoch 94 summary: train_loss: 0.025741
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep94
DEEPTB INFO iteration:95 train_loss: 0.025535 (0.026067) lr: 0.009102
DEEPTB INFO checkpoint saved as nnsk.iter95
DEEPTB INFO Epoch 95 summary: train_loss: 0.025535
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep95
DEEPTB INFO iteration:96 train_loss: 0.025321 (0.025843) lr: 0.009093
DEEPTB INFO checkpoint saved as nnsk.iter96
DEEPTB INFO Epoch 96 summary: train_loss: 0.025321
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep96
DEEPTB INFO iteration:97 train_loss: 0.025096 (0.025619) lr: 0.009084
DEEPTB INFO checkpoint saved as nnsk.iter97
DEEPTB INFO Epoch 97 summary: train_loss: 0.025096
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep97
DEEPTB INFO iteration:98 train_loss: 0.024869 (0.025394) lr: 0.009075
DEEPTB INFO checkpoint saved as nnsk.iter98
DEEPTB INFO Epoch 98 summary: train_loss: 0.024869
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep98
DEEPTB INFO iteration:99 train_loss: 0.024652 (0.025171) lr: 0.009066
DEEPTB INFO checkpoint saved as nnsk.iter99
DEEPTB INFO Epoch 99 summary: train_loss: 0.024652
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep99
DEEPTB INFO iteration:100 train_loss: 0.024449 (0.024955) lr: 0.009057
DEEPTB INFO checkpoint saved as nnsk.iter100
DEEPTB INFO Epoch 100 summary: train_loss: 0.024449
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep100
DEEPTB INFO iteration:101 train_loss: 0.024254 (0.024744) lr: 0.009048
DEEPTB INFO checkpoint saved as nnsk.iter101
DEEPTB INFO Epoch 101 summary: train_loss: 0.024254
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep101
DEEPTB INFO iteration:102 train_loss: 0.024060 (0.024539) lr: 0.009039
DEEPTB INFO checkpoint saved as nnsk.iter102
DEEPTB INFO Epoch 102 summary: train_loss: 0.024060
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep102
DEEPTB INFO iteration:103 train_loss: 0.023862 (0.024336) lr: 0.00903
DEEPTB INFO checkpoint saved as nnsk.iter103
DEEPTB INFO Epoch 103 summary: train_loss: 0.023862
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep103
DEEPTB INFO iteration:104 train_loss: 0.023663 (0.024134) lr: 0.009021
DEEPTB INFO checkpoint saved as nnsk.iter104
DEEPTB INFO Epoch 104 summary: train_loss: 0.023663
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep104
DEEPTB INFO iteration:105 train_loss: 0.023468 (0.023934) lr: 0.009012
DEEPTB INFO checkpoint saved as nnsk.iter105
DEEPTB INFO Epoch 105 summary: train_loss: 0.023468
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep105
DEEPTB INFO iteration:106 train_loss: 0.023281 (0.023738) lr: 0.009003
DEEPTB INFO checkpoint saved as nnsk.iter106
DEEPTB INFO Epoch 106 summary: train_loss: 0.023281
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep106
DEEPTB INFO iteration:107 train_loss: 0.023100 (0.023547) lr: 0.008994
DEEPTB INFO checkpoint saved as nnsk.iter107
DEEPTB INFO Epoch 107 summary: train_loss: 0.023100
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep107
DEEPTB INFO iteration:108 train_loss: 0.022920 (0.023359) lr: 0.008985
DEEPTB INFO checkpoint saved as nnsk.iter108
DEEPTB INFO Epoch 108 summary: train_loss: 0.022920
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep108
DEEPTB INFO iteration:109 train_loss: 0.022740 (0.023173) lr: 0.008976
DEEPTB INFO checkpoint saved as nnsk.iter109
DEEPTB INFO Epoch 109 summary: train_loss: 0.022740
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep109
DEEPTB INFO iteration:110 train_loss: 0.022560 (0.022989) lr: 0.008967
DEEPTB INFO checkpoint saved as nnsk.iter110
DEEPTB INFO Epoch 110 summary: train_loss: 0.022560
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep110
DEEPTB INFO iteration:111 train_loss: 0.022383 (0.022807) lr: 0.008958
DEEPTB INFO checkpoint saved as nnsk.iter111
DEEPTB INFO Epoch 111 summary: train_loss: 0.022383
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep111
DEEPTB INFO iteration:112 train_loss: 0.022211 (0.022628) lr: 0.008949
DEEPTB INFO checkpoint saved as nnsk.iter112
DEEPTB INFO Epoch 112 summary: train_loss: 0.022211
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep112
DEEPTB INFO iteration:113 train_loss: 0.022044 (0.022453) lr: 0.00894
DEEPTB INFO checkpoint saved as nnsk.iter113
DEEPTB INFO Epoch 113 summary: train_loss: 0.022044
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep113
DEEPTB INFO iteration:114 train_loss: 0.021881 (0.022281) lr: 0.008931
DEEPTB INFO checkpoint saved as nnsk.iter114
DEEPTB INFO Epoch 114 summary: train_loss: 0.021881
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep114
DEEPTB INFO iteration:115 train_loss: 0.021718 (0.022112) lr: 0.008922
DEEPTB INFO checkpoint saved as nnsk.iter115
DEEPTB INFO Epoch 115 summary: train_loss: 0.021718
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep115
DEEPTB INFO iteration:116 train_loss: 0.021555 (0.021945) lr: 0.008913
DEEPTB INFO checkpoint saved as nnsk.iter116
DEEPTB INFO Epoch 116 summary: train_loss: 0.021555
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep116
DEEPTB INFO iteration:117 train_loss: 0.021394 (0.021780) lr: 0.008904
DEEPTB INFO checkpoint saved as nnsk.iter117
DEEPTB INFO Epoch 117 summary: train_loss: 0.021394
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep117
DEEPTB INFO iteration:118 train_loss: 0.021236 (0.021617) lr: 0.008895
DEEPTB INFO checkpoint saved as nnsk.iter118
DEEPTB INFO Epoch 118 summary: train_loss: 0.021236
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep118
DEEPTB INFO iteration:119 train_loss: 0.021082 (0.021456) lr: 0.008886
DEEPTB INFO checkpoint saved as nnsk.iter119
DEEPTB INFO Epoch 119 summary: train_loss: 0.021082
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep119
DEEPTB INFO iteration:120 train_loss: 0.020931 (0.021299) lr: 0.008878
DEEPTB INFO checkpoint saved as nnsk.iter120
DEEPTB INFO Epoch 120 summary: train_loss: 0.020931
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep120
DEEPTB INFO iteration:121 train_loss: 0.020781 (0.021143) lr: 0.008869
DEEPTB INFO checkpoint saved as nnsk.iter121
DEEPTB INFO Epoch 121 summary: train_loss: 0.020781
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep121
DEEPTB INFO iteration:122 train_loss: 0.020633 (0.020990) lr: 0.00886
DEEPTB INFO checkpoint saved as nnsk.iter122
DEEPTB INFO Epoch 122 summary: train_loss: 0.020633
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep122
DEEPTB INFO iteration:123 train_loss: 0.020487 (0.020839) lr: 0.008851
DEEPTB INFO checkpoint saved as nnsk.iter123
DEEPTB INFO Epoch 123 summary: train_loss: 0.020487
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep123
DEEPTB INFO iteration:124 train_loss: 0.020343 (0.020690) lr: 0.008842
DEEPTB INFO checkpoint saved as nnsk.iter124
DEEPTB INFO Epoch 124 summary: train_loss: 0.020343
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep124
DEEPTB INFO iteration:125 train_loss: 0.020202 (0.020544) lr: 0.008833
DEEPTB INFO checkpoint saved as nnsk.iter125
DEEPTB INFO Epoch 125 summary: train_loss: 0.020202
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep125
DEEPTB INFO iteration:126 train_loss: 0.020063 (0.020400) lr: 0.008824
DEEPTB INFO checkpoint saved as nnsk.iter126
DEEPTB INFO Epoch 126 summary: train_loss: 0.020063
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep126
DEEPTB INFO iteration:127 train_loss: 0.019926 (0.020257) lr: 0.008816
DEEPTB INFO checkpoint saved as nnsk.iter127
DEEPTB INFO Epoch 127 summary: train_loss: 0.019926
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep127
DEEPTB INFO iteration:128 train_loss: 0.019790 (0.020117) lr: 0.008807
DEEPTB INFO checkpoint saved as nnsk.iter128
DEEPTB INFO Epoch 128 summary: train_loss: 0.019790
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep128
DEEPTB INFO iteration:129 train_loss: 0.019656 (0.019979) lr: 0.008798
DEEPTB INFO checkpoint saved as nnsk.iter129
DEEPTB INFO Epoch 129 summary: train_loss: 0.019656
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep129
DEEPTB INFO iteration:130 train_loss: 0.019524 (0.019842) lr: 0.008789
DEEPTB INFO checkpoint saved as nnsk.iter130
DEEPTB INFO Epoch 130 summary: train_loss: 0.019524
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep130
DEEPTB INFO iteration:131 train_loss: 0.019394 (0.019708) lr: 0.00878
DEEPTB INFO checkpoint saved as nnsk.iter131
DEEPTB INFO Epoch 131 summary: train_loss: 0.019394
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep131
DEEPTB INFO iteration:132 train_loss: 0.019267 (0.019575) lr: 0.008772
DEEPTB INFO checkpoint saved as nnsk.iter132
DEEPTB INFO Epoch 132 summary: train_loss: 0.019267
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep132
DEEPTB INFO iteration:133 train_loss: 0.019141 (0.019445) lr: 0.008763
DEEPTB INFO checkpoint saved as nnsk.iter133
DEEPTB INFO Epoch 133 summary: train_loss: 0.019141
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep133
DEEPTB INFO iteration:134 train_loss: 0.019016 (0.019316) lr: 0.008754
DEEPTB INFO checkpoint saved as nnsk.iter134
DEEPTB INFO Epoch 134 summary: train_loss: 0.019016
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep134
DEEPTB INFO iteration:135 train_loss: 0.018893 (0.019189) lr: 0.008745
DEEPTB INFO checkpoint saved as nnsk.iter135
DEEPTB INFO Epoch 135 summary: train_loss: 0.018893
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep135
DEEPTB INFO iteration:136 train_loss: 0.018772 (0.019064) lr: 0.008737
DEEPTB INFO checkpoint saved as nnsk.iter136
DEEPTB INFO Epoch 136 summary: train_loss: 0.018772
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep136
DEEPTB INFO iteration:137 train_loss: 0.018653 (0.018941) lr: 0.008728
DEEPTB INFO checkpoint saved as nnsk.iter137
DEEPTB INFO Epoch 137 summary: train_loss: 0.018653
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep137
DEEPTB INFO iteration:138 train_loss: 0.018536 (0.018819) lr: 0.008719
DEEPTB INFO checkpoint saved as nnsk.iter138
DEEPTB INFO Epoch 138 summary: train_loss: 0.018536
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep138
DEEPTB INFO iteration:139 train_loss: 0.018420 (0.018699) lr: 0.00871
DEEPTB INFO checkpoint saved as nnsk.iter139
DEEPTB INFO Epoch 139 summary: train_loss: 0.018420
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep139
DEEPTB INFO iteration:140 train_loss: 0.018305 (0.018581) lr: 0.008702
DEEPTB INFO checkpoint saved as nnsk.iter140
DEEPTB INFO Epoch 140 summary: train_loss: 0.018305
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep140
DEEPTB INFO iteration:141 train_loss: 0.018192 (0.018464) lr: 0.008693
DEEPTB INFO checkpoint saved as nnsk.iter141
DEEPTB INFO Epoch 141 summary: train_loss: 0.018192
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep141
DEEPTB INFO iteration:142 train_loss: 0.018081 (0.018349) lr: 0.008684
DEEPTB INFO checkpoint saved as nnsk.iter142
DEEPTB INFO Epoch 142 summary: train_loss: 0.018081
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep142
DEEPTB INFO iteration:143 train_loss: 0.017971 (0.018236) lr: 0.008676
DEEPTB INFO checkpoint saved as nnsk.iter143
DEEPTB INFO Epoch 143 summary: train_loss: 0.017971
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep143
DEEPTB INFO iteration:144 train_loss: 0.017863 (0.018124) lr: 0.008667
DEEPTB INFO checkpoint saved as nnsk.iter144
DEEPTB INFO Epoch 144 summary: train_loss: 0.017863
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep144
DEEPTB INFO iteration:145 train_loss: 0.017757 (0.018014) lr: 0.008658
DEEPTB INFO checkpoint saved as nnsk.iter145
DEEPTB INFO Epoch 145 summary: train_loss: 0.017757
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep145
DEEPTB INFO iteration:146 train_loss: 0.017652 (0.017905) lr: 0.00865
DEEPTB INFO checkpoint saved as nnsk.iter146
DEEPTB INFO Epoch 146 summary: train_loss: 0.017652
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep146
DEEPTB INFO iteration:147 train_loss: 0.017548 (0.017798) lr: 0.008641
DEEPTB INFO checkpoint saved as nnsk.iter147
DEEPTB INFO Epoch 147 summary: train_loss: 0.017548
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep147
DEEPTB INFO iteration:148 train_loss: 0.017445 (0.017692) lr: 0.008632
DEEPTB INFO checkpoint saved as nnsk.iter148
DEEPTB INFO Epoch 148 summary: train_loss: 0.017445
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep148
DEEPTB INFO iteration:149 train_loss: 0.017345 (0.017588) lr: 0.008624
DEEPTB INFO checkpoint saved as nnsk.iter149
DEEPTB INFO Epoch 149 summary: train_loss: 0.017345
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep149
DEEPTB INFO iteration:150 train_loss: 0.017245 (0.017485) lr: 0.008615
DEEPTB INFO checkpoint saved as nnsk.iter150
DEEPTB INFO Epoch 150 summary: train_loss: 0.017245
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep150
DEEPTB INFO iteration:151 train_loss: 0.017147 (0.017384) lr: 0.008606
DEEPTB INFO checkpoint saved as nnsk.iter151
DEEPTB INFO Epoch 151 summary: train_loss: 0.017147
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep151
DEEPTB INFO iteration:152 train_loss: 0.017050 (0.017284) lr: 0.008598
DEEPTB INFO checkpoint saved as nnsk.iter152
DEEPTB INFO Epoch 152 summary: train_loss: 0.017050
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep152
DEEPTB INFO iteration:153 train_loss: 0.016955 (0.017185) lr: 0.008589
DEEPTB INFO checkpoint saved as nnsk.iter153
DEEPTB INFO Epoch 153 summary: train_loss: 0.016955
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep153
DEEPTB INFO iteration:154 train_loss: 0.016860 (0.017088) lr: 0.008581
DEEPTB INFO checkpoint saved as nnsk.iter154
DEEPTB INFO Epoch 154 summary: train_loss: 0.016860
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep154
DEEPTB INFO iteration:155 train_loss: 0.016768 (0.016992) lr: 0.008572
DEEPTB INFO checkpoint saved as nnsk.iter155
DEEPTB INFO Epoch 155 summary: train_loss: 0.016768
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep155
DEEPTB INFO iteration:156 train_loss: 0.016676 (0.016897) lr: 0.008563
DEEPTB INFO checkpoint saved as nnsk.iter156
DEEPTB INFO Epoch 156 summary: train_loss: 0.016676
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep156
DEEPTB INFO iteration:157 train_loss: 0.016586 (0.016804) lr: 0.008555
DEEPTB INFO checkpoint saved as nnsk.iter157
DEEPTB INFO Epoch 157 summary: train_loss: 0.016586
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep157
DEEPTB INFO iteration:158 train_loss: 0.016497 (0.016712) lr: 0.008546
DEEPTB INFO checkpoint saved as nnsk.iter158
DEEPTB INFO Epoch 158 summary: train_loss: 0.016497
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep158
DEEPTB INFO iteration:159 train_loss: 0.016409 (0.016621) lr: 0.008538
DEEPTB INFO checkpoint saved as nnsk.iter159
DEEPTB INFO Epoch 159 summary: train_loss: 0.016409
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep159
DEEPTB INFO iteration:160 train_loss: 0.016322 (0.016531) lr: 0.008529
DEEPTB INFO checkpoint saved as nnsk.iter160
DEEPTB INFO Epoch 160 summary: train_loss: 0.016322
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep160
DEEPTB INFO iteration:161 train_loss: 0.016236 (0.016443) lr: 0.008521
DEEPTB INFO checkpoint saved as nnsk.iter161
DEEPTB INFO Epoch 161 summary: train_loss: 0.016236
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep161
DEEPTB INFO iteration:162 train_loss: 0.016152 (0.016355) lr: 0.008512
DEEPTB INFO checkpoint saved as nnsk.iter162
DEEPTB INFO Epoch 162 summary: train_loss: 0.016152
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep162
DEEPTB INFO iteration:163 train_loss: 0.016069 (0.016269) lr: 0.008504
DEEPTB INFO checkpoint saved as nnsk.iter163
DEEPTB INFO Epoch 163 summary: train_loss: 0.016069
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep163
DEEPTB INFO iteration:164 train_loss: 0.015986 (0.016185) lr: 0.008495
DEEPTB INFO checkpoint saved as nnsk.iter164
DEEPTB INFO Epoch 164 summary: train_loss: 0.015986
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep164
DEEPTB INFO iteration:165 train_loss: 0.015905 (0.016101) lr: 0.008487
DEEPTB INFO checkpoint saved as nnsk.iter165
DEEPTB INFO Epoch 165 summary: train_loss: 0.015905
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep165
DEEPTB INFO iteration:166 train_loss: 0.015825 (0.016018) lr: 0.008478
DEEPTB INFO checkpoint saved as nnsk.iter166
DEEPTB INFO Epoch 166 summary: train_loss: 0.015825
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep166
DEEPTB INFO iteration:167 train_loss: 0.015747 (0.015937) lr: 0.00847
DEEPTB INFO checkpoint saved as nnsk.iter167
DEEPTB INFO Epoch 167 summary: train_loss: 0.015747
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep167
DEEPTB INFO iteration:168 train_loss: 0.015669 (0.015856) lr: 0.008461
DEEPTB INFO checkpoint saved as nnsk.iter168
DEEPTB INFO Epoch 168 summary: train_loss: 0.015669
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep168
DEEPTB INFO iteration:169 train_loss: 0.015592 (0.015777) lr: 0.008453
DEEPTB INFO checkpoint saved as nnsk.iter169
DEEPTB INFO Epoch 169 summary: train_loss: 0.015592
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep169
DEEPTB INFO iteration:170 train_loss: 0.015516 (0.015699) lr: 0.008444
DEEPTB INFO checkpoint saved as nnsk.iter170
DEEPTB INFO Epoch 170 summary: train_loss: 0.015516
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep170
DEEPTB INFO iteration:171 train_loss: 0.015441 (0.015622) lr: 0.008436
DEEPTB INFO checkpoint saved as nnsk.iter171
DEEPTB INFO Epoch 171 summary: train_loss: 0.015441
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep171
DEEPTB INFO iteration:172 train_loss: 0.015368 (0.015545) lr: 0.008427
DEEPTB INFO checkpoint saved as nnsk.iter172
DEEPTB INFO Epoch 172 summary: train_loss: 0.015368
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep172
DEEPTB INFO iteration:173 train_loss: 0.015295 (0.015470) lr: 0.008419
DEEPTB INFO checkpoint saved as nnsk.iter173
DEEPTB INFO Epoch 173 summary: train_loss: 0.015295
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep173
DEEPTB INFO iteration:174 train_loss: 0.015223 (0.015396) lr: 0.008411
DEEPTB INFO checkpoint saved as nnsk.iter174
DEEPTB INFO Epoch 174 summary: train_loss: 0.015223
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep174
DEEPTB INFO iteration:175 train_loss: 0.015152 (0.015323) lr: 0.008402
DEEPTB INFO checkpoint saved as nnsk.iter175
DEEPTB INFO Epoch 175 summary: train_loss: 0.015152
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep175
DEEPTB INFO iteration:176 train_loss: 0.015082 (0.015251) lr: 0.008394
DEEPTB INFO checkpoint saved as nnsk.iter176
DEEPTB INFO Epoch 176 summary: train_loss: 0.015082
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep176
DEEPTB INFO iteration:177 train_loss: 0.015013 (0.015179) lr: 0.008385
DEEPTB INFO checkpoint saved as nnsk.iter177
DEEPTB INFO Epoch 177 summary: train_loss: 0.015013
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep177
DEEPTB INFO iteration:178 train_loss: 0.014945 (0.015109) lr: 0.008377
DEEPTB INFO checkpoint saved as nnsk.iter178
DEEPTB INFO Epoch 178 summary: train_loss: 0.014945
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep178
DEEPTB INFO iteration:179 train_loss: 0.014877 (0.015040) lr: 0.008369
DEEPTB INFO checkpoint saved as nnsk.iter179
DEEPTB INFO Epoch 179 summary: train_loss: 0.014877
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep179
DEEPTB INFO iteration:180 train_loss: 0.014811 (0.014971) lr: 0.00836
DEEPTB INFO checkpoint saved as nnsk.iter180
DEEPTB INFO Epoch 180 summary: train_loss: 0.014811
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep180
DEEPTB INFO iteration:181 train_loss: 0.014745 (0.014903) lr: 0.008352
DEEPTB INFO checkpoint saved as nnsk.iter181
DEEPTB INFO Epoch 181 summary: train_loss: 0.014745
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep181
DEEPTB INFO iteration:182 train_loss: 0.014681 (0.014837) lr: 0.008344
DEEPTB INFO checkpoint saved as nnsk.iter182
DEEPTB INFO Epoch 182 summary: train_loss: 0.014681
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep182
DEEPTB INFO iteration:183 train_loss: 0.014617 (0.014771) lr: 0.008335
DEEPTB INFO checkpoint saved as nnsk.iter183
DEEPTB INFO Epoch 183 summary: train_loss: 0.014617
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep183
DEEPTB INFO iteration:184 train_loss: 0.014554 (0.014705) lr: 0.008327
DEEPTB INFO checkpoint saved as nnsk.iter184
DEEPTB INFO Epoch 184 summary: train_loss: 0.014554
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep184
DEEPTB INFO iteration:185 train_loss: 0.014491 (0.014641) lr: 0.008319
DEEPTB INFO checkpoint saved as nnsk.iter185
DEEPTB INFO Epoch 185 summary: train_loss: 0.014491
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep185
DEEPTB INFO iteration:186 train_loss: 0.014430 (0.014578) lr: 0.00831
DEEPTB INFO checkpoint saved as nnsk.iter186
DEEPTB INFO Epoch 186 summary: train_loss: 0.014430
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep186
DEEPTB INFO iteration:187 train_loss: 0.014369 (0.014515) lr: 0.008302
DEEPTB INFO checkpoint saved as nnsk.iter187
DEEPTB INFO Epoch 187 summary: train_loss: 0.014369
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep187
DEEPTB INFO iteration:188 train_loss: 0.014309 (0.014453) lr: 0.008294
DEEPTB INFO checkpoint saved as nnsk.iter188
DEEPTB INFO Epoch 188 summary: train_loss: 0.014309
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep188
DEEPTB INFO iteration:189 train_loss: 0.014249 (0.014392) lr: 0.008285
DEEPTB INFO checkpoint saved as nnsk.iter189
DEEPTB INFO Epoch 189 summary: train_loss: 0.014249
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep189
DEEPTB INFO iteration:190 train_loss: 0.014191 (0.014332) lr: 0.008277
DEEPTB INFO checkpoint saved as nnsk.iter190
DEEPTB INFO Epoch 190 summary: train_loss: 0.014191
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep190
DEEPTB INFO iteration:191 train_loss: 0.014133 (0.014272) lr: 0.008269
DEEPTB INFO checkpoint saved as nnsk.iter191
DEEPTB INFO Epoch 191 summary: train_loss: 0.014133
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep191
DEEPTB INFO iteration:192 train_loss: 0.014077 (0.014213) lr: 0.008261
DEEPTB INFO checkpoint saved as nnsk.iter192
DEEPTB INFO Epoch 192 summary: train_loss: 0.014077
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep192
DEEPTB INFO iteration:193 train_loss: 0.014023 (0.014156) lr: 0.008252
DEEPTB INFO checkpoint saved as nnsk.iter193
DEEPTB INFO Epoch 193 summary: train_loss: 0.014023
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep193
DEEPTB INFO iteration:194 train_loss: 0.013966 (0.014099) lr: 0.008244
DEEPTB INFO checkpoint saved as nnsk.iter194
DEEPTB INFO Epoch 194 summary: train_loss: 0.013966
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep194
DEEPTB INFO iteration:195 train_loss: 0.013915 (0.014044) lr: 0.008236
DEEPTB INFO checkpoint saved as nnsk.iter195
DEEPTB INFO Epoch 195 summary: train_loss: 0.013915
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep195
DEEPTB INFO iteration:196 train_loss: 0.013864 (0.013990) lr: 0.008228
DEEPTB INFO checkpoint saved as nnsk.iter196
DEEPTB INFO Epoch 196 summary: train_loss: 0.013864
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep196
DEEPTB INFO iteration:197 train_loss: 0.013813 (0.013937) lr: 0.008219
DEEPTB INFO checkpoint saved as nnsk.iter197
DEEPTB INFO Epoch 197 summary: train_loss: 0.013813
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep197
DEEPTB INFO iteration:198 train_loss: 0.013762 (0.013885) lr: 0.008211
DEEPTB INFO checkpoint saved as nnsk.iter198
DEEPTB INFO Epoch 198 summary: train_loss: 0.013762
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep198
DEEPTB INFO iteration:199 train_loss: 0.013711 (0.013833) lr: 0.008203
DEEPTB INFO checkpoint saved as nnsk.iter199
DEEPTB INFO Epoch 199 summary: train_loss: 0.013711
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep199
DEEPTB INFO iteration:200 train_loss: 0.013660 (0.013781) lr: 0.008195
DEEPTB INFO checkpoint saved as nnsk.iter200
DEEPTB INFO Epoch 200 summary: train_loss: 0.013660
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep200
DEEPTB INFO iteration:201 train_loss: 0.013611 (0.013730) lr: 0.008186
DEEPTB INFO checkpoint saved as nnsk.iter201
DEEPTB INFO Epoch 201 summary: train_loss: 0.013611
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep201
DEEPTB INFO iteration:202 train_loss: 0.013565 (0.013680) lr: 0.008178
DEEPTB INFO checkpoint saved as nnsk.iter202
DEEPTB INFO Epoch 202 summary: train_loss: 0.013565
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep202
DEEPTB INFO iteration:203 train_loss: 0.013516 (0.013631) lr: 0.00817
DEEPTB INFO checkpoint saved as nnsk.iter203
DEEPTB INFO Epoch 203 summary: train_loss: 0.013516
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep203
DEEPTB INFO iteration:204 train_loss: 0.013468 (0.013582) lr: 0.008162
DEEPTB INFO checkpoint saved as nnsk.iter204
DEEPTB INFO Epoch 204 summary: train_loss: 0.013468
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep204
DEEPTB INFO iteration:205 train_loss: 0.013424 (0.013535) lr: 0.008154
DEEPTB INFO checkpoint saved as nnsk.iter205
DEEPTB INFO Epoch 205 summary: train_loss: 0.013424
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep205
DEEPTB INFO iteration:206 train_loss: 0.013380 (0.013488) lr: 0.008146
DEEPTB INFO checkpoint saved as nnsk.iter206
DEEPTB INFO Epoch 206 summary: train_loss: 0.013380
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep206
DEEPTB INFO iteration:207 train_loss: 0.013334 (0.013442) lr: 0.008137
DEEPTB INFO checkpoint saved as nnsk.iter207
DEEPTB INFO Epoch 207 summary: train_loss: 0.013334
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep207
DEEPTB INFO iteration:208 train_loss: 0.013288 (0.013396) lr: 0.008129
DEEPTB INFO checkpoint saved as nnsk.iter208
DEEPTB INFO Epoch 208 summary: train_loss: 0.013288
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep208
DEEPTB INFO iteration:209 train_loss: 0.013246 (0.013351) lr: 0.008121
DEEPTB INFO checkpoint saved as nnsk.iter209
DEEPTB INFO Epoch 209 summary: train_loss: 0.013246
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep209
DEEPTB INFO iteration:210 train_loss: 0.013204 (0.013307) lr: 0.008113
DEEPTB INFO checkpoint saved as nnsk.iter210
DEEPTB INFO Epoch 210 summary: train_loss: 0.013204
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep210
DEEPTB INFO iteration:211 train_loss: 0.013161 (0.013263) lr: 0.008105
DEEPTB INFO checkpoint saved as nnsk.iter211
DEEPTB INFO Epoch 211 summary: train_loss: 0.013161
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep211
DEEPTB INFO iteration:212 train_loss: 0.013117 (0.013219) lr: 0.008097
DEEPTB INFO checkpoint saved as nnsk.iter212
DEEPTB INFO Epoch 212 summary: train_loss: 0.013117
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep212
DEEPTB INFO iteration:213 train_loss: 0.013076 (0.013176) lr: 0.008089
DEEPTB INFO checkpoint saved as nnsk.iter213
DEEPTB INFO Epoch 213 summary: train_loss: 0.013076
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep213
DEEPTB INFO iteration:214 train_loss: 0.013036 (0.013134) lr: 0.008081
DEEPTB INFO checkpoint saved as nnsk.iter214
DEEPTB INFO Epoch 214 summary: train_loss: 0.013036
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep214
DEEPTB INFO iteration:215 train_loss: 0.012994 (0.013092) lr: 0.008073
DEEPTB INFO checkpoint saved as nnsk.iter215
DEEPTB INFO Epoch 215 summary: train_loss: 0.012994
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep215
DEEPTB INFO iteration:216 train_loss: 0.012953 (0.013050) lr: 0.008065
DEEPTB INFO checkpoint saved as nnsk.iter216
DEEPTB INFO Epoch 216 summary: train_loss: 0.012953
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep216
DEEPTB INFO iteration:217 train_loss: 0.012914 (0.013009) lr: 0.008056
DEEPTB INFO checkpoint saved as nnsk.iter217
DEEPTB INFO Epoch 217 summary: train_loss: 0.012914
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep217
DEEPTB INFO iteration:218 train_loss: 0.012874 (0.012969) lr: 0.008048
DEEPTB INFO checkpoint saved as nnsk.iter218
DEEPTB INFO Epoch 218 summary: train_loss: 0.012874
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep218
DEEPTB INFO iteration:219 train_loss: 0.012835 (0.012929) lr: 0.00804
DEEPTB INFO checkpoint saved as nnsk.iter219
DEEPTB INFO Epoch 219 summary: train_loss: 0.012835
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep219
DEEPTB INFO iteration:220 train_loss: 0.012797 (0.012889) lr: 0.008032
DEEPTB INFO checkpoint saved as nnsk.iter220
DEEPTB INFO Epoch 220 summary: train_loss: 0.012797
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep220
DEEPTB INFO iteration:221 train_loss: 0.012759 (0.012850) lr: 0.008024
DEEPTB INFO checkpoint saved as nnsk.iter221
DEEPTB INFO Epoch 221 summary: train_loss: 0.012759
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep221
DEEPTB INFO iteration:222 train_loss: 0.012722 (0.012812) lr: 0.008016
DEEPTB INFO checkpoint saved as nnsk.iter222
DEEPTB INFO Epoch 222 summary: train_loss: 0.012722
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep222
DEEPTB INFO iteration:223 train_loss: 0.012684 (0.012773) lr: 0.008008
DEEPTB INFO checkpoint saved as nnsk.iter223
DEEPTB INFO Epoch 223 summary: train_loss: 0.012684
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep223
DEEPTB INFO iteration:224 train_loss: 0.012647 (0.012735) lr: 0.008
DEEPTB INFO checkpoint saved as nnsk.iter224
DEEPTB INFO Epoch 224 summary: train_loss: 0.012647
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep224
DEEPTB INFO iteration:225 train_loss: 0.012610 (0.012698) lr: 0.007992
DEEPTB INFO checkpoint saved as nnsk.iter225
DEEPTB INFO Epoch 225 summary: train_loss: 0.012610
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep225
DEEPTB INFO iteration:226 train_loss: 0.012575 (0.012661) lr: 0.007984
DEEPTB INFO checkpoint saved as nnsk.iter226
DEEPTB INFO Epoch 226 summary: train_loss: 0.012575
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep226
DEEPTB INFO iteration:227 train_loss: 0.012540 (0.012624) lr: 0.007976
DEEPTB INFO checkpoint saved as nnsk.iter227
DEEPTB INFO Epoch 227 summary: train_loss: 0.012540
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep227
DEEPTB INFO iteration:228 train_loss: 0.012504 (0.012588) lr: 0.007968
DEEPTB INFO checkpoint saved as nnsk.iter228
DEEPTB INFO Epoch 228 summary: train_loss: 0.012504
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep228
DEEPTB INFO iteration:229 train_loss: 0.012468 (0.012552) lr: 0.00796
DEEPTB INFO checkpoint saved as nnsk.iter229
DEEPTB INFO Epoch 229 summary: train_loss: 0.012468
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep229
DEEPTB INFO iteration:230 train_loss: 0.012433 (0.012516) lr: 0.007952
DEEPTB INFO checkpoint saved as nnsk.iter230
DEEPTB INFO Epoch 230 summary: train_loss: 0.012433
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep230
DEEPTB INFO iteration:231 train_loss: 0.012399 (0.012481) lr: 0.007944
DEEPTB INFO checkpoint saved as nnsk.iter231
DEEPTB INFO Epoch 231 summary: train_loss: 0.012399
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep231
DEEPTB INFO iteration:232 train_loss: 0.012364 (0.012446) lr: 0.007936
DEEPTB INFO checkpoint saved as nnsk.iter232
DEEPTB INFO Epoch 232 summary: train_loss: 0.012364
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep232
DEEPTB INFO iteration:233 train_loss: 0.012330 (0.012411) lr: 0.007929
DEEPTB INFO checkpoint saved as nnsk.iter233
DEEPTB INFO Epoch 233 summary: train_loss: 0.012330
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep233
DEEPTB INFO iteration:234 train_loss: 0.012297 (0.012377) lr: 0.007921
DEEPTB INFO checkpoint saved as nnsk.iter234
DEEPTB INFO Epoch 234 summary: train_loss: 0.012297
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep234
DEEPTB INFO iteration:235 train_loss: 0.012263 (0.012343) lr: 0.007913
DEEPTB INFO checkpoint saved as nnsk.iter235
DEEPTB INFO Epoch 235 summary: train_loss: 0.012263
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep235
DEEPTB INFO iteration:236 train_loss: 0.012230 (0.012309) lr: 0.007905
DEEPTB INFO checkpoint saved as nnsk.iter236
DEEPTB INFO Epoch 236 summary: train_loss: 0.012230
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep236
DEEPTB INFO iteration:237 train_loss: 0.012197 (0.012275) lr: 0.007897
DEEPTB INFO checkpoint saved as nnsk.iter237
DEEPTB INFO Epoch 237 summary: train_loss: 0.012197
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep237
DEEPTB INFO iteration:238 train_loss: 0.012165 (0.012242) lr: 0.007889
DEEPTB INFO checkpoint saved as nnsk.iter238
DEEPTB INFO Epoch 238 summary: train_loss: 0.012165
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep238
DEEPTB INFO iteration:239 train_loss: 0.012133 (0.012209) lr: 0.007881
DEEPTB INFO checkpoint saved as nnsk.iter239
DEEPTB INFO Epoch 239 summary: train_loss: 0.012133
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep239
DEEPTB INFO iteration:240 train_loss: 0.012101 (0.012177) lr: 0.007873
DEEPTB INFO checkpoint saved as nnsk.iter240
DEEPTB INFO Epoch 240 summary: train_loss: 0.012101
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep240
DEEPTB INFO iteration:241 train_loss: 0.012069 (0.012144) lr: 0.007865
DEEPTB INFO checkpoint saved as nnsk.iter241
DEEPTB INFO Epoch 241 summary: train_loss: 0.012069
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep241
DEEPTB INFO iteration:242 train_loss: 0.012036 (0.012112) lr: 0.007857
DEEPTB INFO checkpoint saved as nnsk.iter242
DEEPTB INFO Epoch 242 summary: train_loss: 0.012036
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep242
DEEPTB INFO iteration:243 train_loss: 0.012004 (0.012080) lr: 0.00785
DEEPTB INFO checkpoint saved as nnsk.iter243
DEEPTB INFO Epoch 243 summary: train_loss: 0.012004
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep243
DEEPTB INFO iteration:244 train_loss: 0.011973 (0.012048) lr: 0.007842
DEEPTB INFO checkpoint saved as nnsk.iter244
DEEPTB INFO Epoch 244 summary: train_loss: 0.011973
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep244
DEEPTB INFO iteration:245 train_loss: 0.011943 (0.012016) lr: 0.007834
DEEPTB INFO checkpoint saved as nnsk.iter245
DEEPTB INFO Epoch 245 summary: train_loss: 0.011943
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep245
DEEPTB INFO iteration:246 train_loss: 0.011912 (0.011985) lr: 0.007826
DEEPTB INFO checkpoint saved as nnsk.iter246
DEEPTB INFO Epoch 246 summary: train_loss: 0.011912
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep246
DEEPTB INFO iteration:247 train_loss: 0.011882 (0.011954) lr: 0.007818
DEEPTB INFO checkpoint saved as nnsk.iter247
DEEPTB INFO Epoch 247 summary: train_loss: 0.011882
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep247
DEEPTB INFO iteration:248 train_loss: 0.011851 (0.011923) lr: 0.00781
DEEPTB INFO checkpoint saved as nnsk.iter248
DEEPTB INFO Epoch 248 summary: train_loss: 0.011851
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep248
DEEPTB INFO iteration:249 train_loss: 0.011820 (0.011892) lr: 0.007803
DEEPTB INFO checkpoint saved as nnsk.iter249
DEEPTB INFO Epoch 249 summary: train_loss: 0.011820
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep249
DEEPTB INFO iteration:250 train_loss: 0.011790 (0.011862) lr: 0.007795
DEEPTB INFO checkpoint saved as nnsk.iter250
DEEPTB INFO Epoch 250 summary: train_loss: 0.011790
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep250
DEEPTB INFO iteration:251 train_loss: 0.011762 (0.011832) lr: 0.007787
DEEPTB INFO checkpoint saved as nnsk.iter251
DEEPTB INFO Epoch 251 summary: train_loss: 0.011762
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep251
DEEPTB INFO iteration:252 train_loss: 0.011732 (0.011802) lr: 0.007779
DEEPTB INFO checkpoint saved as nnsk.iter252
DEEPTB INFO Epoch 252 summary: train_loss: 0.011732
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep252
DEEPTB INFO iteration:253 train_loss: 0.011701 (0.011772) lr: 0.007771
DEEPTB INFO checkpoint saved as nnsk.iter253
DEEPTB INFO Epoch 253 summary: train_loss: 0.011701
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep253
DEEPTB INFO iteration:254 train_loss: 0.011672 (0.011742) lr: 0.007764
DEEPTB INFO checkpoint saved as nnsk.iter254
DEEPTB INFO Epoch 254 summary: train_loss: 0.011672
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep254
DEEPTB INFO iteration:255 train_loss: 0.011643 (0.011712) lr: 0.007756
DEEPTB INFO checkpoint saved as nnsk.iter255
DEEPTB INFO Epoch 255 summary: train_loss: 0.011643
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep255
DEEPTB INFO iteration:256 train_loss: 0.011614 (0.011683) lr: 0.007748
DEEPTB INFO checkpoint saved as nnsk.iter256
DEEPTB INFO Epoch 256 summary: train_loss: 0.011614
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep256
DEEPTB INFO iteration:257 train_loss: 0.011585 (0.011653) lr: 0.00774
DEEPTB INFO checkpoint saved as nnsk.iter257
DEEPTB INFO Epoch 257 summary: train_loss: 0.011585
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep257
DEEPTB INFO iteration:258 train_loss: 0.011556 (0.011624) lr: 0.007733
DEEPTB INFO checkpoint saved as nnsk.iter258
DEEPTB INFO Epoch 258 summary: train_loss: 0.011556
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep258
DEEPTB INFO iteration:259 train_loss: 0.011528 (0.011595) lr: 0.007725
DEEPTB INFO checkpoint saved as nnsk.iter259
DEEPTB INFO Epoch 259 summary: train_loss: 0.011528
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep259
DEEPTB INFO iteration:260 train_loss: 0.011499 (0.011566) lr: 0.007717
DEEPTB INFO checkpoint saved as nnsk.iter260
DEEPTB INFO Epoch 260 summary: train_loss: 0.011499
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep260
DEEPTB INFO iteration:261 train_loss: 0.011470 (0.011537) lr: 0.00771
DEEPTB INFO checkpoint saved as nnsk.iter261
DEEPTB INFO Epoch 261 summary: train_loss: 0.011470
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep261
DEEPTB INFO iteration:262 train_loss: 0.011442 (0.011509) lr: 0.007702
DEEPTB INFO checkpoint saved as nnsk.iter262
DEEPTB INFO Epoch 262 summary: train_loss: 0.011442
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep262
DEEPTB INFO iteration:263 train_loss: 0.011415 (0.011481) lr: 0.007694
DEEPTB INFO checkpoint saved as nnsk.iter263
DEEPTB INFO Epoch 263 summary: train_loss: 0.011415
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep263
DEEPTB INFO iteration:264 train_loss: 0.011387 (0.011453) lr: 0.007686
DEEPTB INFO checkpoint saved as nnsk.iter264
DEEPTB INFO Epoch 264 summary: train_loss: 0.011387
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep264
DEEPTB INFO iteration:265 train_loss: 0.011360 (0.011425) lr: 0.007679
DEEPTB INFO checkpoint saved as nnsk.iter265
DEEPTB INFO Epoch 265 summary: train_loss: 0.011360
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep265
DEEPTB INFO iteration:266 train_loss: 0.011332 (0.011397) lr: 0.007671
DEEPTB INFO checkpoint saved as nnsk.iter266
DEEPTB INFO Epoch 266 summary: train_loss: 0.011332
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep266
DEEPTB INFO iteration:267 train_loss: 0.011304 (0.011369) lr: 0.007663
DEEPTB INFO checkpoint saved as nnsk.iter267
DEEPTB INFO Epoch 267 summary: train_loss: 0.011304
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep267
DEEPTB INFO iteration:268 train_loss: 0.011276 (0.011341) lr: 0.007656
DEEPTB INFO checkpoint saved as nnsk.iter268
DEEPTB INFO Epoch 268 summary: train_loss: 0.011276
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep268
DEEPTB INFO iteration:269 train_loss: 0.011251 (0.011314) lr: 0.007648
DEEPTB INFO checkpoint saved as nnsk.iter269
DEEPTB INFO Epoch 269 summary: train_loss: 0.011251
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep269
DEEPTB INFO iteration:270 train_loss: 0.011223 (0.011287) lr: 0.00764
DEEPTB INFO checkpoint saved as nnsk.iter270
DEEPTB INFO Epoch 270 summary: train_loss: 0.011223
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep270
DEEPTB INFO iteration:271 train_loss: 0.011196 (0.011260) lr: 0.007633
DEEPTB INFO checkpoint saved as nnsk.iter271
DEEPTB INFO Epoch 271 summary: train_loss: 0.011196
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep271
DEEPTB INFO iteration:272 train_loss: 0.011170 (0.011233) lr: 0.007625
DEEPTB INFO checkpoint saved as nnsk.iter272
DEEPTB INFO Epoch 272 summary: train_loss: 0.011170
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep272
DEEPTB INFO iteration:273 train_loss: 0.011144 (0.011206) lr: 0.007618
DEEPTB INFO checkpoint saved as nnsk.iter273
DEEPTB INFO Epoch 273 summary: train_loss: 0.011144
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep273
DEEPTB INFO iteration:274 train_loss: 0.011117 (0.011179) lr: 0.00761
DEEPTB INFO checkpoint saved as nnsk.iter274
DEEPTB INFO Epoch 274 summary: train_loss: 0.011117
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep274
DEEPTB INFO iteration:275 train_loss: 0.011091 (0.011153) lr: 0.007602
DEEPTB INFO checkpoint saved as nnsk.iter275
DEEPTB INFO Epoch 275 summary: train_loss: 0.011091
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep275
DEEPTB INFO iteration:276 train_loss: 0.011064 (0.011126) lr: 0.007595
DEEPTB INFO checkpoint saved as nnsk.iter276
DEEPTB INFO Epoch 276 summary: train_loss: 0.011064
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep276
DEEPTB INFO iteration:277 train_loss: 0.011037 (0.011099) lr: 0.007587
DEEPTB INFO checkpoint saved as nnsk.iter277
DEEPTB INFO Epoch 277 summary: train_loss: 0.011037
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep277
DEEPTB INFO iteration:278 train_loss: 0.011011 (0.011073) lr: 0.007579
DEEPTB INFO checkpoint saved as nnsk.iter278
DEEPTB INFO Epoch 278 summary: train_loss: 0.011011
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep278
DEEPTB INFO iteration:279 train_loss: 0.010985 (0.011046) lr: 0.007572
DEEPTB INFO checkpoint saved as nnsk.iter279
DEEPTB INFO Epoch 279 summary: train_loss: 0.010985
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep279
DEEPTB INFO iteration:280 train_loss: 0.010958 (0.011020) lr: 0.007564
DEEPTB INFO checkpoint saved as nnsk.iter280
DEEPTB INFO Epoch 280 summary: train_loss: 0.010958
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep280
DEEPTB INFO iteration:281 train_loss: 0.010933 (0.010994) lr: 0.007557
DEEPTB INFO checkpoint saved as nnsk.iter281
DEEPTB INFO Epoch 281 summary: train_loss: 0.010933
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep281
DEEPTB INFO iteration:282 train_loss: 0.010908 (0.010968) lr: 0.007549
DEEPTB INFO checkpoint saved as nnsk.iter282
DEEPTB INFO Epoch 282 summary: train_loss: 0.010908
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep282
DEEPTB INFO iteration:283 train_loss: 0.010883 (0.010942) lr: 0.007542
DEEPTB INFO checkpoint saved as nnsk.iter283
DEEPTB INFO Epoch 283 summary: train_loss: 0.010883
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep283
DEEPTB INFO iteration:284 train_loss: 0.010857 (0.010917) lr: 0.007534
DEEPTB INFO checkpoint saved as nnsk.iter284
DEEPTB INFO Epoch 284 summary: train_loss: 0.010857
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep284
DEEPTB INFO iteration:285 train_loss: 0.010832 (0.010891) lr: 0.007527
DEEPTB INFO checkpoint saved as nnsk.iter285
DEEPTB INFO Epoch 285 summary: train_loss: 0.010832
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep285
DEEPTB INFO iteration:286 train_loss: 0.010806 (0.010866) lr: 0.007519
DEEPTB INFO checkpoint saved as nnsk.iter286
DEEPTB INFO Epoch 286 summary: train_loss: 0.010806
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep286
DEEPTB INFO iteration:287 train_loss: 0.010781 (0.010840) lr: 0.007512
DEEPTB INFO checkpoint saved as nnsk.iter287
DEEPTB INFO Epoch 287 summary: train_loss: 0.010781
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep287
DEEPTB INFO iteration:288 train_loss: 0.010755 (0.010815) lr: 0.007504
DEEPTB INFO checkpoint saved as nnsk.iter288
DEEPTB INFO Epoch 288 summary: train_loss: 0.010755
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep288
DEEPTB INFO iteration:289 train_loss: 0.010730 (0.010789) lr: 0.007497
DEEPTB INFO checkpoint saved as nnsk.iter289
DEEPTB INFO Epoch 289 summary: train_loss: 0.010730
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep289
DEEPTB INFO iteration:290 train_loss: 0.010706 (0.010764) lr: 0.007489
DEEPTB INFO checkpoint saved as nnsk.iter290
DEEPTB INFO Epoch 290 summary: train_loss: 0.010706
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep290
DEEPTB INFO iteration:291 train_loss: 0.010680 (0.010739) lr: 0.007482
DEEPTB INFO checkpoint saved as nnsk.iter291
DEEPTB INFO Epoch 291 summary: train_loss: 0.010680
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep291
DEEPTB INFO iteration:292 train_loss: 0.010655 (0.010714) lr: 0.007474
DEEPTB INFO checkpoint saved as nnsk.iter292
DEEPTB INFO Epoch 292 summary: train_loss: 0.010655
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep292
DEEPTB INFO iteration:293 train_loss: 0.010631 (0.010689) lr: 0.007467
DEEPTB INFO checkpoint saved as nnsk.iter293
DEEPTB INFO Epoch 293 summary: train_loss: 0.010631
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep293
DEEPTB INFO iteration:294 train_loss: 0.010607 (0.010665) lr: 0.007459
DEEPTB INFO checkpoint saved as nnsk.iter294
DEEPTB INFO Epoch 294 summary: train_loss: 0.010607
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep294
DEEPTB INFO iteration:295 train_loss: 0.010583 (0.010640) lr: 0.007452
DEEPTB INFO checkpoint saved as nnsk.iter295
DEEPTB INFO Epoch 295 summary: train_loss: 0.010583
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep295
DEEPTB INFO iteration:296 train_loss: 0.010558 (0.010615) lr: 0.007444
DEEPTB INFO checkpoint saved as nnsk.iter296
DEEPTB INFO Epoch 296 summary: train_loss: 0.010558
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep296
DEEPTB INFO iteration:297 train_loss: 0.010534 (0.010591) lr: 0.007437
DEEPTB INFO checkpoint saved as nnsk.iter297
DEEPTB INFO Epoch 297 summary: train_loss: 0.010534
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep297
DEEPTB INFO iteration:298 train_loss: 0.010509 (0.010566) lr: 0.007429
DEEPTB INFO checkpoint saved as nnsk.iter298
DEEPTB INFO Epoch 298 summary: train_loss: 0.010509
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep298
DEEPTB INFO iteration:299 train_loss: 0.010484 (0.010542) lr: 0.007422
DEEPTB INFO checkpoint saved as nnsk.iter299
DEEPTB INFO Epoch 299 summary: train_loss: 0.010484
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep299
DEEPTB INFO iteration:300 train_loss: 0.010460 (0.010517) lr: 0.007414
DEEPTB INFO checkpoint saved as nnsk.iter300
DEEPTB INFO Epoch 300 summary: train_loss: 0.010460
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep300
DEEPTB INFO iteration:301 train_loss: 0.010436 (0.010493) lr: 0.007407
DEEPTB INFO checkpoint saved as nnsk.iter301
DEEPTB INFO Epoch 301 summary: train_loss: 0.010436
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep301
DEEPTB INFO iteration:302 train_loss: 0.010412 (0.010468) lr: 0.0074
DEEPTB INFO checkpoint saved as nnsk.iter302
DEEPTB INFO Epoch 302 summary: train_loss: 0.010412
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep302
DEEPTB INFO iteration:303 train_loss: 0.010387 (0.010444) lr: 0.007392
DEEPTB INFO checkpoint saved as nnsk.iter303
DEEPTB INFO Epoch 303 summary: train_loss: 0.010387
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep303
DEEPTB INFO iteration:304 train_loss: 0.010364 (0.010420) lr: 0.007385
DEEPTB INFO checkpoint saved as nnsk.iter304
DEEPTB INFO Epoch 304 summary: train_loss: 0.010364
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep304
DEEPTB INFO iteration:305 train_loss: 0.010340 (0.010396) lr: 0.007377
DEEPTB INFO checkpoint saved as nnsk.iter305
DEEPTB INFO Epoch 305 summary: train_loss: 0.010340
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep305
DEEPTB INFO iteration:306 train_loss: 0.010316 (0.010372) lr: 0.00737
DEEPTB INFO checkpoint saved as nnsk.iter306
DEEPTB INFO Epoch 306 summary: train_loss: 0.010316
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep306
DEEPTB INFO iteration:307 train_loss: 0.010292 (0.010348) lr: 0.007363
DEEPTB INFO checkpoint saved as nnsk.iter307
DEEPTB INFO Epoch 307 summary: train_loss: 0.010292
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep307
DEEPTB INFO iteration:308 train_loss: 0.010269 (0.010324) lr: 0.007355
DEEPTB INFO checkpoint saved as nnsk.iter308
DEEPTB INFO Epoch 308 summary: train_loss: 0.010269
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep308
DEEPTB INFO iteration:309 train_loss: 0.010245 (0.010301) lr: 0.007348
DEEPTB INFO checkpoint saved as nnsk.iter309
DEEPTB INFO Epoch 309 summary: train_loss: 0.010245
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep309
DEEPTB INFO iteration:310 train_loss: 0.010222 (0.010277) lr: 0.007341
DEEPTB INFO checkpoint saved as nnsk.iter310
DEEPTB INFO Epoch 310 summary: train_loss: 0.010222
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep310
DEEPTB INFO iteration:311 train_loss: 0.010199 (0.010254) lr: 0.007333
DEEPTB INFO checkpoint saved as nnsk.iter311
DEEPTB INFO Epoch 311 summary: train_loss: 0.010199
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep311
DEEPTB INFO iteration:312 train_loss: 0.010176 (0.010230) lr: 0.007326
DEEPTB INFO checkpoint saved as nnsk.iter312
DEEPTB INFO Epoch 312 summary: train_loss: 0.010176
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep312
DEEPTB INFO iteration:313 train_loss: 0.010152 (0.010207) lr: 0.007319
DEEPTB INFO checkpoint saved as nnsk.iter313
DEEPTB INFO Epoch 313 summary: train_loss: 0.010152
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep313
DEEPTB INFO iteration:314 train_loss: 0.010129 (0.010183) lr: 0.007311
DEEPTB INFO checkpoint saved as nnsk.iter314
DEEPTB INFO Epoch 314 summary: train_loss: 0.010129
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep314
DEEPTB INFO iteration:315 train_loss: 0.010106 (0.010160) lr: 0.007304
DEEPTB INFO checkpoint saved as nnsk.iter315
DEEPTB INFO Epoch 315 summary: train_loss: 0.010106
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep315
DEEPTB INFO iteration:316 train_loss: 0.010083 (0.010137) lr: 0.007297
DEEPTB INFO checkpoint saved as nnsk.iter316
DEEPTB INFO Epoch 316 summary: train_loss: 0.010083
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep316
DEEPTB INFO iteration:317 train_loss: 0.010060 (0.010114) lr: 0.007289
DEEPTB INFO checkpoint saved as nnsk.iter317
DEEPTB INFO Epoch 317 summary: train_loss: 0.010060
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep317
DEEPTB INFO iteration:318 train_loss: 0.010038 (0.010091) lr: 0.007282
DEEPTB INFO checkpoint saved as nnsk.iter318
DEEPTB INFO Epoch 318 summary: train_loss: 0.010038
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep318
DEEPTB INFO iteration:319 train_loss: 0.010014 (0.010068) lr: 0.007275
DEEPTB INFO checkpoint saved as nnsk.iter319
DEEPTB INFO Epoch 319 summary: train_loss: 0.010014
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep319
DEEPTB INFO iteration:320 train_loss: 0.009991 (0.010045) lr: 0.007268
DEEPTB INFO checkpoint saved as nnsk.iter320
DEEPTB INFO Epoch 320 summary: train_loss: 0.009991
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep320
DEEPTB INFO iteration:321 train_loss: 0.009969 (0.010022) lr: 0.00726
DEEPTB INFO checkpoint saved as nnsk.iter321
DEEPTB INFO Epoch 321 summary: train_loss: 0.009969
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep321
DEEPTB INFO iteration:322 train_loss: 0.009946 (0.009999) lr: 0.007253
DEEPTB INFO checkpoint saved as nnsk.iter322
DEEPTB INFO Epoch 322 summary: train_loss: 0.009946
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep322
DEEPTB INFO iteration:323 train_loss: 0.009924 (0.009977) lr: 0.007246
DEEPTB INFO checkpoint saved as nnsk.iter323
DEEPTB INFO Epoch 323 summary: train_loss: 0.009924
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep323
DEEPTB INFO iteration:324 train_loss: 0.009901 (0.009954) lr: 0.007239
DEEPTB INFO checkpoint saved as nnsk.iter324
DEEPTB INFO Epoch 324 summary: train_loss: 0.009901
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep324
DEEPTB INFO iteration:325 train_loss: 0.009878 (0.009931) lr: 0.007231
DEEPTB INFO checkpoint saved as nnsk.iter325
DEEPTB INFO Epoch 325 summary: train_loss: 0.009878
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep325
DEEPTB INFO iteration:326 train_loss: 0.009856 (0.009909) lr: 0.007224
DEEPTB INFO checkpoint saved as nnsk.iter326
DEEPTB INFO Epoch 326 summary: train_loss: 0.009856
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep326
DEEPTB INFO iteration:327 train_loss: 0.009834 (0.009886) lr: 0.007217
DEEPTB INFO checkpoint saved as nnsk.iter327
DEEPTB INFO Epoch 327 summary: train_loss: 0.009834
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep327
DEEPTB INFO iteration:328 train_loss: 0.009811 (0.009864) lr: 0.00721
DEEPTB INFO checkpoint saved as nnsk.iter328
DEEPTB INFO Epoch 328 summary: train_loss: 0.009811
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep328
DEEPTB INFO iteration:329 train_loss: 0.009789 (0.009841) lr: 0.007202
DEEPTB INFO checkpoint saved as nnsk.iter329
DEEPTB INFO Epoch 329 summary: train_loss: 0.009789
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep329
DEEPTB INFO iteration:330 train_loss: 0.009767 (0.009819) lr: 0.007195
DEEPTB INFO checkpoint saved as nnsk.iter330
DEEPTB INFO Epoch 330 summary: train_loss: 0.009767
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep330
DEEPTB INFO iteration:331 train_loss: 0.009745 (0.009797) lr: 0.007188
DEEPTB INFO checkpoint saved as nnsk.iter331
DEEPTB INFO Epoch 331 summary: train_loss: 0.009745
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep331
DEEPTB INFO iteration:332 train_loss: 0.009722 (0.009774) lr: 0.007181
DEEPTB INFO checkpoint saved as nnsk.iter332
DEEPTB INFO Epoch 332 summary: train_loss: 0.009722
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep332
DEEPTB INFO iteration:333 train_loss: 0.009700 (0.009752) lr: 0.007174
DEEPTB INFO checkpoint saved as nnsk.iter333
DEEPTB INFO Epoch 333 summary: train_loss: 0.009700
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep333
DEEPTB INFO iteration:334 train_loss: 0.009678 (0.009730) lr: 0.007167
DEEPTB INFO checkpoint saved as nnsk.iter334
DEEPTB INFO Epoch 334 summary: train_loss: 0.009678
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep334
DEEPTB INFO iteration:335 train_loss: 0.009656 (0.009708) lr: 0.007159
DEEPTB INFO checkpoint saved as nnsk.iter335
DEEPTB INFO Epoch 335 summary: train_loss: 0.009656
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep335
DEEPTB INFO iteration:336 train_loss: 0.009634 (0.009685) lr: 0.007152
DEEPTB INFO checkpoint saved as nnsk.iter336
DEEPTB INFO Epoch 336 summary: train_loss: 0.009634
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep336
DEEPTB INFO iteration:337 train_loss: 0.009612 (0.009664) lr: 0.007145
DEEPTB INFO checkpoint saved as nnsk.iter337
DEEPTB INFO Epoch 337 summary: train_loss: 0.009612
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep337
DEEPTB INFO iteration:338 train_loss: 0.009590 (0.009641) lr: 0.007138
DEEPTB INFO checkpoint saved as nnsk.iter338
DEEPTB INFO Epoch 338 summary: train_loss: 0.009590
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep338
DEEPTB INFO iteration:339 train_loss: 0.009568 (0.009620) lr: 0.007131
DEEPTB INFO checkpoint saved as nnsk.iter339
DEEPTB INFO Epoch 339 summary: train_loss: 0.009568
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep339
DEEPTB INFO iteration:340 train_loss: 0.009547 (0.009598) lr: 0.007124
DEEPTB INFO checkpoint saved as nnsk.iter340
DEEPTB INFO Epoch 340 summary: train_loss: 0.009547
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep340
DEEPTB INFO iteration:341 train_loss: 0.009525 (0.009576) lr: 0.007116
DEEPTB INFO checkpoint saved as nnsk.iter341
DEEPTB INFO Epoch 341 summary: train_loss: 0.009525
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep341
DEEPTB INFO iteration:342 train_loss: 0.009503 (0.009554) lr: 0.007109
DEEPTB INFO checkpoint saved as nnsk.iter342
DEEPTB INFO Epoch 342 summary: train_loss: 0.009503
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep342
DEEPTB INFO iteration:343 train_loss: 0.009481 (0.009532) lr: 0.007102
DEEPTB INFO checkpoint saved as nnsk.iter343
DEEPTB INFO Epoch 343 summary: train_loss: 0.009481
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep343
DEEPTB INFO iteration:344 train_loss: 0.009459 (0.009510) lr: 0.007095
DEEPTB INFO checkpoint saved as nnsk.iter344
DEEPTB INFO Epoch 344 summary: train_loss: 0.009459
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep344
DEEPTB INFO iteration:345 train_loss: 0.009438 (0.009489) lr: 0.007088
DEEPTB INFO checkpoint saved as nnsk.iter345
DEEPTB INFO Epoch 345 summary: train_loss: 0.009438
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep345
DEEPTB INFO iteration:346 train_loss: 0.009416 (0.009467) lr: 0.007081
DEEPTB INFO checkpoint saved as nnsk.iter346
DEEPTB INFO Epoch 346 summary: train_loss: 0.009416
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep346
DEEPTB INFO iteration:347 train_loss: 0.009395 (0.009445) lr: 0.007074
DEEPTB INFO checkpoint saved as nnsk.iter347
DEEPTB INFO Epoch 347 summary: train_loss: 0.009395
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep347
DEEPTB INFO iteration:348 train_loss: 0.009373 (0.009424) lr: 0.007067
DEEPTB INFO checkpoint saved as nnsk.iter348
DEEPTB INFO Epoch 348 summary: train_loss: 0.009373
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep348
DEEPTB INFO iteration:349 train_loss: 0.009352 (0.009402) lr: 0.00706
DEEPTB INFO checkpoint saved as nnsk.iter349
DEEPTB INFO Epoch 349 summary: train_loss: 0.009352
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep349
DEEPTB INFO iteration:350 train_loss: 0.009331 (0.009381) lr: 0.007053
DEEPTB INFO checkpoint saved as nnsk.iter350
DEEPTB INFO Epoch 350 summary: train_loss: 0.009331
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep350
DEEPTB INFO iteration:351 train_loss: 0.009309 (0.009359) lr: 0.007046
DEEPTB INFO checkpoint saved as nnsk.iter351
DEEPTB INFO Epoch 351 summary: train_loss: 0.009309
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep351
DEEPTB INFO iteration:352 train_loss: 0.009288 (0.009338) lr: 0.007039
DEEPTB INFO checkpoint saved as nnsk.iter352
DEEPTB INFO Epoch 352 summary: train_loss: 0.009288
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep352
DEEPTB INFO iteration:353 train_loss: 0.009267 (0.009316) lr: 0.007032
DEEPTB INFO checkpoint saved as nnsk.iter353
DEEPTB INFO Epoch 353 summary: train_loss: 0.009267
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep353
DEEPTB INFO iteration:354 train_loss: 0.009245 (0.009295) lr: 0.007025
DEEPTB INFO checkpoint saved as nnsk.iter354
DEEPTB INFO Epoch 354 summary: train_loss: 0.009245
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep354
DEEPTB INFO iteration:355 train_loss: 0.009224 (0.009274) lr: 0.007018
DEEPTB INFO checkpoint saved as nnsk.iter355
DEEPTB INFO Epoch 355 summary: train_loss: 0.009224
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep355
DEEPTB INFO iteration:356 train_loss: 0.009203 (0.009252) lr: 0.00701
DEEPTB INFO checkpoint saved as nnsk.iter356
DEEPTB INFO Epoch 356 summary: train_loss: 0.009203
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep356
DEEPTB INFO iteration:357 train_loss: 0.009181 (0.009231) lr: 0.007003
DEEPTB INFO checkpoint saved as nnsk.iter357
DEEPTB INFO Epoch 357 summary: train_loss: 0.009181
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep357
DEEPTB INFO iteration:358 train_loss: 0.009160 (0.009210) lr: 0.006996
DEEPTB INFO checkpoint saved as nnsk.iter358
DEEPTB INFO Epoch 358 summary: train_loss: 0.009160
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep358
DEEPTB INFO iteration:359 train_loss: 0.009139 (0.009189) lr: 0.006989
DEEPTB INFO checkpoint saved as nnsk.iter359
DEEPTB INFO Epoch 359 summary: train_loss: 0.009139
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep359
DEEPTB INFO iteration:360 train_loss: 0.009118 (0.009167) lr: 0.006982
DEEPTB INFO checkpoint saved as nnsk.iter360
DEEPTB INFO Epoch 360 summary: train_loss: 0.009118
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep360
DEEPTB INFO iteration:361 train_loss: 0.009097 (0.009146) lr: 0.006976
DEEPTB INFO checkpoint saved as nnsk.iter361
DEEPTB INFO Epoch 361 summary: train_loss: 0.009097
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep361
DEEPTB INFO iteration:362 train_loss: 0.009076 (0.009125) lr: 0.006969
DEEPTB INFO checkpoint saved as nnsk.iter362
DEEPTB INFO Epoch 362 summary: train_loss: 0.009076
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep362
DEEPTB INFO iteration:363 train_loss: 0.009055 (0.009104) lr: 0.006962
DEEPTB INFO checkpoint saved as nnsk.iter363
DEEPTB INFO Epoch 363 summary: train_loss: 0.009055
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep363
DEEPTB INFO iteration:364 train_loss: 0.009033 (0.009083) lr: 0.006955
DEEPTB INFO checkpoint saved as nnsk.iter364
DEEPTB INFO Epoch 364 summary: train_loss: 0.009033
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep364
DEEPTB INFO iteration:365 train_loss: 0.009012 (0.009062) lr: 0.006948
DEEPTB INFO checkpoint saved as nnsk.iter365
DEEPTB INFO Epoch 365 summary: train_loss: 0.009012
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep365
DEEPTB INFO iteration:366 train_loss: 0.008991 (0.009041) lr: 0.006941
DEEPTB INFO checkpoint saved as nnsk.iter366
DEEPTB INFO Epoch 366 summary: train_loss: 0.008991
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep366
DEEPTB INFO iteration:367 train_loss: 0.008970 (0.009019) lr: 0.006934
DEEPTB INFO checkpoint saved as nnsk.iter367
DEEPTB INFO Epoch 367 summary: train_loss: 0.008970
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep367
DEEPTB INFO iteration:368 train_loss: 0.008949 (0.008998) lr: 0.006927
DEEPTB INFO checkpoint saved as nnsk.iter368
DEEPTB INFO Epoch 368 summary: train_loss: 0.008949
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep368
DEEPTB INFO iteration:369 train_loss: 0.008928 (0.008977) lr: 0.00692
DEEPTB INFO checkpoint saved as nnsk.iter369
DEEPTB INFO Epoch 369 summary: train_loss: 0.008928
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep369
DEEPTB INFO iteration:370 train_loss: 0.008907 (0.008956) lr: 0.006913
DEEPTB INFO checkpoint saved as nnsk.iter370
DEEPTB INFO Epoch 370 summary: train_loss: 0.008907
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep370
DEEPTB INFO iteration:371 train_loss: 0.008887 (0.008935) lr: 0.006906
DEEPTB INFO checkpoint saved as nnsk.iter371
DEEPTB INFO Epoch 371 summary: train_loss: 0.008887
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep371
DEEPTB INFO iteration:372 train_loss: 0.008865 (0.008914) lr: 0.006899
DEEPTB INFO checkpoint saved as nnsk.iter372
DEEPTB INFO Epoch 372 summary: train_loss: 0.008865
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep372
DEEPTB INFO iteration:373 train_loss: 0.008845 (0.008893) lr: 0.006892
DEEPTB INFO checkpoint saved as nnsk.iter373
DEEPTB INFO Epoch 373 summary: train_loss: 0.008845
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep373
DEEPTB INFO iteration:374 train_loss: 0.008824 (0.008873) lr: 0.006885
DEEPTB INFO checkpoint saved as nnsk.iter374
DEEPTB INFO Epoch 374 summary: train_loss: 0.008824
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep374
DEEPTB INFO iteration:375 train_loss: 0.008803 (0.008852) lr: 0.006878
DEEPTB INFO checkpoint saved as nnsk.iter375
DEEPTB INFO Epoch 375 summary: train_loss: 0.008803
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep375
DEEPTB INFO iteration:376 train_loss: 0.008782 (0.008831) lr: 0.006872
DEEPTB INFO checkpoint saved as nnsk.iter376
DEEPTB INFO Epoch 376 summary: train_loss: 0.008782
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep376
DEEPTB INFO iteration:377 train_loss: 0.008761 (0.008810) lr: 0.006865
DEEPTB INFO checkpoint saved as nnsk.iter377
DEEPTB INFO Epoch 377 summary: train_loss: 0.008761
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep377
DEEPTB INFO iteration:378 train_loss: 0.008740 (0.008789) lr: 0.006858
DEEPTB INFO checkpoint saved as nnsk.iter378
DEEPTB INFO Epoch 378 summary: train_loss: 0.008740
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep378
DEEPTB INFO iteration:379 train_loss: 0.008720 (0.008768) lr: 0.006851
DEEPTB INFO checkpoint saved as nnsk.iter379
DEEPTB INFO Epoch 379 summary: train_loss: 0.008720
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep379
DEEPTB INFO iteration:380 train_loss: 0.008699 (0.008747) lr: 0.006844
DEEPTB INFO checkpoint saved as nnsk.iter380
DEEPTB INFO Epoch 380 summary: train_loss: 0.008699
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep380
DEEPTB INFO iteration:381 train_loss: 0.008678 (0.008727) lr: 0.006837
DEEPTB INFO checkpoint saved as nnsk.iter381
DEEPTB INFO Epoch 381 summary: train_loss: 0.008678
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep381
DEEPTB INFO iteration:382 train_loss: 0.008657 (0.008706) lr: 0.00683
DEEPTB INFO checkpoint saved as nnsk.iter382
DEEPTB INFO Epoch 382 summary: train_loss: 0.008657
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep382
DEEPTB INFO iteration:383 train_loss: 0.008637 (0.008685) lr: 0.006824
DEEPTB INFO checkpoint saved as nnsk.iter383
DEEPTB INFO Epoch 383 summary: train_loss: 0.008637
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep383
DEEPTB INFO iteration:384 train_loss: 0.008616 (0.008664) lr: 0.006817
DEEPTB INFO checkpoint saved as nnsk.iter384
DEEPTB INFO Epoch 384 summary: train_loss: 0.008616
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep384
DEEPTB INFO iteration:385 train_loss: 0.008596 (0.008644) lr: 0.00681
DEEPTB INFO checkpoint saved as nnsk.iter385
DEEPTB INFO Epoch 385 summary: train_loss: 0.008596
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep385
DEEPTB INFO iteration:386 train_loss: 0.008575 (0.008623) lr: 0.006803
DEEPTB INFO checkpoint saved as nnsk.iter386
DEEPTB INFO Epoch 386 summary: train_loss: 0.008575
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep386
DEEPTB INFO iteration:387 train_loss: 0.008555 (0.008603) lr: 0.006796
DEEPTB INFO checkpoint saved as nnsk.iter387
DEEPTB INFO Epoch 387 summary: train_loss: 0.008555
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep387
DEEPTB INFO iteration:388 train_loss: 0.008534 (0.008582) lr: 0.00679
DEEPTB INFO checkpoint saved as nnsk.iter388
DEEPTB INFO Epoch 388 summary: train_loss: 0.008534
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep388
DEEPTB INFO iteration:389 train_loss: 0.008514 (0.008561) lr: 0.006783
DEEPTB INFO checkpoint saved as nnsk.iter389
DEEPTB INFO Epoch 389 summary: train_loss: 0.008514
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep389
DEEPTB INFO iteration:390 train_loss: 0.008493 (0.008541) lr: 0.006776
DEEPTB INFO checkpoint saved as nnsk.iter390
DEEPTB INFO Epoch 390 summary: train_loss: 0.008493
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep390
DEEPTB INFO iteration:391 train_loss: 0.008473 (0.008520) lr: 0.006769
DEEPTB INFO checkpoint saved as nnsk.iter391
DEEPTB INFO Epoch 391 summary: train_loss: 0.008473
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep391
DEEPTB INFO iteration:392 train_loss: 0.008452 (0.008500) lr: 0.006762
DEEPTB INFO checkpoint saved as nnsk.iter392
DEEPTB INFO Epoch 392 summary: train_loss: 0.008452
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep392
DEEPTB INFO iteration:393 train_loss: 0.008432 (0.008479) lr: 0.006756
DEEPTB INFO checkpoint saved as nnsk.iter393
DEEPTB INFO Epoch 393 summary: train_loss: 0.008432
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep393
DEEPTB INFO iteration:394 train_loss: 0.008411 (0.008459) lr: 0.006749
DEEPTB INFO checkpoint saved as nnsk.iter394
DEEPTB INFO Epoch 394 summary: train_loss: 0.008411
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep394
DEEPTB INFO iteration:395 train_loss: 0.008391 (0.008438) lr: 0.006742
DEEPTB INFO checkpoint saved as nnsk.iter395
DEEPTB INFO Epoch 395 summary: train_loss: 0.008391
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep395
DEEPTB INFO iteration:396 train_loss: 0.008371 (0.008418) lr: 0.006735
DEEPTB INFO checkpoint saved as nnsk.iter396
DEEPTB INFO Epoch 396 summary: train_loss: 0.008371
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep396
DEEPTB INFO iteration:397 train_loss: 0.008351 (0.008398) lr: 0.006729
DEEPTB INFO checkpoint saved as nnsk.iter397
DEEPTB INFO Epoch 397 summary: train_loss: 0.008351
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep397
DEEPTB INFO iteration:398 train_loss: 0.008332 (0.008378) lr: 0.006722
DEEPTB INFO checkpoint saved as nnsk.iter398
DEEPTB INFO Epoch 398 summary: train_loss: 0.008332
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep398
DEEPTB INFO iteration:399 train_loss: 0.008312 (0.008358) lr: 0.006715
DEEPTB INFO checkpoint saved as nnsk.iter399
DEEPTB INFO Epoch 399 summary: train_loss: 0.008312
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep399
DEEPTB INFO iteration:400 train_loss: 0.008292 (0.008339) lr: 0.006709
DEEPTB INFO checkpoint saved as nnsk.iter400
DEEPTB INFO Epoch 400 summary: train_loss: 0.008292
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep400
DEEPTB INFO iteration:401 train_loss: 0.008272 (0.008319) lr: 0.006702
DEEPTB INFO checkpoint saved as nnsk.iter401
DEEPTB INFO Epoch 401 summary: train_loss: 0.008272
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep401
DEEPTB INFO iteration:402 train_loss: 0.008252 (0.008299) lr: 0.006695
DEEPTB INFO checkpoint saved as nnsk.iter402
DEEPTB INFO Epoch 402 summary: train_loss: 0.008252
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep402
DEEPTB INFO iteration:403 train_loss: 0.008232 (0.008279) lr: 0.006688
DEEPTB INFO checkpoint saved as nnsk.iter403
DEEPTB INFO Epoch 403 summary: train_loss: 0.008232
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep403
DEEPTB INFO iteration:404 train_loss: 0.008212 (0.008259) lr: 0.006682
DEEPTB INFO checkpoint saved as nnsk.iter404
DEEPTB INFO Epoch 404 summary: train_loss: 0.008212
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep404
DEEPTB INFO iteration:405 train_loss: 0.008192 (0.008239) lr: 0.006675
DEEPTB INFO checkpoint saved as nnsk.iter405
DEEPTB INFO Epoch 405 summary: train_loss: 0.008192
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep405
DEEPTB INFO iteration:406 train_loss: 0.008172 (0.008219) lr: 0.006668
DEEPTB INFO checkpoint saved as nnsk.iter406
DEEPTB INFO Epoch 406 summary: train_loss: 0.008172
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep406
DEEPTB INFO iteration:407 train_loss: 0.008152 (0.008199) lr: 0.006662
DEEPTB INFO checkpoint saved as nnsk.iter407
DEEPTB INFO Epoch 407 summary: train_loss: 0.008152
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep407
DEEPTB INFO iteration:408 train_loss: 0.008133 (0.008179) lr: 0.006655
DEEPTB INFO checkpoint saved as nnsk.iter408
DEEPTB INFO Epoch 408 summary: train_loss: 0.008133
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep408
DEEPTB INFO iteration:409 train_loss: 0.008113 (0.008159) lr: 0.006648
DEEPTB INFO checkpoint saved as nnsk.iter409
DEEPTB INFO Epoch 409 summary: train_loss: 0.008113
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep409
DEEPTB INFO iteration:410 train_loss: 0.008094 (0.008140) lr: 0.006642
DEEPTB INFO checkpoint saved as nnsk.iter410
DEEPTB INFO Epoch 410 summary: train_loss: 0.008094
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep410
DEEPTB INFO iteration:411 train_loss: 0.008074 (0.008120) lr: 0.006635
DEEPTB INFO checkpoint saved as nnsk.iter411
DEEPTB INFO Epoch 411 summary: train_loss: 0.008074
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep411
DEEPTB INFO iteration:412 train_loss: 0.008054 (0.008100) lr: 0.006629
DEEPTB INFO checkpoint saved as nnsk.iter412
DEEPTB INFO Epoch 412 summary: train_loss: 0.008054
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep412
DEEPTB INFO iteration:413 train_loss: 0.008035 (0.008080) lr: 0.006622
DEEPTB INFO checkpoint saved as nnsk.iter413
DEEPTB INFO Epoch 413 summary: train_loss: 0.008035
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep413
DEEPTB INFO iteration:414 train_loss: 0.008015 (0.008061) lr: 0.006615
DEEPTB INFO checkpoint saved as nnsk.iter414
DEEPTB INFO Epoch 414 summary: train_loss: 0.008015
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep414
DEEPTB INFO iteration:415 train_loss: 0.007995 (0.008041) lr: 0.006609
DEEPTB INFO checkpoint saved as nnsk.iter415
DEEPTB INFO Epoch 415 summary: train_loss: 0.007995
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep415
DEEPTB INFO iteration:416 train_loss: 0.007975 (0.008021) lr: 0.006602
DEEPTB INFO checkpoint saved as nnsk.iter416
DEEPTB INFO Epoch 416 summary: train_loss: 0.007975
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep416
DEEPTB INFO iteration:417 train_loss: 0.007955 (0.008002) lr: 0.006595
DEEPTB INFO checkpoint saved as nnsk.iter417
DEEPTB INFO Epoch 417 summary: train_loss: 0.007955
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep417
DEEPTB INFO iteration:418 train_loss: 0.007936 (0.007982) lr: 0.006589
DEEPTB INFO checkpoint saved as nnsk.iter418
DEEPTB INFO Epoch 418 summary: train_loss: 0.007936
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep418
DEEPTB INFO iteration:419 train_loss: 0.007917 (0.007962) lr: 0.006582
DEEPTB INFO checkpoint saved as nnsk.iter419
DEEPTB INFO Epoch 419 summary: train_loss: 0.007917
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep419
DEEPTB INFO iteration:420 train_loss: 0.007897 (0.007943) lr: 0.006576
DEEPTB INFO checkpoint saved as nnsk.iter420
DEEPTB INFO Epoch 420 summary: train_loss: 0.007897
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep420
DEEPTB INFO iteration:421 train_loss: 0.007878 (0.007923) lr: 0.006569
DEEPTB INFO checkpoint saved as nnsk.iter421
DEEPTB INFO Epoch 421 summary: train_loss: 0.007878
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep421
DEEPTB INFO iteration:422 train_loss: 0.007858 (0.007904) lr: 0.006563
DEEPTB INFO checkpoint saved as nnsk.iter422
DEEPTB INFO Epoch 422 summary: train_loss: 0.007858
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep422
DEEPTB INFO iteration:423 train_loss: 0.007838 (0.007884) lr: 0.006556
DEEPTB INFO checkpoint saved as nnsk.iter423
DEEPTB INFO Epoch 423 summary: train_loss: 0.007838
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep423
DEEPTB INFO iteration:424 train_loss: 0.007819 (0.007864) lr: 0.006549
DEEPTB INFO checkpoint saved as nnsk.iter424
DEEPTB INFO Epoch 424 summary: train_loss: 0.007819
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep424
DEEPTB INFO iteration:425 train_loss: 0.007799 (0.007845) lr: 0.006543
DEEPTB INFO checkpoint saved as nnsk.iter425
DEEPTB INFO Epoch 425 summary: train_loss: 0.007799
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep425
DEEPTB INFO iteration:426 train_loss: 0.007780 (0.007825) lr: 0.006536
DEEPTB INFO checkpoint saved as nnsk.iter426
DEEPTB INFO Epoch 426 summary: train_loss: 0.007780
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep426
DEEPTB INFO iteration:427 train_loss: 0.007761 (0.007806) lr: 0.00653
DEEPTB INFO checkpoint saved as nnsk.iter427
DEEPTB INFO Epoch 427 summary: train_loss: 0.007761
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep427
DEEPTB INFO iteration:428 train_loss: 0.007741 (0.007787) lr: 0.006523
DEEPTB INFO checkpoint saved as nnsk.iter428
DEEPTB INFO Epoch 428 summary: train_loss: 0.007741
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep428
DEEPTB INFO iteration:429 train_loss: 0.007722 (0.007767) lr: 0.006517
DEEPTB INFO checkpoint saved as nnsk.iter429
DEEPTB INFO Epoch 429 summary: train_loss: 0.007722
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep429
DEEPTB INFO iteration:430 train_loss: 0.007702 (0.007748) lr: 0.00651
DEEPTB INFO checkpoint saved as nnsk.iter430
DEEPTB INFO Epoch 430 summary: train_loss: 0.007702
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep430
DEEPTB INFO iteration:431 train_loss: 0.007683 (0.007728) lr: 0.006504
DEEPTB INFO checkpoint saved as nnsk.iter431
DEEPTB INFO Epoch 431 summary: train_loss: 0.007683
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep431
DEEPTB INFO iteration:432 train_loss: 0.007664 (0.007709) lr: 0.006497
DEEPTB INFO checkpoint saved as nnsk.iter432
DEEPTB INFO Epoch 432 summary: train_loss: 0.007664
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep432
DEEPTB INFO iteration:433 train_loss: 0.007645 (0.007690) lr: 0.006491
DEEPTB INFO checkpoint saved as nnsk.iter433
DEEPTB INFO Epoch 433 summary: train_loss: 0.007645
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep433
DEEPTB INFO iteration:434 train_loss: 0.007626 (0.007671) lr: 0.006484
DEEPTB INFO checkpoint saved as nnsk.iter434
DEEPTB INFO Epoch 434 summary: train_loss: 0.007626
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep434
DEEPTB INFO iteration:435 train_loss: 0.007606 (0.007651) lr: 0.006478
DEEPTB INFO checkpoint saved as nnsk.iter435
DEEPTB INFO Epoch 435 summary: train_loss: 0.007606
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep435
DEEPTB INFO iteration:436 train_loss: 0.007587 (0.007632) lr: 0.006471
DEEPTB INFO checkpoint saved as nnsk.iter436
DEEPTB INFO Epoch 436 summary: train_loss: 0.007587
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep436
DEEPTB INFO iteration:437 train_loss: 0.007568 (0.007613) lr: 0.006465
DEEPTB INFO checkpoint saved as nnsk.iter437
DEEPTB INFO Epoch 437 summary: train_loss: 0.007568
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep437
DEEPTB INFO iteration:438 train_loss: 0.007549 (0.007594) lr: 0.006458
DEEPTB INFO checkpoint saved as nnsk.iter438
DEEPTB INFO Epoch 438 summary: train_loss: 0.007549
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep438
DEEPTB INFO iteration:439 train_loss: 0.007530 (0.007575) lr: 0.006452
DEEPTB INFO checkpoint saved as nnsk.iter439
DEEPTB INFO Epoch 439 summary: train_loss: 0.007530
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep439
DEEPTB INFO iteration:440 train_loss: 0.007511 (0.007556) lr: 0.006445
DEEPTB INFO checkpoint saved as nnsk.iter440
DEEPTB INFO Epoch 440 summary: train_loss: 0.007511
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep440
DEEPTB INFO iteration:441 train_loss: 0.007492 (0.007537) lr: 0.006439
DEEPTB INFO checkpoint saved as nnsk.iter441
DEEPTB INFO Epoch 441 summary: train_loss: 0.007492
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep441
DEEPTB INFO iteration:442 train_loss: 0.007473 (0.007518) lr: 0.006433
DEEPTB INFO checkpoint saved as nnsk.iter442
DEEPTB INFO Epoch 442 summary: train_loss: 0.007473
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep442
DEEPTB INFO iteration:443 train_loss: 0.007454 (0.007499) lr: 0.006426
DEEPTB INFO checkpoint saved as nnsk.iter443
DEEPTB INFO Epoch 443 summary: train_loss: 0.007454
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep443
DEEPTB INFO iteration:444 train_loss: 0.007435 (0.007480) lr: 0.00642
DEEPTB INFO checkpoint saved as nnsk.iter444
DEEPTB INFO Epoch 444 summary: train_loss: 0.007435
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep444
DEEPTB INFO iteration:445 train_loss: 0.007417 (0.007461) lr: 0.006413
DEEPTB INFO checkpoint saved as nnsk.iter445
DEEPTB INFO Epoch 445 summary: train_loss: 0.007417
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep445
DEEPTB INFO iteration:446 train_loss: 0.007398 (0.007442) lr: 0.006407
DEEPTB INFO checkpoint saved as nnsk.iter446
DEEPTB INFO Epoch 446 summary: train_loss: 0.007398
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep446
DEEPTB INFO iteration:447 train_loss: 0.007379 (0.007423) lr: 0.0064
DEEPTB INFO checkpoint saved as nnsk.iter447
DEEPTB INFO Epoch 447 summary: train_loss: 0.007379
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep447
DEEPTB INFO iteration:448 train_loss: 0.007360 (0.007404) lr: 0.006394
DEEPTB INFO checkpoint saved as nnsk.iter448
DEEPTB INFO Epoch 448 summary: train_loss: 0.007360
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep448
DEEPTB INFO iteration:449 train_loss: 0.007342 (0.007386) lr: 0.006388
DEEPTB INFO checkpoint saved as nnsk.iter449
DEEPTB INFO Epoch 449 summary: train_loss: 0.007342
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep449
DEEPTB INFO iteration:450 train_loss: 0.007323 (0.007367) lr: 0.006381
DEEPTB INFO checkpoint saved as nnsk.iter450
DEEPTB INFO Epoch 450 summary: train_loss: 0.007323
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep450
DEEPTB INFO iteration:451 train_loss: 0.007305 (0.007348) lr: 0.006375
DEEPTB INFO checkpoint saved as nnsk.iter451
DEEPTB INFO Epoch 451 summary: train_loss: 0.007305
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep451
DEEPTB INFO iteration:452 train_loss: 0.007286 (0.007330) lr: 0.006368
DEEPTB INFO checkpoint saved as nnsk.iter452
DEEPTB INFO Epoch 452 summary: train_loss: 0.007286
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep452
DEEPTB INFO iteration:453 train_loss: 0.007268 (0.007311) lr: 0.006362
DEEPTB INFO checkpoint saved as nnsk.iter453
DEEPTB INFO Epoch 453 summary: train_loss: 0.007268
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep453
DEEPTB INFO iteration:454 train_loss: 0.007250 (0.007293) lr: 0.006356
DEEPTB INFO checkpoint saved as nnsk.iter454
DEEPTB INFO Epoch 454 summary: train_loss: 0.007250
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep454
DEEPTB INFO iteration:455 train_loss: 0.007231 (0.007274) lr: 0.006349
DEEPTB INFO checkpoint saved as nnsk.iter455
DEEPTB INFO Epoch 455 summary: train_loss: 0.007231
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep455
DEEPTB INFO iteration:456 train_loss: 0.007213 (0.007256) lr: 0.006343
DEEPTB INFO checkpoint saved as nnsk.iter456
DEEPTB INFO Epoch 456 summary: train_loss: 0.007213
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep456
DEEPTB INFO iteration:457 train_loss: 0.007195 (0.007237) lr: 0.006337
DEEPTB INFO checkpoint saved as nnsk.iter457
DEEPTB INFO Epoch 457 summary: train_loss: 0.007195
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep457
DEEPTB INFO iteration:458 train_loss: 0.007176 (0.007219) lr: 0.00633
DEEPTB INFO checkpoint saved as nnsk.iter458
DEEPTB INFO Epoch 458 summary: train_loss: 0.007176
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep458
DEEPTB INFO iteration:459 train_loss: 0.007158 (0.007201) lr: 0.006324
DEEPTB INFO checkpoint saved as nnsk.iter459
DEEPTB INFO Epoch 459 summary: train_loss: 0.007158
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep459
DEEPTB INFO iteration:460 train_loss: 0.007140 (0.007183) lr: 0.006318
DEEPTB INFO checkpoint saved as nnsk.iter460
DEEPTB INFO Epoch 460 summary: train_loss: 0.007140
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep460
DEEPTB INFO iteration:461 train_loss: 0.007123 (0.007165) lr: 0.006311
DEEPTB INFO checkpoint saved as nnsk.iter461
DEEPTB INFO Epoch 461 summary: train_loss: 0.007123
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep461
DEEPTB INFO iteration:462 train_loss: 0.007105 (0.007147) lr: 0.006305
DEEPTB INFO checkpoint saved as nnsk.iter462
DEEPTB INFO Epoch 462 summary: train_loss: 0.007105
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep462
DEEPTB INFO iteration:463 train_loss: 0.007087 (0.007129) lr: 0.006299
DEEPTB INFO checkpoint saved as nnsk.iter463
DEEPTB INFO Epoch 463 summary: train_loss: 0.007087
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep463
DEEPTB INFO iteration:464 train_loss: 0.007069 (0.007111) lr: 0.006292
DEEPTB INFO checkpoint saved as nnsk.iter464
DEEPTB INFO Epoch 464 summary: train_loss: 0.007069
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep464
DEEPTB INFO iteration:465 train_loss: 0.007051 (0.007093) lr: 0.006286
DEEPTB INFO checkpoint saved as nnsk.iter465
DEEPTB INFO Epoch 465 summary: train_loss: 0.007051
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep465
DEEPTB INFO iteration:466 train_loss: 0.007034 (0.007075) lr: 0.00628
DEEPTB INFO checkpoint saved as nnsk.iter466
DEEPTB INFO Epoch 466 summary: train_loss: 0.007034
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep466
DEEPTB INFO iteration:467 train_loss: 0.007016 (0.007057) lr: 0.006274
DEEPTB INFO checkpoint saved as nnsk.iter467
DEEPTB INFO Epoch 467 summary: train_loss: 0.007016
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep467
DEEPTB INFO iteration:468 train_loss: 0.006999 (0.007040) lr: 0.006267
DEEPTB INFO checkpoint saved as nnsk.iter468
DEEPTB INFO Epoch 468 summary: train_loss: 0.006999
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep468
DEEPTB INFO iteration:469 train_loss: 0.006981 (0.007022) lr: 0.006261
DEEPTB INFO checkpoint saved as nnsk.iter469
DEEPTB INFO Epoch 469 summary: train_loss: 0.006981
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep469
DEEPTB INFO iteration:470 train_loss: 0.006964 (0.007005) lr: 0.006255
DEEPTB INFO checkpoint saved as nnsk.iter470
DEEPTB INFO Epoch 470 summary: train_loss: 0.006964
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep470
DEEPTB INFO iteration:471 train_loss: 0.006947 (0.006987) lr: 0.006249
DEEPTB INFO checkpoint saved as nnsk.iter471
DEEPTB INFO Epoch 471 summary: train_loss: 0.006947
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep471
DEEPTB INFO iteration:472 train_loss: 0.006929 (0.006970) lr: 0.006242
DEEPTB INFO checkpoint saved as nnsk.iter472
DEEPTB INFO Epoch 472 summary: train_loss: 0.006929
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep472
DEEPTB INFO iteration:473 train_loss: 0.006912 (0.006953) lr: 0.006236
DEEPTB INFO checkpoint saved as nnsk.iter473
DEEPTB INFO Epoch 473 summary: train_loss: 0.006912
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep473
DEEPTB INFO iteration:474 train_loss: 0.006895 (0.006935) lr: 0.00623
DEEPTB INFO checkpoint saved as nnsk.iter474
DEEPTB INFO Epoch 474 summary: train_loss: 0.006895
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep474
DEEPTB INFO iteration:475 train_loss: 0.006878 (0.006918) lr: 0.006224
DEEPTB INFO checkpoint saved as nnsk.iter475
DEEPTB INFO Epoch 475 summary: train_loss: 0.006878
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep475
DEEPTB INFO iteration:476 train_loss: 0.006861 (0.006901) lr: 0.006217
DEEPTB INFO checkpoint saved as nnsk.iter476
DEEPTB INFO Epoch 476 summary: train_loss: 0.006861
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep476
DEEPTB INFO iteration:477 train_loss: 0.006845 (0.006884) lr: 0.006211
DEEPTB INFO checkpoint saved as nnsk.iter477
DEEPTB INFO Epoch 477 summary: train_loss: 0.006845
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep477
DEEPTB INFO iteration:478 train_loss: 0.006828 (0.006867) lr: 0.006205
DEEPTB INFO checkpoint saved as nnsk.iter478
DEEPTB INFO Epoch 478 summary: train_loss: 0.006828
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep478
DEEPTB INFO iteration:479 train_loss: 0.006811 (0.006850) lr: 0.006199
DEEPTB INFO checkpoint saved as nnsk.iter479
DEEPTB INFO Epoch 479 summary: train_loss: 0.006811
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep479
DEEPTB INFO iteration:480 train_loss: 0.006795 (0.006834) lr: 0.006193
DEEPTB INFO checkpoint saved as nnsk.iter480
DEEPTB INFO Epoch 480 summary: train_loss: 0.006795
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep480
DEEPTB INFO iteration:481 train_loss: 0.006778 (0.006817) lr: 0.006186
DEEPTB INFO checkpoint saved as nnsk.iter481
DEEPTB INFO Epoch 481 summary: train_loss: 0.006778
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep481
DEEPTB INFO iteration:482 train_loss: 0.006762 (0.006800) lr: 0.00618
DEEPTB INFO checkpoint saved as nnsk.iter482
DEEPTB INFO Epoch 482 summary: train_loss: 0.006762
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep482
DEEPTB INFO iteration:483 train_loss: 0.006745 (0.006784) lr: 0.006174
DEEPTB INFO checkpoint saved as nnsk.iter483
DEEPTB INFO Epoch 483 summary: train_loss: 0.006745
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep483
DEEPTB INFO iteration:484 train_loss: 0.006729 (0.006767) lr: 0.006168
DEEPTB INFO checkpoint saved as nnsk.iter484
DEEPTB INFO Epoch 484 summary: train_loss: 0.006729
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep484
DEEPTB INFO iteration:485 train_loss: 0.006713 (0.006751) lr: 0.006162
DEEPTB INFO checkpoint saved as nnsk.iter485
DEEPTB INFO Epoch 485 summary: train_loss: 0.006713
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep485
DEEPTB INFO iteration:486 train_loss: 0.006697 (0.006735) lr: 0.006155
DEEPTB INFO checkpoint saved as nnsk.iter486
DEEPTB INFO Epoch 486 summary: train_loss: 0.006697
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep486
DEEPTB INFO iteration:487 train_loss: 0.006681 (0.006719) lr: 0.006149
DEEPTB INFO checkpoint saved as nnsk.iter487
DEEPTB INFO Epoch 487 summary: train_loss: 0.006681
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep487
DEEPTB INFO iteration:488 train_loss: 0.006665 (0.006702) lr: 0.006143
DEEPTB INFO checkpoint saved as nnsk.iter488
DEEPTB INFO Epoch 488 summary: train_loss: 0.006665
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep488
DEEPTB INFO iteration:489 train_loss: 0.006649 (0.006686) lr: 0.006137
DEEPTB INFO checkpoint saved as nnsk.iter489
DEEPTB INFO Epoch 489 summary: train_loss: 0.006649
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep489
DEEPTB INFO iteration:490 train_loss: 0.006633 (0.006670) lr: 0.006131
DEEPTB INFO checkpoint saved as nnsk.iter490
DEEPTB INFO Epoch 490 summary: train_loss: 0.006633
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep490
DEEPTB INFO iteration:491 train_loss: 0.006618 (0.006655) lr: 0.006125
DEEPTB INFO checkpoint saved as nnsk.iter491
DEEPTB INFO Epoch 491 summary: train_loss: 0.006618
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep491
DEEPTB INFO iteration:492 train_loss: 0.006602 (0.006639) lr: 0.006119
DEEPTB INFO checkpoint saved as nnsk.iter492
DEEPTB INFO Epoch 492 summary: train_loss: 0.006602
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep492
DEEPTB INFO iteration:493 train_loss: 0.006586 (0.006623) lr: 0.006113
DEEPTB INFO checkpoint saved as nnsk.iter493
DEEPTB INFO Epoch 493 summary: train_loss: 0.006586
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep493
DEEPTB INFO iteration:494 train_loss: 0.006571 (0.006607) lr: 0.006106
DEEPTB INFO checkpoint saved as nnsk.iter494
DEEPTB INFO Epoch 494 summary: train_loss: 0.006571
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep494
DEEPTB INFO iteration:495 train_loss: 0.006556 (0.006592) lr: 0.0061
DEEPTB INFO checkpoint saved as nnsk.iter495
DEEPTB INFO Epoch 495 summary: train_loss: 0.006556
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep495
DEEPTB INFO iteration:496 train_loss: 0.006540 (0.006576) lr: 0.006094
DEEPTB INFO checkpoint saved as nnsk.iter496
DEEPTB INFO Epoch 496 summary: train_loss: 0.006540
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep496
DEEPTB INFO iteration:497 train_loss: 0.006525 (0.006561) lr: 0.006088
DEEPTB INFO checkpoint saved as nnsk.iter497
DEEPTB INFO Epoch 497 summary: train_loss: 0.006525
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep497
DEEPTB INFO iteration:498 train_loss: 0.006510 (0.006546) lr: 0.006082
DEEPTB INFO checkpoint saved as nnsk.iter498
DEEPTB INFO Epoch 498 summary: train_loss: 0.006510
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep498
DEEPTB INFO iteration:499 train_loss: 0.006495 (0.006530) lr: 0.006076
DEEPTB INFO checkpoint saved as nnsk.iter499
DEEPTB INFO Epoch 499 summary: train_loss: 0.006495
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep499
DEEPTB INFO iteration:500 train_loss: 0.006480 (0.006515) lr: 0.00607
DEEPTB INFO checkpoint saved as nnsk.iter500
DEEPTB INFO Epoch 500 summary: train_loss: 0.006480
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep500
DEEPTB INFO finished training
DEEPTB INFO wall time: 94.206 s
# !dptb run band.json -i ./nnsk1/checkpoint/nnsk.best.pth -o band1 -stu ../data/silicon.vasp
!dptb run band.json -i ./ref_ckpt/nnsk_tr1.pth -o band1 -stu ../data/silicon.vasp
# display the band plot:
from IPython.display import Image, display
import os
image_path = f'./band1/results/band.png'
display(Image(filename=image_path,width=400))
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO The ['overlap_param'] are frozen!
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
DEEPTB INFO KPOINTS klist: 302 kpoints
DEEPTB INFO The eigenvalues are already in data. will use them.
DEEPTB INFO Calculating Fermi energy in the case of spin-degeneracy.
DEEPTB INFO Fermi energy converged after 20 iterations.
DEEPTB INFO q_cal: 8.000000000080137, total_electrons: 8.0, diff q: 8.01367860958635e-11
DEEPTB INFO Estimated E_fermi: -3.864934537298578 based on the valence electrons setting nel_atom : {'Si': 4} .
DEEPTB INFO Using input Fermi energy: -4.7220 eV (estimated: -3.8649 eV)
Figure(640x560)
DEEPTB INFO band calculation successfully completed.
2.3 Training on MD data (bond length dependence)#
In DeePTB, the SK integral based on physical images is parameterized by various bond length-related functions. For example, in the above forms of powerlaw and poly4pow, the bond integral is an explicit function of bond length. This provides good transferability for the NNSK model, allowing it to fully simulate the changes in electronic structure caused by structural distortion.
To further improve the transferability of such models, we strongly recommend training bond length-dependent parameters. This type of training can be easily obtained from MD trajectory datasets. Additionally, it is important to ensure that the training dataset is diverse and representative of the various bond lengths encountered in practical scenarios.
We provide datasets for 10 MD frames at 25K, 100K, and 300K. Users can easily obtain bond length-dependent NNSK models by modifying the data_options/train/prefix in the input configuration to kpathmd25/kpathmd100/kpathmd300, and using the -i option to initialize the checkpoint for training.
During the training on MD trajectory datasets, in addition to providing the train dataset of MD trajectory data, it is also recommended to provide the single structure dataset used in the previous training as a reference dataset. This can help stabilize the training process for the MD trajectory. When using the reference dataset, it is necessary to specify the ref_batch_size in train_option and the corresponding loss calculation method for the reference dataset in train_loss. The rest of the input content remains unchanged. For specific input details, please refer to input_2.json.
After training the model, users can use the same band plotting API as mentioned earlier to visualize the band structure. The plotting parameters are located in ./run/band_2.json.
We now adjust the training input parameters. For details, please refer to input_2.json. We change the prefix of the dataset to kpathmd100.0. We can also set other parameters such as learning rate and number of iterations. For specific parameter settings, please refer to input_2.json.
main changes in data_options：
{
"data_options": {
"train": {
"root": "./data/",
"prefix": "kpathmd100",
"get_eigenvalues": true,
"get_Hamiltonian": false
},
"reference": {
"root": "./data/",
"prefix": "kpath_spk",
"type": "DefaultDataset",
"get_eigenvalues": true,
"get_Hamiltonian": false
}
}
}
and loss_options
{
"loss_options": {
"train": {
"method": "eigvals",
"diff_on": false,
"eout_weight": 0.001,
"diff_weight": 0.01
},
"reference": {
"method": "eigvals",
"diff_on": false,
"eout_weight": 0.001,
"diff_weight": 0.01
}
},
}
Here, just as a demonstration, we also reduced the number of training iterations.
# v100, 4m4s
# !dptb train input_2.json -i nnsk1/checkpoint/nnsk.best.pth -o nnskmd100
!dptb train input_2.json -i ./ref_ckpt/nnsk_tr1.pth -o nnskmd100
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO ------------------------------------------------------------------
DEEPTB INFO Cutoff options:
DEEPTB INFO
DEEPTB INFO r_max : {'Si-Si': 6.24}
DEEPTB INFO er_max : None
DEEPTB INFO oer_max : None
DEEPTB INFO ------------------------------------------------------------------
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
DEEPTB WARNING The cutoffs in data and model are not checked. be careful!
DEEPTB WARNING The cutoffs in data and model are not checked. be careful!
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO iteration:1 train_loss: 0.127402 (0.038221) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter1
DEEPTB INFO iteration:2 train_loss: 0.469684 (0.167660) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter2
DEEPTB INFO iteration:3 train_loss: 0.126340 (0.155264) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter3
DEEPTB INFO iteration:4 train_loss: 0.213768 (0.172815) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter4
DEEPTB INFO iteration:5 train_loss: 0.307562 (0.213239) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter5
DEEPTB INFO iteration:6 train_loss: 0.205336 (0.210868) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter6
DEEPTB INFO iteration:7 train_loss: 0.102732 (0.178427) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter7
DEEPTB INFO iteration:8 train_loss: 0.112134 (0.158539) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter8
DEEPTB INFO iteration:9 train_loss: 0.176445 (0.163911) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter9
DEEPTB INFO iteration:10 train_loss: 0.185834 (0.170488) lr: 0.01
DEEPTB INFO checkpoint saved as nnsk.iter10
DEEPTB INFO Epoch 1 summary: train_loss: 0.202724
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as nnsk.ep1
DEEPTB INFO iteration:11 train_loss: 0.134079 (0.159565) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter11
DEEPTB INFO iteration:12 train_loss: 0.085558 (0.137363) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter12
DEEPTB INFO iteration:13 train_loss: 0.087609 (0.122437) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter13
DEEPTB INFO iteration:14 train_loss: 0.119255 (0.121482) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter14
DEEPTB INFO iteration:15 train_loss: 0.131781 (0.124572) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter15
DEEPTB INFO iteration:16 train_loss: 0.107232 (0.119370) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter16
DEEPTB INFO iteration:17 train_loss: 0.075381 (0.106173) lr: 0.00999
DEEPTB INFO checkpoint saved as nnsk.iter17
^C
Traceback (most recent call last):
File "/root/dptb_venv/bin/dptb", line 8, in <module>
sys.exit(main())
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/__main__.py", line 35, in main
entry_main()
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/entrypoints/main.py", line 473, in main
train(**dict_args)
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/entrypoints/train.py", line 234, in train
trainer.run(trainer.train_options["num_epoch"])
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/base_trainer.py", line 52, in run
self.epoch()
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/trainer.py", line 198, in epoch
self.iteration(ibatch, next(iter(self.reference_loader)))
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/trainer.py", line 130, in iteration
loss.backward()
File "/root/dptb_venv/lib/python3.10/site-packages/torch/_tensor.py", line 581, in backward
torch.autograd.backward(
File "/root/dptb_venv/lib/python3.10/site-packages/torch/autograd/__init__.py", line 347, in backward
_engine_run_backward(
File "/root/dptb_venv/lib/python3.10/site-packages/torch/autograd/graph.py", line 825, in _engine_run_backward
return Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
KeyboardInterrupt
# !dptb run band.json -i ./nnskmd100/checkpoint/nnsk.best.pth -o band2 -stu ../data/silicon.vasp
!dptb run band.json -i ./ref_ckpt/nnsk.md100.pth -o band2 -stu ../data/silicon.vasp
# display the band plot:
from IPython.display import Image, display
import os
image_path = f'./band2/results/band.png'
display(Image(filename=image_path,width=400))
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO The ['overlap_param'] are frozen!
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
DEEPTB INFO KPOINTS klist: 302 kpoints
DEEPTB INFO The eigenvalues are already in data. will use them.
DEEPTB INFO Calculating Fermi energy in the case of spin-degeneracy.
DEEPTB INFO Fermi energy converged after 26 iterations.
DEEPTB INFO q_cal: 8.000000000186008, total_electrons: 8.0, diff q: 1.8600765372411843e-10
DEEPTB INFO Estimated E_fermi: -4.091689614180318 based on the valence electrons setting nel_atom : {'Si': 4} .
DEEPTB INFO Using input Fermi energy: -4.7220 eV (estimated: -4.0917 eV)
Figure(640x560)
DEEPTB INFO band calculation successfully completed.
# !dptb run band_2.json -i ./nnskmd100/checkpoint/nnsk.best.pth -stu ./data/kpathmd100.0/struct.vasp -o band3
!dptb run band_2.json -i ./ref_ckpt/nnsk.md100.pth -stu ../data/kpathmd100.0/struct.vasp -o band3
# display the band plot:
from IPython.display import Image, display
import os
image_path = f'./band3/results/band.png'
display(Image(filename=image_path,width=400))
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO The ['overlap_param'] are frozen!
DEEPTB INFO The ['overlap_param'] are frozen!
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
DEEPTB INFO KPOINTS klist: 354 kpoints
DEEPTB INFO The eigenvalues are already in data. will use them.
DEEPTB INFO Calculating Fermi energy in the case of spin-degeneracy.
DEEPTB WARNING Fermi level bisection did not converge under tolerance 1e-10 after 55 iterations.
DEEPTB INFO q_cal: 32.00000000075238, total_electrons: 32.0, diff q: 7.523794920416549e-10
DEEPTB INFO Estimated E_fermi: -4.1064770221710205 based on the valence electrons setting nel_atom : {'Si': 4} .
DEEPTB INFO Using Fermi energy: -4.1065 eV (matches estimated value)
Figure(640x560)
DEEPTB INFO band calculation successfully completed.
2.4 Training Environment Correction#
The DeePTB-SK module provides powerful environment-dependent modeling with symmetry-preserving neural networks. Based on the previously constructed nnsk model, we can further enhance the TB model’s descriptive ability by adding an environment-dependent component to overcome the accuracy limitations imposed by the two-center approximation. The model that incorporates environment dependence into the nnsk model is referred to as the mix model, and its expression is as follows: $$
(1)#\[\begin{equation} h^{\text{env}}_{ll^\prime{\zeta}} = h_{ll^\prime{\zeta}}(r_{ij}) \times \left[1+\Phi_{ll^\prime\zeta}^{o_i,o_j}\left(r_{ij},\mathcal{D}^{ij}\right)\right] \end{equation}\]
$\( where \)\mathcal{D}^{ij}\( is the environment descriptor defined by the `embedding` keyword, and \)\Phi_{ll^\prime\zeta}^{o_i,o_j}$ is the neural network that provides the environment correction prediction value.
To define the mix correction model, you need to provide the following keywords in the model_options section of the training input file:
embedding: Themethodhere specifies the form of the atomic environment used in thedptbmodel. In this example, we use these2form of descriptor similar to that used in DeePMD.prediction: Themethodspecifies the prediction method of the model, which is set tosktbhere. Theneuronskeyword specifies the size of the prediction network.nnsk: This section is consistent with the content in thennskmodel. Thefreezeoption should be set totrue, indicating that the trained SK parameters of thennskmodel are fixed, and only the neural network parameters of the environment-dependent part are trained. This fixing is crucial; otherwise, the initialization of themixmodel may completely destroy the parameters of thennskmodel, leading to non-convergence during training.
For example:
"model_options": {
"embedding":{
"method": "se2",
"rs": 2.5,
"rc": 5.0,
"radial_net": {
"neurons": [10,20,30]
}
},
"prediction":{
"method": "sktb",
"neurons": [16,16,16]
},
"nnsk": {
...
"freeze": true
...
}
}
The complete input content can be found in input_3.json. The environment-dependent mix model training requires reading the nnsk model. We can run:
# !dptb train input_3.json -i ./nnskmd100/checkpoint/nnsk.ep20.pth -o ./mix
!dptb train input_3.json -i ./ref_ckpt/nnsk.md100.pth -o ./mix
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB WARNING The model options embedding is not defined in checkpoint, set to {'method': 'se2', 'rs': 2.5, 'rc': 5.0, 'radial_net': {'neurons': [10, 20, 30], 'activation': 'tanh', 'if_batch_normalized': False}, 'n_axis': None}.
DEEPTB WARNING The model options prediction is not defined in checkpoint, set to {'method': 'sktb', 'neurons': [16, 16, 16], 'activation': 'tanh', 'if_batch_normalized': False}.
DEEPTB WARNING The model option freeze is set to True, but in checkpoint it is ['overlap'], make sure it it correct!
DEEPTB INFO ------------------------------------------------------------------
DEEPTB INFO Cutoff options:
DEEPTB INFO
DEEPTB INFO r_max : {'Si-Si': 6.24}
DEEPTB INFO er_max : 5.0
DEEPTB INFO oer_max : None
DEEPTB INFO ------------------------------------------------------------------
Processing dataset...
Loading data: 0%| | 0/1 [00:00<?, ?it/s]/root/dptb_venv/lib/python3.10/site-packages/dptb/data/dataset/_default_dataset.py:254: UserWarning: The given NumPy array is not writable, and PyTorch does not support non-writable tensors. This means writing to this tensor will result in undefined behavior. You may want to copy the array to protect its data or make it writable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at ../torch/csrc/utils/tensor_numpy.cpp:206.)
kwargs[AtomicDataDict.KPOINT_KEY] = torch.as_tensor(self.data[AtomicDataDict.KPOINT_KEY][frame], dtype=torch.get_default_dtype())
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
Loading data: 100%|███████████████████████████████| 1/1 [00:00<00:00, 2.31it/s]
DEEPTB INFO Loaded data: Batch(atomic_numbers=[80, 1], batch=[80], bwindow=[10, 2], cell=[10, 3, 3], edge_cell_shift=[3680, 3], edge_features=[3680, 1], edge_index=[2, 3680], edge_overlap=[3680, 1], eigenvalue=nested, env_cell_shift=[2240, 3], env_index=[2, 2240], kpoint=nested, node_features=[80, 1], node_overlap=[80, 1], node_soc=[80, 1], node_soc_switch=[10, 1], pbc=[10, 3], pos=[80, 3], ptr=[11])
processed data size: ~1.04 MB
DEEPTB INFO Cached processed data to disk
Done!
DEEPTB WARNING The cutoffs in data and model are not checked. be careful!
Processing dataset...
Loading data: 100%|███████████████████████████████| 1/1 [00:00<00:00, 38.25it/s]
DEEPTB INFO Loaded data: Batch(atomic_numbers=[2, 1], batch=[2], bwindow=[1, 2], cell=[1, 3, 3], edge_cell_shift=[92, 3], edge_features=[92, 1], edge_index=[2, 92], edge_overlap=[92, 1], eigenvalue=nested, env_cell_shift=[56, 3], env_index=[2, 56], kpoint=nested, node_features=[2, 1], node_overlap=[2, 1], node_soc=[2, 1], node_soc_switch=[1, 1], pbc=[1, 3], pos=[2, 3], ptr=[2])
processed data size: ~0.01 MB
DEEPTB INFO Cached processed data to disk
Done!
DEEPTB WARNING The cutoffs in data and model are not checked. be careful!
DEEPTB WARNING The push option is not supported in the mixed model. The push option is only supported in the nnsk model.
DEEPTB INFO The ['hopping_param', 'overlap_param', 'onsite_param'] are frozen!
DEEPTB INFO iteration:1 train_loss: 0.012923 (0.003877) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter1
DEEPTB INFO iteration:2 train_loss: 0.016042 (0.007526) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter2
DEEPTB INFO iteration:3 train_loss: 0.012602 (0.009049) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter3
DEEPTB INFO iteration:4 train_loss: 0.012335 (0.010035) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter4
DEEPTB INFO iteration:5 train_loss: 0.013277 (0.011008) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter5
DEEPTB INFO iteration:6 train_loss: 0.012853 (0.011561) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter6
DEEPTB INFO iteration:7 train_loss: 0.012130 (0.011732) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter7
DEEPTB INFO iteration:8 train_loss: 0.011827 (0.011760) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter8
DEEPTB INFO iteration:9 train_loss: 0.012503 (0.011983) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter9
DEEPTB INFO iteration:10 train_loss: 0.012240 (0.012060) lr: 0.001
DEEPTB INFO checkpoint saved as mix.iter10
DEEPTB INFO Epoch 1 summary: train_loss: 0.012873
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as mix.ep1
DEEPTB INFO iteration:11 train_loss: 0.011943 (0.012025) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter11
DEEPTB INFO iteration:12 train_loss: 0.011564 (0.011887) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter12
DEEPTB INFO iteration:13 train_loss: 0.011930 (0.011900) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter13
DEEPTB INFO iteration:14 train_loss: 0.012044 (0.011943) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter14
DEEPTB INFO iteration:15 train_loss: 0.011880 (0.011924) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter15
DEEPTB INFO iteration:16 train_loss: 0.011474 (0.011789) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter16
DEEPTB INFO iteration:17 train_loss: 0.011509 (0.011705) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter17
DEEPTB INFO iteration:18 train_loss: 0.011747 (0.011718) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter18
DEEPTB INFO iteration:19 train_loss: 0.011595 (0.011681) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter19
DEEPTB INFO iteration:20 train_loss: 0.011615 (0.011661) lr: 0.000999
DEEPTB INFO checkpoint saved as mix.iter20
DEEPTB INFO Epoch 2 summary: train_loss: 0.011730
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as mix.ep2
DEEPTB INFO iteration:21 train_loss: 0.011626 (0.011651) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter21
DEEPTB INFO iteration:22 train_loss: 0.011402 (0.011576) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter22
DEEPTB INFO iteration:23 train_loss: 0.011556 (0.011570) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter23
DEEPTB INFO iteration:24 train_loss: 0.011458 (0.011536) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter24
DEEPTB INFO iteration:25 train_loss: 0.011385 (0.011491) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter25
DEEPTB INFO iteration:26 train_loss: 0.011412 (0.011467) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter26
DEEPTB INFO iteration:27 train_loss: 0.011503 (0.011478) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter27
DEEPTB INFO iteration:28 train_loss: 0.011466 (0.011474) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter28
DEEPTB INFO iteration:29 train_loss: 0.011669 (0.011533) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter29
DEEPTB INFO iteration:30 train_loss: 0.011285 (0.011459) lr: 0.000998
DEEPTB INFO checkpoint saved as mix.iter30
DEEPTB INFO Epoch 3 summary: train_loss: 0.011476
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as mix.ep3
DEEPTB INFO iteration:31 train_loss: 0.011497 (0.011470) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter31
DEEPTB INFO iteration:32 train_loss: 0.011230 (0.011398) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter32
DEEPTB INFO iteration:33 train_loss: 0.011447 (0.011413) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter33
DEEPTB INFO iteration:34 train_loss: 0.011389 (0.011405) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter34
DEEPTB INFO iteration:35 train_loss: 0.011500 (0.011434) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter35
DEEPTB INFO iteration:36 train_loss: 0.011169 (0.011354) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter36
DEEPTB INFO iteration:37 train_loss: 0.011279 (0.011332) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter37
DEEPTB INFO iteration:38 train_loss: 0.011217 (0.011297) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter38
DEEPTB INFO iteration:39 train_loss: 0.011352 (0.011314) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter39
DEEPTB INFO iteration:40 train_loss: 0.011480 (0.011364) lr: 0.000997
DEEPTB INFO checkpoint saved as mix.iter40
DEEPTB INFO Epoch 4 summary: train_loss: 0.011356
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as mix.ep4
DEEPTB INFO iteration:41 train_loss: 0.011409 (0.011377) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter41
DEEPTB INFO iteration:42 train_loss: 0.011185 (0.011320) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter42
DEEPTB INFO iteration:43 train_loss: 0.011341 (0.011326) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter43
DEEPTB INFO iteration:44 train_loss: 0.011174 (0.011280) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter44
DEEPTB INFO iteration:45 train_loss: 0.011342 (0.011299) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter45
DEEPTB INFO iteration:46 train_loss: 0.011222 (0.011276) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter46
DEEPTB INFO iteration:47 train_loss: 0.011125 (0.011231) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter47
DEEPTB INFO iteration:48 train_loss: 0.011265 (0.011241) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter48
DEEPTB INFO iteration:49 train_loss: 0.011438 (0.011300) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter49
DEEPTB INFO iteration:50 train_loss: 0.011371 (0.011321) lr: 0.000996
DEEPTB INFO checkpoint saved as mix.iter50
DEEPTB INFO Epoch 5 summary: train_loss: 0.011287
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as mix.ep5
DEEPTB INFO iteration:51 train_loss: 0.011115 (0.011259) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter51
DEEPTB INFO iteration:52 train_loss: 0.011362 (0.011290) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter52
DEEPTB INFO iteration:53 train_loss: 0.011376 (0.011316) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter53
DEEPTB INFO iteration:54 train_loss: 0.011208 (0.011284) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter54
DEEPTB INFO iteration:55 train_loss: 0.011114 (0.011233) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter55
DEEPTB INFO iteration:56 train_loss: 0.011220 (0.011229) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter56
DEEPTB INFO iteration:57 train_loss: 0.011264 (0.011239) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter57
DEEPTB INFO iteration:58 train_loss: 0.011359 (0.011275) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter58
DEEPTB INFO iteration:59 train_loss: 0.011103 (0.011224) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter59
DEEPTB INFO iteration:60 train_loss: 0.011354 (0.011263) lr: 0.000995
DEEPTB INFO checkpoint saved as mix.iter60
DEEPTB INFO Epoch 6 summary: train_loss: 0.011248
---------------------------------------------------------------------------------
DEEPTB INFO checkpoint saved as mix.ep6
DEEPTB INFO iteration:61 train_loss: 0.011180 (0.011238) lr: 0.000994
DEEPTB INFO checkpoint saved as mix.iter61
DEEPTB INFO iteration:62 train_loss: 0.011182 (0.011221) lr: 0.000994
DEEPTB INFO checkpoint saved as mix.iter62
DEEPTB INFO iteration:63 train_loss: 0.011087 (0.011181) lr: 0.000994
DEEPTB INFO checkpoint saved as mix.iter63
DEEPTB INFO iteration:64 train_loss: 0.011344 (0.011230) lr: 0.000994
DEEPTB INFO checkpoint saved as mix.iter64
DEEPTB INFO iteration:65 train_loss: 0.011371 (0.011272) lr: 0.000994
DEEPTB INFO checkpoint saved as mix.iter65
Traceback (most recent call last):
File "/root/dptb_venv/bin/dptb", line 8, in <module>
sys.exit(main())
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/__main__.py", line 35, in main
entry_main()
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/entrypoints/main.py", line 473, in main
train(**dict_args)
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/entrypoints/train.py", line 234, in train
trainer.run(trainer.train_options["num_epoch"])
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/base_trainer.py", line 52, in run
self.epoch()
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/trainer.py", line 198, in epoch
self.iteration(ibatch, next(iter(self.reference_loader)))
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/trainer.py", line 108, in iteration
loss = self.train_lossfunc(batch, batch_for_loss)
File "/root/dptb_venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/root/dptb_venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nnops/loss.py", line 154, in forward
data = self.eigenvalue(AtomicData.to_AtomicDataDict(data))
File "/root/dptb_venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/root/dptb_venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nn/energy.py", line 74, in forward
data = self.s2k(data)
File "/root/dptb_venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/root/dptb_venv/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
File "/root/dptb_venv/lib/python3.10/site-packages/dptb/nn/hr2hk.py", line 155, in forward
block[:,iatom_indices,jatom_indices] += masked_hblock.squeeze(0).type_as(block) * \
KeyboardInterrupt
^C
# !dptb run band_2.json -i ./mix/checkpoint/mix.best.pth -stu ./data/kpathmd100.0/struct.vasp -o band4
!dptb run band_2.json -i ./ref_ckpt/mix.md100.pth -stu ../data/kpathmd100.0/struct.vasp -o band4
# display the band plot:
from IPython.display import Image, display
import os
image_path = f'./band4/results/band.png'
display(Image(filename=image_path,width=400))
TBPLaS is not installed. Thus the TBPLaS is not available, Please install it first.
#################################################################################
# #
# #
# ██████████ ███████████ ███████████ ███████████ #
# ░░███░░░░███ ░░███░░░░░███░█░░░███░░░█░░███░░░░░███ #
# ░███ ░░███ ██████ ██████ ░███ ░███░ ░███ ░ ░███ ░███ #
# ░███ ░███ ███░░███ ███░░███ ░██████████ ░███ ░██████████ #
# ░███ ░███░███████ ░███████ ░███░░░░░░ ░███ ░███░░░░░███ #
# ░███ ███ ░███░░░ ░███░░░ ░███ ░███ ░███ ░███ #
# ██████████ ░░██████ ░░██████ █████ █████ ███████████ #
# ░░░░░░░░░░ ░░░░░░ ░░░░░░ ░░░░░ ░░░░░ ░░░░░░░░░░░ #
# #
# Version: 2.0.4.dev87+5ed8d35 #
# #
#################################################################################
DEEPTB INFO The ['hopping_param', 'overlap_param', 'onsite_param'] are frozen!
/root/dptb_venv/lib/python3.10/site-packages/torch/nested/__init__.py:107: UserWarning: The PyTorch API of nested tensors is in prototype stage and will change in the near future. (Triggered internally at ../aten/src/ATen/NestedTensorImpl.cpp:178.)
return torch._nested_tensor_from_tensor_list(ts, dtype, None, device, None)
DEEPTB INFO KPOINTS klist: 354 kpoints
DEEPTB INFO The eigenvalues are already in data. will use them.
DEEPTB INFO Calculating Fermi energy in the case of spin-degeneracy.
DEEPTB WARNING Fermi level bisection did not converge under tolerance 1e-10 after 55 iterations.
DEEPTB INFO q_cal: 32.0000000006585, total_electrons: 32.0, diff q: 6.585025857930304e-10
DEEPTB INFO Estimated E_fermi: -4.093235731124878 based on the valence electrons setting nel_atom : {'Si': 4} .
DEEPTB INFO Using input Fermi energy: -4.1065 eV (estimated: -4.0932 eV)
Figure(640x560)
DEEPTB INFO band calculation successfully completed.
Author: Gu, Qiangqiang : guqq@ustc.edu.cn
Thank you for reading!