PyTorch model - optimizer - state_dict() - torch.save(config, save_path) - torch.load(load_path)
1. state_dict()
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# yongqiang cheng
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
# Define model
class TheModelClass(nn.Module):
def __init__(self):
super(TheModelClass, self).__init__()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16 * 5 * 5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
# Initialize model
model = TheModelClass()
# Initialize optimizer
optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)
# Print model's state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
print(param_tensor, "\t", model.state_dict()[param_tensor].size())
# Print optimizer's state_dict
print("Optimizer's state_dict:")
for var_name in optimizer.state_dict():
print(var_name, "\t", optimizer.state_dict()[var_name])
/home/yongqiang/miniconda3/envs/pt-1.4_py-3.6/bin/python /home/yongqiang/pytorch_work/end2end-asr-pytorch-example/yongqiang.py
Model's state_dict:
conv1.weight torch.Size([6, 3, 5, 5])
conv1.bias torch.Size([6])
conv2.weight torch.Size([16, 6, 5, 5])
conv2.bias torch.Size([16])
fc1.weight torch.Size([120, 400])
fc1.bias torch.Size([120])
fc2.weight torch.Size([84, 120])
fc2.bias torch.Size([84])
fc3.weight torch.Size([10, 84])
fc3.bias torch.Size([10])
Optimizer's state_dict:
state {}
param_groups [{'lr': 0.001, 'momentum': 0.9, 'dampening': 0, 'weight_decay': 0, 'nesterov': False, 'params': [140726061376352, 140726040176608, 140726040176680, 140726040176752, 140726040176824, 140726040176896, 140726040176968, 140726040177040, 140726040177112, 140726040177184]}]
Process finished with exit code 0
2. state_dict()
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# yongqiang cheng
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
# Define model
class TheModelClass(nn.Module):
def __init__(self):
super(TheModelClass, self).__init__()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16 * 5 * 5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
# Initialize model
model = TheModelClass()
# Initialize optimizer
optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)
# Print model's state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
print(param_tensor, "\t", model.state_dict()[param_tensor].size())
# Print optimizer's state_dict
print("Optimizer's state_dict:")
for var_name in optimizer.state_dict():
print(var_name, "\t", optimizer.state_dict()[var_name])
# Print model's state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
print(param_tensor, "\n", model.state_dict()[param_tensor])
/home/yongqiang/miniconda3/envs/pt-1.4_py-3.6/bin/python /home/yongqiang/pytorch_work/end2end-asr-pytorch-example/yongqiang.py
Model's state_dict:
conv1.weight torch.Size([6, 3, 5, 5])
conv1.bias torch.Size([6])
conv2.weight torch.Size([16, 6, 5, 5])
conv2.bias torch.Size([16])
fc1.weight torch.Size([120, 400])
fc1.bias torch.Size([120])
fc2.weight torch.Size([84, 120])
fc2.bias torch.Size([84])
fc3.weight torch.Size([10, 84])
fc3.bias torch.Size([10])
Optimizer's state_dict:
state {}
param_groups [{'lr': 0.001, 'momentum': 0.9, 'dampening': 0, 'weight_decay': 0, 'nesterov': False, 'params': [140109876585384, 140109855434792, 140109855434864, 140109855434936, 140109855435008, 140109855435080, 140109855435152, 140109855435224, 140109855435296, 140109855435368]}]
Model's state_dict:
conv1.weight
......
conv1.bias
tensor([-0.0674, -0.1050, -0.0501, -0.0006, -0.0412, 0.0723])
conv2.weight
......
conv2.bias
tensor([-0.0759, -0.0430, -0.0360, 0.0326, -0.0434, -0.0319, -0.0762, -0.0777,
-0.0670, 0.0813, -0.0459, -0.0032, 0.0653, -0.0217, -0.0804, -0.0423])
fc1.weight
tensor([[-0.0215, 0.0353, 0.0108, ..., 0.0245, -0.0217, 0.0306],
[-0.0069, -0.0206, -0.0316, ..., -0.0325, 0.0455, -0.0178],
[ 0.0082, 0.0180, 0.0067, ..., -0.0385, 0.0237, 0.0232],
...,
[ 0.0438, -0.0409, -0.0337, ..., 0.0358, -0.0055, 0.0378],
[ 0.0077, -0.0468, 0.0162, ..., -0.0433, -0.0359, -0.0240],
[-0.0498, 0.0463, -0.0128, ..., -0.0427, 0.0169, 0.0093]])
fc1.bias
......
fc2.weight
tensor([[ 0.0489, 0.0608, 0.0596, ..., -0.0331, -0.0158, 0.0263],
[-0.0729, -0.0118, -0.0794, ..., 0.0427, -0.0092, -0.0524],
[-0.0814, 0.0552, 0.0365, ..., 0.0676, 0.0044, 0.0455],
...,
[ 0.0636, 0.0371, -0.0887, ..., -0.0207, -0.0367, -0.0761],
[-0.0584, 0.0579, -0.0076, ..., 0.0863, -0.0167, -0.0223],
[ 0.0247, -0.0500, -0.0751, ..., -0.0557, -0.0673, 0.0164]])
fc2.bias
......
fc3.weight
......
fc3.bias
tensor([ 0.0501, 0.0460, -0.1056, -0.0683, 0.0583, -0.0780, 0.0997, 0.0550,
0.0777, -0.0156])
Process finished with exit code 0
3. torch.save(config, save_path)
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# yongqiang cheng
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
def save_model(model, epoch, optimizer, save_folder, name, loss, best_model=False):
"""
Saving model
"""
if best_model:
save_path = "{}/{}/best_model.th".format(save_folder, name)
else:
save_path = "{}/{}/epoch_{}.th".format(save_folder, name, epoch)
if not os.path.exists(save_folder + "/" + name):
os.makedirs(save_folder + "/" + name)
print("SAVE MODEL to", save_path)
if loss == "ce":
config = {
'label2id': "label2id",
'id2label': "id2label",
'args': "args",
'epoch': "epoch",
'model_state_dict': model.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
'optimizer_params': {
'_step': "_step",
'_rate': "_rate",
'warmup': "warmup",
'factor': "factor",
'model_size': "model_size"
},
'metrics': "metrics"
}
elif loss == "ctc":
config = {
'label2id': "label2id",
'id2label': "id2label",
'args': "args",
'epoch': epoch,
'model_state_dict': model.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
'optimizer_params': {
'lr': "lr",
'lr_anneal': "lr_anneal"
},
'metrics': "metrics"
}
else:
print("Loss is not defined")
torch.save(config, save_path)
# Define model
class TheModelClass(nn.Module):
def __init__(self):
super(TheModelClass, self).__init__()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16 * 5 * 5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
# Initialize model
model = TheModelClass()
# Initialize optimizer
optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)
# Print model's state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
print(param_tensor, "\t", model.state_dict()[param_tensor].size())
# Print optimizer's state_dict
print("Optimizer's state_dict:")
for var_name in optimizer.state_dict():
print(var_name, "\t", optimizer.state_dict()[var_name])
# Print model's state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
print(param_tensor, "\n", model.state_dict()[param_tensor])
model_folder = "/home/yongqiang/pytorch_work/end2end-asr-pytorch-example"
save_model(model=model, epoch=9, optimizer=optimizer, save_folder=model_folder, name="log", loss="ctc", best_model=False)
/home/yongqiang/miniconda3/envs/pt-1.4_py-3.6/bin/python /home/yongqiang/pytorch_work/end2end-asr-pytorch-example/yongqiang.py
Model's state_dict:
conv1.weight torch.Size([6, 3, 5, 5])
conv1.bias torch.Size([6])
conv2.weight torch.Size([16, 6, 5, 5])
conv2.bias torch.Size([16])
fc1.weight torch.Size([120, 400])
fc1.bias torch.Size([120])
fc2.weight torch.Size([84, 120])
fc2.bias torch.Size([84])
fc3.weight torch.Size([10, 84])
fc3.bias torch.Size([10])
Optimizer's state_dict:
state {}
param_groups [{'lr': 0.001, 'momentum': 0.9, 'dampening': 0, 'weight_decay': 0, 'nesterov': False, 'params': [139858574714200, 139858553707328, 139858553707400, 139858553707472, 139858553707544, 139858553707616, 139858553707688, 139858553707760, 139858553707832, 139858553707904]}]
Model's state_dict:
conv1.weight
tensor([[[[ 0.0134, -0.0569, -0.0762, -0.0897, -0.0430],
[ 0.0423, -0.0034, -0.0318, 0.0950, -0.0198],
[ 0.0963, 0.0397, -0.0560, 0.0744, 0.0987],
[ 0.0997, 0.0042, -0.0353, -0.0677, 0.0536],
[ 0.0791, 0.0338, -0.0949, -0.0367, 0.0982]],
[[ 0.0951, 0.0567, 0.1046, -0.0704, -0.0307],
[ 0.0991, 0.0478, 0.0778, 0.0336, 0.0927],
[ 0.0694, 0.0703, 0.1084, 0.1068, 0.0833],
[-0.0444, 0.0951, 0.0369, -0.1047, 0.0575],
[ 0.0866, 0.1128, 0.0547, -0.0558, 0.0164]],
[[ 0.0562, -0.1008, -0.0253, 0.0131, 0.0880],
[-0.0422, -0.0437, 0.0181, -0.0247, -0.0727],
[-0.0072, 0.0113, 0.0032, 0.0571, 0.0431],
[-0.0011, -0.0778, 0.0533, 0.0441, 0.0252],
[-0.0267, 0.0294, -0.0218, -0.1044, -0.0390]]],
[[[ 0.0848, 0.1063, -0.0568, -0.0719, 0.0082],
[-0.0797, 0.0945, -0.0328, -0.0290, -0.0759],
[ 0.0794, -0.0985, -0.0748, -0.0041, 0.0019],
[-0.0375, -0.0602, -0.0819, -0.1079, 0.0773],
[ 0.0944, -0.0515, -0.0621, 0.0124, 0.0627]],
[[ 0.0582, -0.1018, 0.0449, -0.0540, 0.0171],
[-0.0551, 0.1112, -0.0233, -0.1066, 0.0431],
[ 0.0936, -0.1010, -0.0952, -0.0989, -0.0224],
[-0.0374, -0.0663, -0.1145, -0.0551, -0.0156],
[-0.0178, -0.0025, -0.0656, -0.0878, 0.0368]],
[[ 0.0199, -0.0584, 0.0842, -0.1144, -0.0983],
[ 0.0355, 0.0959, -0.0010, -0.0964, 0.0487],
[-0.0616, -0.0055, 0.0847, 0.0304, -0.0560],
[ 0.0960, -0.1062, 0.1005, -0.0658, -0.0791],
[ 0.0265, -0.0321, -0.0826, -0.1112, -0.0275]]],
[[[-0.1136, -0.0062, -0.0536, -0.0138, -0.1113],
[ 0.0939, 0.0356, -0.0073, 0.0578, 0.0617],
[ 0.0470, 0.0967, -0.0801, 0.1070, 0.0272],
[ 0.0662, 0.0654, 0.0242, 0.1028, -0.0372],
[ 0.0699, 0.0887, 0.1111, 0.0259, -0.0816]],
[[ 0.0348, 0.0764, 0.0776, -0.0557, 0.0993],
[ 0.0266, -0.0372, -0.0937, 0.0853, -0.0370],
[ 0.0783, -0.0296, 0.0722, -0.1091, -0.1092],
[ 0.1108, 0.1095, -0.0578, 0.0673, 0.0882],
[ 0.0088, -0.0430, 0.0211, 0.1035, -0.0614]],
[[ 0.0309, 0.0961, -0.0820, 0.0557, 0.1115],
[ 0.0167, 0.0800, 0.0058, 0.1004, -0.0041],
[-0.0032, -0.0768, 0.0156, 0.0554, 0.0967],
[-0.0623, -0.0133, 0.0419, 0.0855, -0.0093],
[-0.0782, 0.0276, -0.0715, -0.0109, 0.0090]]],
[[[-0.0351, 0.0579, -0.0033, 0.0753, 0.0762],
[ 0.0846, 0.0981, -0.0195, -0.0937, 0.0385],
[-0.0608, -0.0624, 0.0757, 0.0834, -0.0371],
[ 0.0948, -0.0822, -0.1009, 0.1132, -0.1114],
[ 0.0245, 0.0276, 0.0120, -0.0866, 0.0686]],
[[ 0.0277, 0.0420, 0.0820, 0.0180, -0.0904],
[-0.0340, 0.0228, 0.1026, 0.0177, 0.0052],
[-0.0608, -0.0869, -0.1016, 0.0253, 0.1009],
[-0.0366, -0.1122, 0.0936, -0.0223, -0.0959],
[-0.0824, 0.1027, 0.0007, -0.0691, -0.0679]],
[[-0.0165, -0.0798, 0.0052, -0.0003, -0.1124],
[-0.0151, 0.0668, 0.1082, -0.0442, -0.1077],
[-0.0055, -0.0172, 0.0627, 0.0263, -0.0847],
[ 0.0741, 0.1126, 0.0694, -0.0641, -0.0290],
[-0.0695, -0.0944, 0.0760, 0.1055, 0.1067]]],
[[[-0.0462, -0.0732, -0.0339, -0.0569, 0.0344],
[ 0.0217, -0.0960, -0.1135, -0.0820, 0.0353],
[ 0.0226, 0.0066, 0.0191, 0.0522, 0.0986],
[-0.0946, 0.0257, -0.0353, 0.0866, -0.0039],
[ 0.0354, -0.0436, -0.0857, 0.0711, 0.0421]],
[[-0.0715, 0.0448, -0.0478, 0.0755, -0.0578],
[ 0.1127, -0.0526, -0.1046, 0.0710, -0.0189],
[-0.0079, 0.1007, -0.0464, -0.0967, 0.0137],
[ 0.0587, -0.0915, 0.0212, 0.0212, 0.0259],
[ 0.0572, 0.0283, 0.0013, 0.1119, -0.0051]],
[[ 0.0268, 0.0909, -0.0809, -0.0398, 0.1101],
[ 0.1020, 0.0993, 0.0176, -0.0125, -0.0770],
[ 0.0786, -0.1061, 0.0761, -0.0004, 0.0296],
[-0.0290, -0.0337, 0.0569, -0.0355, -0.0754],
[ 0.0760, -0.0028, 0.0969, -0.0407, 0.0886]]],
[[[ 0.0132, -0.0045, -0.0614, 0.0071, -0.1035],
[ 0.0208, 0.0518, -0.1080, -0.0129, -0.0524],
[-0.0328, 0.0093, -0.0668, 0.0273, 0.0307],
[ 0.0311, 0.0798, 0.0596, -0.0256, -0.0979],
[-0.0542, -0.0553, -0.0613, -0.0782, -0.0579]],
[[-0.0875, -0.0210, 0.0966, -0.0768, -0.0592],
[ 0.0897, 0.0828, 0.0343, 0.0133, -0.0669],
[-0.0986, -0.0866, -0.1075, 0.0613, 0.0486],
[-0.1153, 0.1053, 0.0163, -0.0813, -0.0268],
[-0.0267, 0.1103, -0.0289, -0.0212, -0.0427]],
[[-0.0471, -0.0681, 0.0008, 0.0685, -0.0726],
[-0.0161, -0.0378, 0.0142, -0.0448, -0.0542],
[-0.0594, -0.1057, -0.0388, 0.0282, -0.0045],
[-0.0942, 0.1069, -0.0601, -0.1092, 0.0658],
[ 0.0058, -0.0673, -0.0139, 0.0544, 0.0266]]]])
conv1.bias
tensor([-0.0518, 0.0757, -0.0106, 0.0676, -0.0081, 0.0287])
conv2.weight
tensor([[[[ 3.7345e-02, 3.6917e-02, 7.1036e-04, -1.8674e-02, -2.0425e-02],
[ 6.0888e-02, 3.9115e-02, 6.0703e-02, -4.9426e-02, 2.0759e-03],
[ 4.4816e-02, 2.4605e-02, 7.5611e-02, 7.0319e-02, -4.8377e-03],
[-5.8067e-02, 2.4481e-02, -7.9937e-02, -6.9166e-02, 4.6737e-02],
[-1.4972e-02, -7.3878e-02, -7.6790e-02, -1.9032e-02, -1.9925e-02]],
[[ 2.2720e-02, -2.5971e-02, -6.3477e-02, 1.7530e-02, 2.5905e-02],
[-6.4675e-02, -6.6284e-02, -8.0954e-02, 3.6396e-02, -6.3034e-02],
[-1.8645e-03, 4.3630e-02, -2.6132e-02, -7.6036e-02, 7.8512e-02],
[ 4.3217e-02, -3.1471e-02, -9.8552e-03, 3.9861e-02, -2.3391e-02],
[ 2.8649e-02, 5.9830e-02, 2.7637e-02, 1.8659e-02, -5.6407e-02]],
[[-4.5381e-02, 5.4505e-03, 4.0909e-03, 5.6494e-02, -5.9140e-02],
[ 2.3393e-02, -3.3347e-02, -6.6707e-02, 1.9643e-02, -2.0795e-02],
[-7.7018e-02, 3.0887e-02, -1.8568e-02, 6.2216e-02, 1.9621e-02],
[ 3.4469e-02, -8.1499e-03, 3.7170e-03, -4.2050e-02, -1.1584e-02],
[ 1.5811e-02, 7.1654e-02, -6.3917e-02, 2.9590e-02, 2.9235e-02]],
[[ 3.9611e-02, -2.3142e-02, 3.7166e-02, -3.2922e-02, -7.0504e-02],
[-5.1643e-03, 7.2051e-02, 4.7072e-02, 4.3785e-02, 3.0908e-02],
[ 4.5853e-02, -5.6516e-02, -7.1385e-02, 4.2695e-02, 2.6823e-02],
[-3.3464e-03, -5.1895e-02, -5.9452e-02, -1.4120e-02, -4.2594e-02],
[-5.9216e-03, -7.9317e-02, 3.2249e-02, 2.3419e-02, 1.3252e-02]],
[[ 1.1583e-02, 7.0880e-02, 1.9811e-02, 8.0881e-02, -4.4006e-02],
[ 3.5248e-02, -2.2885e-02, 2.7275e-02, 4.8847e-02, -2.0026e-02],
[ 3.0439e-02, 7.2296e-02, 7.4953e-02, 4.2624e-02, 3.7546e-02],
[ 6.0431e-02, -7.8233e-02, -2.1510e-02, 7.9771e-02, 2.0746e-02],
[-6.6449e-02, 8.1094e-02, 1.7321e-03, 9.2274e-03, 4.6537e-02]],
[[-4.3946e-02, 3.4462e-02, 4.2152e-02, -3.2618e-02, -3.1919e-02],
[ 3.7446e-02, -1.6665e-02, -4.6682e-02, -4.2671e-02, 3.8549e-02],
[ 2.0830e-02, 3.1240e-02, 1.4524e-02, 6.9791e-02, 6.4641e-02],
[-5.1961e-02, 2.9224e-02, -8.1169e-02, 5.0892e-02, 7.3306e-02],
[-5.3219e-02, 4.4223e-02, 1.6751e-02, 1.0149e-02, 2.7877e-02]]],
[[[ 4.0394e-02, -7.5706e-02, 4.0839e-02, 4.1527e-02, 7.5264e-02],
[ 5.0450e-02, 2.5265e-02, 5.6446e-02, -9.1349e-03, -2.0079e-02],
[ 4.0942e-03, 1.9982e-02, 7.8088e-02, 1.3781e-02, 5.9151e-02],
[ 1.5582e-03, -7.1641e-02, -3.9163e-02, 2.0775e-02, 6.9971e-04],
[ 1.4160e-02, -2.1367e-02, 7.2005e-02, 3.0974e-02, 5.5173e-02]],
[[ 2.6252e-02, 2.1418e-02, -7.9839e-03, -6.4778e-02, -1.7517e-02],
[ 3.0320e-02, -7.2231e-03, -6.9592e-02, -2.1982e-02, -7.9040e-02],
[-6.8035e-02, -4.2568e-02, 6.6144e-03, 3.2699e-02, 4.0103e-03],
[-7.6126e-02, 6.6878e-02, 2.0500e-02, 7.3015e-02, 4.4364e-02],
[ 6.4838e-02, -6.1012e-02, -3.7262e-02, -2.4985e-02, 4.3699e-02]],
[[ 2.9414e-02, -6.4046e-02, -7.9652e-02, 5.7636e-02, 1.8768e-02],
[ 1.3999e-02, 6.1029e-04, -2.5749e-02, 5.8838e-02, -7.4751e-02],
[ 6.9579e-02, 4.0549e-02, 2.6570e-02, 1.9762e-02, 6.2969e-02],
[-3.0153e-02, 4.1174e-02, -1.4508e-02, -7.5177e-02, 6.0151e-02],
[ 3.5434e-02, -3.4538e-02, 9.2284e-04, -6.9311e-02, 5.5276e-02]],
[[-1.1414e-02, 5.8701e-02, -8.5818e-03, 5.3662e-02, -6.6120e-02],
[-6.6017e-02, -7.2396e-02, 3.9557e-02, -3.2917e-02, 2.3871e-02],
[ 3.6651e-02, -5.7215e-02, 7.5414e-02, -2.8045e-02, -5.4822e-03],
[ 6.6979e-03, 4.3628e-02, -7.9500e-02, 1.6581e-02, -2.0443e-03],
[ 4.8128e-02, -2.5281e-03, 6.3143e-02, 5.5622e-02, 5.0823e-02]],
[[ 2.6724e-02, -4.4536e-02, 3.8301e-02, -1.6428e-02, 7.6386e-03],
[ 6.9414e-02, 6.4933e-02, -6.0955e-02, 4.2710e-02, 6.7609e-02],
[ 3.2155e-02, -3.9059e-02, -8.1267e-02, 6.0943e-02, 7.8761e-02],
[ 9.2058e-03, -6.8716e-02, 5.9567e-02, -4.7019e-02, 4.3247e-02],
[-8.0727e-02, -7.7734e-02, 6.4857e-02, 1.4899e-03, 4.7743e-02]],
[[ 4.5334e-02, 1.9591e-02, 4.8368e-02, -7.7771e-02, -2.2575e-02],
[ 2.3952e-02, -7.0499e-02, -6.3645e-02, -1.8650e-02, -2.2908e-02],
[ 4.0042e-02, 5.1475e-02, -4.9496e-02, -7.1370e-02, -7.3195e-02],
[ 7.5049e-02, -5.0792e-02, 7.1668e-02, 2.3936e-02, 7.4133e-02],
[ 7.1541e-03, 3.1869e-02, -7.6662e-02, -1.1101e-02, -6.0137e-02]]],
[[[-3.1704e-02, -6.3101e-02, 5.3376e-02, -7.5087e-02, 7.4585e-02],
[ 3.3090e-03, -1.2829e-02, 4.2248e-03, -3.8829e-02, 2.2651e-02],
[-3.0851e-02, 6.0922e-02, 6.3309e-02, -1.0301e-02, -3.4664e-04],
[ 3.7358e-03, 5.7330e-02, 7.6840e-02, 4.9586e-02, -5.9538e-02],
[ 3.8326e-02, -6.2273e-02, 3.8199e-02, 9.8926e-03, -6.3158e-02]],
[[ 3.4346e-02, -3.0442e-02, 9.5122e-03, 5.9389e-03, 6.4301e-02],
[ 3.7045e-02, -1.4779e-02, 5.3899e-02, 1.2880e-02, 3.2471e-02],
[-4.9418e-02, -7.3648e-02, -1.2379e-02, 6.0696e-02, -4.2189e-03],
[-5.9245e-03, -2.1985e-02, 3.4781e-02, 6.4471e-02, -6.9246e-02],
[-5.3104e-02, 4.7656e-02, -4.7472e-02, 5.0731e-02, 6.2965e-02]],
[[-7.3520e-02, -2.9047e-02, 2.4189e-02, 9.0452e-03, 3.1695e-02],
[-5.4734e-02, 2.8641e-02, -6.1692e-02, -2.3195e-02, 1.1079e-02],
[ 5.4220e-02, -7.6615e-02, 5.6999e-02, 2.3648e-03, -4.8024e-02],
[ 2.4688e-02, 2.6944e-02, -5.6788e-02, 4.2183e-02, -2.4088e-02],
[ 5.4214e-03, -1.7575e-02, 6.7586e-02, 5.1225e-02, 2.3619e-02]],
[[-2.5996e-03, -7.9755e-02, 6.8051e-02, 7.2919e-02, 7.9731e-02],
[-5.1041e-02, -2.0511e-02, 4.2875e-02, -3.9107e-02, 7.6118e-02],
[ 9.0300e-03, 3.5622e-02, 1.0749e-02, -1.6854e-02, -5.9495e-02],
[ 3.1847e-02, 7.7569e-02, 3.8437e-02, -3.2095e-02, 8.4368e-03],
[ 2.2344e-02, 5.6201e-02, 4.6436e-02, -6.0905e-02, -6.2908e-02]],
[[ 1.1728e-02, -2.6272e-03, -3.3887e-02, -4.3492e-02, 1.4497e-02],
[ 3.9038e-02, -6.3376e-02, -2.3013e-02, -4.0424e-02, 1.4657e-02],
[ 4.9225e-03, -3.3368e-02, 1.8460e-02, -3.6485e-02, -6.4865e-02],
[-6.4731e-05, -6.8108e-02, -1.3410e-02, -3.4235e-02, -6.4206e-02],
[-1.7974e-02, 3.3237e-02, 5.5293e-02, -3.7493e-02, 2.4666e-02]],
[[ 6.7512e-02, -8.9689e-03, -2.2933e-02, 1.6010e-02, 6.9638e-02],
[-3.7105e-02, 6.7179e-02, -1.0214e-02, -4.8984e-02, -4.1328e-02],
[-8.6841e-03, -1.7692e-02, -1.4424e-02, 7.3771e-02, 1.2215e-02],
[-7.9845e-02, 6.7047e-02, 3.1997e-02, -5.1304e-02, 6.0482e-02],
[-1.4893e-02, -2.7676e-02, 5.4934e-02, 6.3990e-02, 2.7662e-02]]],
...,
[[[ 1.0199e-02, -5.2322e-02, -3.7206e-02, -1.4544e-02, -4.6219e-02],
[-1.2879e-02, 4.7828e-02, 7.8953e-02, -1.3771e-02, 2.3383e-02],
[-2.0076e-02, -2.9035e-02, 5.4802e-02, 3.0305e-02, -2.6655e-02],
[ 4.6331e-02, 6.4619e-02, 3.6669e-02, -5.7059e-02, -1.2394e-02],
[-4.7441e-02, 3.9712e-02, -6.8904e-02, -7.4229e-02, -5.4093e-03]],
[[-3.7258e-02, 8.1076e-02, -7.6125e-02, -2.9282e-02, 1.0107e-02],
[ 2.3341e-02, 2.7966e-02, 4.9232e-02, 4.2246e-02, -2.2956e-02],
[ 6.1198e-02, 5.5776e-02, -6.1231e-02, -7.3973e-02, -1.8083e-02],
[ 2.4045e-02, 5.3863e-02, 3.1777e-02, -3.6017e-02, 7.9472e-02],
[-6.9948e-02, 6.3699e-02, 4.3250e-02, 4.6033e-02, 8.9336e-04]],
[[ 2.5100e-02, 5.6652e-03, -3.4297e-02, -4.4208e-02, -3.4073e-03],
[ 1.6430e-02, -5.2721e-02, 3.6158e-02, 7.7301e-02, 4.7632e-03],
[ 7.6485e-02, -4.3372e-03, 4.2569e-02, -3.6177e-02, -2.1329e-03],
[-8.4518e-03, -9.9161e-03, -2.1800e-02, -6.9410e-02, 5.5372e-02],
[ 7.8497e-02, 6.9532e-02, -1.9760e-03, -4.9298e-02, -3.0007e-02]],
[[ 5.7926e-02, -1.7255e-02, 5.0323e-02, 7.3531e-03, 7.3418e-02],
[ 5.8616e-02, 3.4145e-02, 6.8296e-02, -8.6052e-03, -1.9639e-02],
[-7.3903e-03, 3.6604e-02, -6.6443e-02, 4.2403e-02, -1.6865e-02],
[-4.4089e-02, -3.2403e-02, -3.5313e-02, 6.2766e-02, 4.4323e-02],
[-7.7840e-02, -5.1307e-02, 3.1470e-02, 2.4083e-02, -3.1171e-02]],
[[-7.2114e-02, 4.7854e-02, -3.7292e-02, 3.6088e-02, -1.2283e-02],
[ 2.4217e-02, 4.2792e-02, 4.0946e-02, -4.8699e-02, -2.0729e-02],
[ 4.8880e-02, -8.1478e-02, -7.3582e-02, -2.0240e-02, 5.9753e-02],
[ 3.2898e-02, -1.7339e-02, -5.6911e-02, 6.5161e-02, 1.7858e-02],
[ 3.9480e-02, -3.2163e-02, 3.1307e-02, -5.5870e-02, -1.6457e-02]],
[[ 2.2531e-02, -8.0621e-02, 6.8040e-02, -5.3297e-02, 6.8076e-02],
[-2.8019e-02, -4.2122e-02, 9.4667e-03, 1.1930e-02, 5.3379e-02],
[ 7.0529e-03, -7.5513e-02, -5.8059e-03, -2.2628e-02, -3.1337e-02],
[ 4.6129e-02, 1.0021e-02, 1.3801e-02, 5.4890e-02, 8.2099e-03],
[-3.4219e-02, -3.5771e-02, 4.9384e-02, 4.2971e-03, 1.0185e-02]]],
[[[ 1.5722e-02, -1.1761e-02, -5.5889e-02, -4.0064e-02, 4.6955e-02],
[ 3.8323e-02, 1.4971e-02, 4.7523e-02, -7.3487e-02, -6.6887e-02],
[ 5.4269e-02, 6.5048e-03, -2.5124e-02, 7.0294e-02, 2.8026e-02],
[-7.3427e-02, -6.3417e-02, -6.9835e-02, 5.1313e-02, -3.0339e-02],
[-3.7274e-02, -6.8501e-02, 1.3327e-02, 7.0011e-03, 7.3245e-02]],
[[ 3.7582e-02, 8.0081e-02, -4.8837e-02, 7.3556e-03, 6.0067e-02],
[-3.2691e-02, -4.5259e-02, -6.4128e-02, -5.0167e-02, -2.3580e-02],
[-5.0516e-02, -7.3288e-02, -5.1057e-02, -4.8879e-02, 1.4947e-02],
[ 5.5937e-02, 2.8608e-02, -5.1257e-02, -9.8808e-03, 7.7368e-03],
[-6.1867e-03, 5.5593e-02, -6.4040e-02, -1.9817e-02, 4.8827e-02]],
[[ 7.0229e-02, 3.4918e-02, 3.4894e-02, -3.7108e-02, 5.1772e-02],
[-8.0578e-02, -1.9076e-02, -5.2325e-02, -3.1958e-02, -1.4594e-02],
[-6.5627e-02, -4.4570e-02, 7.8603e-02, -1.8659e-02, -4.3064e-02],
[-5.8323e-02, 2.8146e-02, 2.7508e-02, -2.0674e-03, -2.3774e-02],
[ 5.3830e-02, 3.1012e-03, -1.3121e-02, 4.2424e-02, 1.3136e-03]],
[[ 6.2128e-02, -8.1062e-02, 5.5346e-03, 4.9415e-02, 1.9914e-02],
[ 4.5311e-02, 4.5756e-02, -1.6908e-02, 1.4688e-02, 3.0094e-02],
[-4.5764e-03, -3.1909e-02, 6.0014e-02, -6.2276e-02, -3.4248e-02],
[ 3.9257e-02, 1.9695e-02, -8.7933e-03, 7.9642e-02, -3.3138e-02],
[-1.9051e-02, 4.8750e-02, -4.9173e-03, 7.9091e-02, 1.4798e-02]],
[[-5.5164e-02, 3.3881e-02, -3.4420e-04, 6.3908e-02, 7.2785e-02],
[ 3.9900e-02, 6.5517e-03, -8.0552e-02, -1.5329e-02, -5.7844e-02],
[-6.8076e-03, 5.0193e-02, 7.4728e-02, -2.9026e-02, 2.0508e-02],
[ 6.2877e-02, 8.0455e-02, 5.1410e-02, -3.8253e-02, -6.9830e-02],
[-1.4001e-02, -6.4940e-02, 4.3407e-03, -2.0574e-03, -6.3191e-02]],
[[ 1.8252e-02, 8.8192e-03, 1.6308e-02, 8.0085e-02, 5.4516e-02],
[ 3.2235e-02, -1.4626e-02, -5.4716e-02, 7.6015e-02, -5.5979e-02],
[ 3.0786e-02, 3.1961e-02, -1.8162e-02, 6.9846e-02, 3.8360e-02],
[ 7.4391e-02, 1.7810e-02, 5.8638e-02, -2.3435e-02, -5.3262e-02],
[ 7.7326e-02, -2.3349e-02, -7.6374e-02, -1.5019e-02, -4.1779e-02]]],
[[[-4.0523e-02, -1.3673e-02, 1.7464e-02, -8.0872e-02, 1.3032e-02],
[-5.7647e-02, 7.6154e-02, 2.9756e-02, 6.1005e-02, -6.2796e-02],
[-1.8865e-02, -3.1772e-02, -3.0434e-02, -6.4698e-02, -7.2306e-02],
[-1.3506e-02, 7.4316e-02, 3.0436e-02, -1.4313e-02, 7.2392e-02],
[-3.5933e-02, 1.1374e-02, -1.8395e-02, -3.0843e-02, 5.5321e-02]],
[[ 1.4394e-02, 3.6295e-02, 4.9757e-02, -4.6856e-03, 9.4505e-03],
[-5.9689e-02, 5.7720e-02, 7.4232e-02, 7.1351e-02, 3.7225e-02],
[ 3.1824e-02, -3.5381e-02, -3.4969e-02, 8.2402e-03, -2.4470e-02],
[-7.7093e-02, 4.4498e-02, 4.7004e-02, -7.2011e-02, -3.8237e-02],
[-6.7163e-02, -1.7111e-02, 5.9214e-02, 7.7507e-02, -4.4913e-02]],
[[-8.0340e-02, 1.9304e-02, 1.0000e-03, 4.0900e-03, -6.6380e-02],
[-6.6810e-02, -6.3729e-03, 2.0123e-02, -4.3774e-02, -3.4981e-02],
[-6.8541e-02, 1.6689e-02, -7.1792e-02, 1.5246e-02, -6.8854e-04],
[ 4.3843e-02, 2.6870e-02, -3.8101e-04, -7.8826e-02, 6.5015e-02],
[-2.0440e-02, 7.9386e-02, -3.8219e-03, -2.4580e-02, 4.0634e-02]],
[[ 5.5544e-02, -3.7836e-02, -7.9338e-02, 7.0619e-02, -3.3446e-02],
[-3.2902e-03, -4.8389e-02, 4.7453e-02, 5.8127e-02, -4.8315e-03],
[-5.6379e-02, -5.8292e-02, -3.9011e-02, -4.0794e-03, 4.6537e-02],
[ 8.0529e-02, 6.2578e-02, 2.7042e-02, -7.3989e-02, 8.9902e-03],
[-3.2401e-02, -6.5050e-02, 6.9688e-02, -6.6803e-02, 3.6991e-02]],
[[-2.2901e-02, 4.9715e-02, -5.3222e-02, -7.6049e-02, -6.6969e-02],
[ 9.8178e-03, -6.9489e-02, -3.7907e-02, 3.8849e-02, -3.6555e-02],
[-4.1870e-02, 1.7604e-02, 1.3747e-02, 1.2642e-02, -2.0011e-02],
[-6.3959e-02, 6.5200e-02, -3.1475e-02, -5.4463e-02, -7.0515e-02],
[-3.3264e-02, 5.0797e-04, 4.7155e-02, 5.2026e-02, -6.2295e-02]],
[[-7.0623e-02, -3.8786e-02, -1.6983e-02, -4.0680e-02, -3.5289e-03],
[ 6.6456e-02, -5.6733e-02, -3.2044e-02, -8.0967e-02, 6.8003e-02],
[ 7.7967e-02, -1.8929e-02, 4.1282e-02, -4.3998e-02, 7.8460e-02],
[ 4.9829e-02, -3.3562e-02, -7.5281e-02, 5.1357e-02, 2.8261e-02],
[ 6.7955e-02, -3.4526e-02, -4.7810e-02, 7.8384e-02, -1.2523e-02]]]])
conv2.bias
tensor([-0.0807, -0.0406, -0.0171, 0.0340, 0.0777, 0.0155, 0.0788, -0.0089,
-0.0693, -0.0276, 0.0578, 0.0165, 0.0525, -0.0673, -0.0171, 0.0804])
fc1.weight
tensor([[-0.0032, -0.0052, -0.0235, ..., -0.0029, -0.0208, 0.0015],
[-0.0460, 0.0396, 0.0050, ..., 0.0194, 0.0410, 0.0174],
[-0.0391, -0.0378, -0.0265, ..., 0.0376, 0.0157, 0.0217],
...,
[-0.0304, -0.0122, 0.0140, ..., -0.0028, 0.0002, -0.0292],
[ 0.0346, 0.0187, 0.0185, ..., -0.0256, 0.0434, -0.0359],
[-0.0116, 0.0302, 0.0412, ..., 0.0163, -0.0479, -0.0163]])
fc1.bias
tensor([-0.0453, 0.0006, 0.0223, 0.0062, -0.0188, 0.0133, 0.0158, -0.0061,
0.0145, 0.0035, -0.0145, -0.0451, 0.0352, 0.0302, -0.0086, 0.0333,
0.0025, 0.0102, 0.0128, -0.0002, -0.0021, 0.0439, 0.0468, 0.0120,
-0.0089, 0.0115, -0.0260, 0.0178, 0.0217, 0.0456, -0.0116, 0.0224,
-0.0207, -0.0060, -0.0386, 0.0405, -0.0180, -0.0116, -0.0009, 0.0239,
-0.0203, -0.0064, 0.0406, -0.0173, 0.0378, 0.0180, 0.0195, 0.0145,
-0.0435, 0.0157, 0.0363, -0.0170, 0.0064, -0.0107, 0.0341, -0.0227,
0.0179, -0.0221, -0.0264, -0.0018, -0.0104, -0.0481, -0.0343, -0.0245,
-0.0190, -0.0425, 0.0329, -0.0488, 0.0279, 0.0034, -0.0491, -0.0422,
0.0237, 0.0211, -0.0260, -0.0380, 0.0010, -0.0025, -0.0367, 0.0386,
-0.0090, 0.0157, -0.0292, -0.0302, 0.0423, -0.0476, -0.0401, 0.0195,
0.0315, 0.0285, 0.0456, 0.0297, 0.0091, -0.0291, 0.0301, -0.0231,
-0.0055, 0.0359, 0.0164, -0.0297, 0.0243, -0.0429, 0.0311, 0.0072,
-0.0400, -0.0473, 0.0470, -0.0068, -0.0160, 0.0092, -0.0327, -0.0040,
0.0245, 0.0178, -0.0249, -0.0146, 0.0401, 0.0019, 0.0216, -0.0497])
fc2.weight
tensor([[ 0.0327, -0.0668, 0.0384, ..., -0.0704, 0.0015, -0.0560],
[-0.0657, -0.0632, 0.0364, ..., 0.0721, 0.0100, 0.0768],
[ 0.0130, 0.0838, -0.0002, ..., 0.0178, 0.0226, -0.0903],
...,
[ 0.0313, -0.0549, 0.0724, ..., 0.0874, -0.0439, -0.0327],
[ 0.0371, -0.0318, 0.0875, ..., 0.0631, -0.0813, 0.0321],
[-0.0071, 0.0583, 0.0542, ..., -0.0285, 0.0750, 0.0572]])
fc2.bias
tensor([ 0.0134, -0.0475, 0.0059, 0.0127, 0.0480, -0.0361, 0.0095, 0.0558,
0.0322, 0.0769, 0.0570, -0.0611, -0.0341, -0.0588, -0.0095, 0.0799,
-0.0563, 0.0562, -0.0833, -0.0296, 0.0359, -0.0193, -0.0529, -0.0428,
-0.0597, -0.0521, 0.0202, -0.0128, -0.0195, 0.0684, -0.0416, 0.0795,
0.0608, 0.0296, -0.0304, -0.0381, -0.0834, 0.0112, 0.0332, 0.0466,
-0.0837, -0.0375, 0.0573, -0.0380, 0.0802, -0.0668, 0.0514, -0.0070,
0.0033, 0.0871, -0.0178, 0.0522, -0.0571, 0.0472, 0.0561, 0.0215,
0.0012, -0.0419, -0.0055, -0.0174, -0.0204, -0.0619, -0.0016, -0.0784,
-0.0819, -0.0562, -0.0807, 0.0765, 0.0316, 0.0423, -0.0063, -0.0210,
0.0315, 0.0806, 0.0109, -0.0175, 0.0705, -0.0466, 0.0097, 0.0098,
0.0798, 0.0796, -0.0783, -0.0730])
fc3.weight
tensor([[-0.0351, 0.0782, 0.0364, -0.0547, 0.0099, -0.0188, -0.0582, -0.0056,
-0.1074, 0.0905, -0.0629, 0.1084, -0.0456, -0.0382, 0.0472, -0.0674,
-0.0414, -0.0523, -0.0009, -0.0946, -0.0463, 0.0898, 0.0631, -0.0910,
0.0091, -0.0383, -0.0311, 0.0978, 0.0204, -0.1040, -0.0793, -0.1019,
0.0631, -0.0136, -0.0283, -0.0260, -0.1050, 0.0029, -0.1016, -0.0253,
-0.0350, 0.0862, 0.0013, 0.0482, -0.0284, 0.0073, 0.0053, 0.0703,
0.0078, -0.0487, 0.0899, -0.0934, 0.0385, -0.0764, 0.0070, -0.0486,
0.0197, 0.0372, -0.0763, 0.0477, -0.1010, -0.0224, -0.0491, -0.0548,
-0.0461, 0.0629, 0.0691, 0.0064, -0.0637, -0.0727, 0.0074, -0.0961,
-0.0429, 0.0362, 0.0140, 0.0659, -0.0640, 0.0336, 0.1020, -0.0572,
0.0779, 0.0741, -0.0588, -0.0856],
[ 0.0477, -0.0715, 0.0228, 0.0505, 0.0311, -0.0720, 0.0702, 0.0573,
0.0241, 0.0151, -0.1036, -0.0023, 0.0145, 0.0096, -0.0538, 0.0805,
0.0951, 0.1079, -0.0928, 0.0317, -0.0034, -0.0983, 0.0781, 0.0109,
-0.0001, -0.0881, 0.0208, 0.0635, 0.0434, 0.1080, -0.0625, -0.0924,
0.0947, 0.0438, 0.0596, -0.0324, -0.0555, -0.0729, 0.0614, -0.0416,
-0.0770, -0.0748, -0.0193, -0.0991, -0.0900, 0.0201, 0.0715, -0.0605,
0.0033, -0.0418, 0.0473, -0.1069, 0.0095, -0.0540, 0.0695, 0.1042,
0.0245, 0.0715, 0.1073, 0.0643, -0.0956, -0.0843, 0.0391, 0.0033,
0.0787, -0.1019, -0.0056, -0.0830, 0.0327, 0.0031, -0.0049, -0.0971,
0.0549, -0.0827, -0.0933, -0.0676, -0.0909, -0.0925, 0.0300, 0.0154,
0.0706, -0.0158, -0.0479, -0.0853],
[-0.0502, -0.0625, 0.0947, -0.0247, 0.0150, -0.0101, 0.0695, 0.0064,
0.0390, -0.0735, -0.0347, 0.0255, -0.0964, 0.0016, -0.0568, 0.1050,
-0.0972, 0.0176, 0.0756, -0.0781, 0.0529, -0.0751, 0.0823, 0.0953,
0.0651, -0.0740, -0.0460, 0.0549, -0.0199, -0.0245, -0.1058, 0.0528,
-0.1070, 0.0086, 0.0665, -0.0895, 0.0798, -0.0902, 0.1072, 0.0173,
0.0257, 0.0481, -0.0289, 0.0291, -0.1057, -0.0429, -0.1033, -0.0716,
-0.0690, 0.0628, 0.0718, -0.1040, -0.1042, 0.0831, 0.0844, 0.1073,
0.0846, 0.0477, 0.0619, 0.0479, -0.0872, -0.0647, 0.0556, 0.0539,
-0.0421, -0.0697, 0.0625, 0.0688, 0.0778, 0.0239, -0.0431, 0.0142,
0.0107, 0.0207, 0.0443, -0.0164, 0.0764, -0.0577, -0.0278, -0.0256,
0.0416, -0.0133, -0.0185, 0.0236],
[ 0.1058, 0.0798, 0.0002, 0.0795, 0.1071, -0.0654, 0.0125, 0.0800,
-0.0396, -0.0300, 0.0885, 0.0236, -0.0025, -0.0938, -0.0494, 0.0230,
0.0239, -0.0673, -0.0466, 0.0425, 0.0175, 0.0395, 0.1038, 0.0651,
-0.0052, 0.0480, -0.0311, -0.0081, 0.0582, 0.0096, 0.0950, 0.0721,
-0.0648, 0.0789, -0.0936, 0.0235, -0.0654, -0.0109, 0.0745, 0.0503,
0.0175, 0.0719, 0.1069, -0.0279, -0.0692, -0.0972, 0.0624, -0.0348,
0.0703, 0.0675, 0.0817, -0.0586, 0.0760, -0.0307, -0.0205, -0.0042,
0.0939, 0.0528, -0.0242, -0.0632, -0.0184, -0.0781, 0.0133, 0.0246,
0.0962, -0.0454, 0.0117, -0.0783, 0.0378, 0.0393, 0.0874, 0.0969,
-0.1045, 0.0167, -0.0041, -0.0450, -0.0738, -0.0291, -0.0480, 0.1086,
0.0284, -0.0981, 0.0933, 0.0159],
[-0.1059, 0.0177, -0.0834, 0.0280, -0.0464, 0.0020, 0.0380, 0.1060,
0.0865, -0.0857, -0.0820, -0.0471, 0.0236, -0.0603, 0.0667, 0.0142,
0.0905, -0.0847, -0.0254, -0.0391, -0.0173, -0.1018, -0.0318, 0.0223,
-0.0592, 0.0776, 0.0137, -0.0570, -0.0139, 0.0646, -0.0869, -0.0624,
-0.0882, 0.0348, 0.0161, -0.0336, 0.0755, 0.1023, -0.1000, -0.0267,
-0.1080, 0.0422, -0.0045, 0.0512, -0.0861, -0.0749, 0.1039, -0.0755,
0.0404, 0.0974, -0.0386, -0.0557, 0.0769, -0.0065, 0.0711, 0.0475,
0.0997, 0.0045, 0.0312, -0.0187, -0.1067, -0.0217, 0.0200, -0.0657,
0.0338, 0.0409, 0.0502, 0.0765, -0.0575, 0.0739, -0.0023, -0.0798,
0.0923, 0.0901, -0.0271, 0.0934, 0.0941, -0.0731, 0.0918, 0.0247,
-0.0655, -0.0685, -0.0875, 0.0479],
[-0.0871, 0.0840, -0.0604, -0.0193, -0.0877, -0.0734, 0.0127, -0.0357,
0.0891, -0.0521, -0.0469, -0.0877, -0.0775, 0.0123, -0.0194, -0.1078,
0.0492, -0.1044, -0.0714, 0.0606, 0.0053, -0.0227, 0.0231, -0.0439,
-0.0554, 0.0112, -0.0717, 0.0829, -0.0140, 0.0908, -0.0258, 0.0669,
-0.0290, -0.0555, 0.1082, -0.0481, -0.0175, -0.0107, -0.0297, -0.0472,
-0.0606, -0.0301, -0.0846, -0.0839, 0.0077, 0.0267, 0.0926, -0.0119,
-0.0195, 0.0077, 0.0336, 0.0981, 0.0040, -0.0447, -0.0131, -0.1037,
-0.1037, 0.0380, 0.0343, 0.1038, 0.0378, -0.0402, -0.0581, -0.0523,
0.0604, 0.0402, -0.0838, -0.0718, -0.0488, 0.0242, 0.0727, -0.0429,
0.0225, -0.0209, 0.0964, 0.0810, -0.0704, 0.0720, -0.0120, 0.0704,
-0.0779, 0.0169, -0.0885, 0.0133],
[ 0.0809, -0.0432, -0.0358, 0.0023, -0.0809, -0.0058, -0.0977, 0.0731,
0.0576, -0.0883, 0.0758, 0.0168, -0.0107, -0.0284, 0.0640, 0.0921,
-0.1068, -0.0934, 0.0853, 0.0940, -0.0152, 0.0899, 0.0431, 0.0047,
-0.0164, -0.0339, 0.0470, 0.0021, -0.0960, 0.1058, 0.0646, -0.0491,
-0.1037, 0.0242, 0.0083, 0.0221, -0.0810, -0.0367, -0.0478, -0.0849,
-0.0836, 0.0972, -0.0792, -0.1085, -0.0072, -0.0512, -0.0573, -0.0067,
0.0944, 0.0517, 0.0925, 0.0974, 0.0239, 0.1011, -0.1045, -0.0396,
0.0565, -0.0717, -0.0250, -0.0896, 0.0717, 0.0155, -0.0818, -0.0084,
-0.0022, -0.1046, -0.0607, -0.0657, -0.0968, 0.0873, -0.0598, -0.0593,
0.0558, -0.1064, 0.0574, 0.1044, -0.0233, -0.0505, 0.1041, 0.1085,
-0.0538, -0.0595, 0.0956, 0.0604],
[ 0.1015, -0.0722, 0.0488, 0.0101, 0.0202, -0.0593, 0.0623, -0.0953,
0.0665, 0.1086, -0.1021, 0.0690, 0.0366, -0.0012, 0.0754, 0.0651,
0.0005, 0.0642, 0.0149, 0.0153, -0.0449, -0.0533, -0.0103, -0.0387,
-0.0602, 0.1004, 0.0283, 0.0476, 0.0596, 0.0504, 0.0366, -0.0626,
-0.0361, -0.0263, -0.1079, 0.0652, -0.1043, 0.0658, -0.1018, 0.0300,
-0.0766, -0.0371, -0.0634, -0.0091, -0.0625, 0.0402, -0.0438, -0.0524,
-0.0687, 0.0636, 0.0641, 0.0817, -0.1090, 0.0409, -0.0558, 0.1062,
0.0414, -0.0629, 0.0131, -0.0563, 0.0378, 0.0774, -0.0750, 0.0777,
0.0578, -0.0936, 0.0269, -0.0396, -0.0964, -0.0894, -0.0125, -0.0236,
0.0014, -0.0819, -0.0124, 0.0247, 0.1023, -0.0884, 0.0624, -0.0268,
0.0912, 0.0312, -0.0989, -0.0456],
[-0.0516, -0.0043, -0.0102, 0.0299, 0.0665, -0.0009, 0.1026, -0.0169,
0.0006, 0.0715, 0.1036, 0.0087, 0.0707, -0.0396, -0.0476, -0.0896,
0.0608, -0.0180, -0.0367, -0.0653, 0.0630, 0.1080, 0.0807, -0.0254,
-0.0631, -0.0055, 0.0611, 0.0971, 0.0037, 0.0689, 0.0861, 0.0770,
0.0932, 0.0045, -0.0973, -0.0262, -0.0513, -0.0835, -0.0036, 0.0355,
-0.0524, -0.0042, 0.0043, 0.0252, 0.0437, -0.0898, -0.0324, 0.0382,
0.0322, 0.0678, 0.0735, 0.0450, -0.0094, -0.0311, 0.0541, -0.0942,
0.0945, 0.0062, 0.0225, 0.0156, -0.0396, 0.0228, 0.0718, -0.0720,
-0.1054, 0.0230, 0.0869, 0.0996, 0.0456, 0.0548, 0.0861, 0.0787,
0.0159, 0.0857, -0.0790, -0.0252, 0.0639, 0.0059, 0.0513, 0.0966,
-0.0251, -0.0521, -0.0341, -0.0758],
[-0.0203, -0.0479, -0.0209, -0.0015, -0.0537, -0.0816, -0.0397, 0.0425,
-0.0706, -0.0772, 0.0870, 0.0587, -0.0714, 0.0805, -0.0104, 0.0030,
-0.0112, -0.0341, 0.0635, -0.0202, 0.0839, -0.0273, 0.0969, -0.1013,
-0.0227, 0.0159, -0.0036, 0.0428, -0.0474, 0.0933, -0.0499, 0.0124,
-0.0043, 0.0435, 0.0785, -0.0419, 0.0231, 0.1041, -0.0128, -0.0573,
0.0456, 0.0235, 0.0565, -0.0868, -0.1073, -0.0534, 0.0074, -0.0648,
-0.0178, -0.0929, 0.1065, 0.0231, -0.0270, -0.0733, 0.0656, 0.0857,
-0.0941, 0.0018, -0.0426, -0.0115, 0.0978, -0.0780, -0.0622, 0.0838,
-0.0606, -0.0661, -0.0331, 0.0457, -0.0763, -0.0903, -0.0381, -0.0198,
-0.0369, -0.0135, -0.1033, 0.1031, 0.0206, 0.0382, -0.1005, -0.0188,
-0.0711, -0.0649, 0.0517, -0.0060]])
fc3.bias
tensor([ 0.0853, -0.1027, -0.0893, -0.0900, 0.0732, 0.0718, -0.0705, 0.0435,
0.0994, 0.0521])
SAVE MODEL to /home/yongqiang/pytorch_work/end2end-asr-pytorch-example/log/epoch_9.th
Process finished with exit code 0
(pt-1.4_py-3.6) yongqiang@yongqiang:~/pytorch_work/end2end-asr-pytorch-example/log$ ll
total 352
drwxrwxrwx 1 yongqiang yongqiang 512 Jun 20 16:01 ./
drwxrwxrwx 1 yongqiang yongqiang 512 Jun 20 16:01 ../
-rw-rw-rw- 1 yongqiang yongqiang 250027 Jun 20 16:01 epoch_9.th
(pt-1.4_py-3.6) yongqiang@yongqiang:~/pytorch_work/end2end-asr-pytorch-example/log$
(pt-1.4_py-3.6) yongqiang@yongqiang:~/pytorch_work/end2end-asr-pytorch-example/log$ du -sh *
352K epoch_9.th
(pt-1.4_py-3.6) yongqiang@yongqiang:~/pytorch_work/end2end-asr-pytorch-example/log$
4. torch.load(load_path)
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# yongqiang cheng
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os
import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
def load_model(load_path):
"""
Loading model
args:
load_path: string
"""
checkpoint = torch.load(load_path)
epoch = checkpoint['epoch']
metrics = checkpoint['metrics']
if 'args' in checkpoint:
args = checkpoint['args']
label2id = checkpoint['label2id']
id2label = checkpoint['id2label']
model = init_transformer_model(args, label2id, id2label)
model.load_state_dict(checkpoint['model_state_dict'])
if args.cuda:
model = model.cuda()
opt = init_optimizer(args, model)
if opt is not None:
opt.optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
if constant.args.loss == "ce":
opt._step = checkpoint['optimizer_params']['_step']
opt._rate = checkpoint['optimizer_params']['_rate']
opt.warmup = checkpoint['optimizer_params']['warmup']
opt.factor = checkpoint['optimizer_params']['factor']
opt.model_size = checkpoint['optimizer_params']['model_size']
elif constant.args.loss == "ctc":
opt.lr = checkpoint['optimizer_params']['lr']
opt.lr_anneal = checkpoint['optimizer_params']['lr_anneal']
else:
print("Need to define loss type")
return model, opt, epoch, metrics, args, label2id, id2label
def save_model(model, epoch, optimizer, save_folder, name, loss, best_model=False):
"""
Saving model
"""
if best_model:
save_path = "{}/{}/best_model.th".format(save_folder, name)
else:
save_path = "{}/{}/epoch_{}.th".format(save_folder, name, epoch)
if not os.path.exists(save_folder + "/" + name):
os.makedirs(save_folder + "/" + name)
print("SAVE MODEL to", save_path)
if loss == "ce":
config = {
'label2id': "label2id",
'id2label': "id2label",
'args': "args",
'epoch': "epoch",
'model_state_dict': model.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
'optimizer_params': {
'_step': "_step",
'_rate': "_rate",
'warmup': "warmup",
'factor': "factor",
'model_size': "model_size"
},
'metrics': "metrics"
}
elif loss == "ctc":
config = {
'label2id': "label2id",
'id2label': "id2label",
'args': "args",
'epoch': epoch,
'model_state_dict': model.state_dict(),
'optimizer_state_dict': optimizer.state_dict(),
'optimizer_params': {
'lr': "lr",
'lr_anneal': "lr_anneal"
},
'metrics': "metrics"
}
else:
print("Loss is not defined")
torch.save(config, save_path)
# Define model
class TheModelClass(nn.Module):
def __init__(self):
super(TheModelClass, self).__init__()
self.conv1 = nn.Conv2d(3, 6, 5)
self.pool = nn.MaxPool2d(2, 2)
self.conv2 = nn.Conv2d(6, 16, 5)
self.fc1 = nn.Linear(16 * 5 * 5, 120)
self.fc2 = nn.Linear(120, 84)
self.fc3 = nn.Linear(84, 10)
def forward(self, x):
x = self.pool(F.relu(self.conv1(x)))
x = self.pool(F.relu(self.conv2(x)))
x = x.view(-1, 16 * 5 * 5)
x = F.relu(self.fc1(x))
x = F.relu(self.fc2(x))
x = self.fc3(x)
return x
# Initialize model
model = TheModelClass()
# Initialize optimizer
optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)
# Print model's state_dict
print("Model's state_dict:")
for param_tensor in model.state_dict():
print(param_tensor, "\t", model.state_dict()[param_tensor].size())
# Print optimizer's state_dict
print("Optimizer's state_dict:")
for var_name in optimizer.state_dict():
print(var_name, "\t", optimizer.state_dict()[var_name])
# # Print model's state_dict
# print("Model's state_dict:")
# for param_tensor in model.state_dict():
# print(param_tensor, "\n", model.state_dict()[param_tensor])
#
# model_folder = "/home/yongqiang/pytorch_work/end2end-asr-pytorch-example"
# save_model(model=model, epoch=9, optimizer=optimizer, save_folder=model_folder, name="log", loss="ctc", best_model=False)
load_path = "/home/yongqiang/pytorch_work/end2end-asr-pytorch-example/log/epoch_9.th"
loss = "ctc"
checkpoint = torch.load(load_path)
print("load_path:", load_path)
epoch = checkpoint['epoch']
metrics = checkpoint['metrics']
if 'args' in checkpoint:
args = checkpoint['args']
label2id = checkpoint['label2id']
id2label = checkpoint['id2label']
model_state_dict_data = checkpoint['model_state_dict']
for k, v in list(model_state_dict_data.items()):
print(k, "\t", v.size())
for k, v in model_state_dict_data.items():
print(k, "\n", v)
if loss == "ce":
_step = checkpoint['optimizer_params']['_step']
_rate = checkpoint['optimizer_params']['_rate']
warmup = checkpoint['optimizer_params']['warmup']
factor = checkpoint['optimizer_params']['factor']
model_size = checkpoint['optimizer_params']['model_size']
elif loss == "ctc":
lr = checkpoint['optimizer_params']['lr']
lr_anneal = checkpoint['optimizer_params']['lr_anneal']
else:
print("Need to define loss type")
/home/yongqiang/miniconda3/envs/pt-1.4_py-3.6/bin/python /home/yongqiang/pytorch_work/end2end-asr-pytorch-example/yongqiang.py
Model's state_dict:
conv1.weight torch.Size([6, 3, 5, 5])
conv1.bias torch.Size([6])
conv2.weight torch.Size([16, 6, 5, 5])
conv2.bias torch.Size([16])
fc1.weight torch.Size([120, 400])
fc1.bias torch.Size([120])
fc2.weight torch.Size([84, 120])
fc2.bias torch.Size([84])
fc3.weight torch.Size([10, 84])
fc3.bias torch.Size([10])
Optimizer's state_dict:
state {}
param_groups [{'lr': 0.001, 'momentum': 0.9, 'dampening': 0, 'weight_decay': 0, 'nesterov': False, 'params': [140319393794736, 140319372878328, 140319372878400, 140319372878472, 140319372878544, 140319372878616, 140319372878688, 140319372878760, 140319372878832, 140319372878904]}]
load_path: /home/yongqiang/pytorch_work/end2end-asr-pytorch-example/log/epoch_9.th
conv1.weight torch.Size([6, 3, 5, 5])
conv1.bias torch.Size([6])
conv2.weight torch.Size([16, 6, 5, 5])
conv2.bias torch.Size([16])
fc1.weight torch.Size([120, 400])
fc1.bias torch.Size([120])
fc2.weight torch.Size([84, 120])
fc2.bias torch.Size([84])
fc3.weight torch.Size([10, 84])
fc3.bias torch.Size([10])
conv1.weight
tensor([[[[ 0.0134, -0.0569, -0.0762, -0.0897, -0.0430],
[ 0.0423, -0.0034, -0.0318, 0.0950, -0.0198],
[ 0.0963, 0.0397, -0.0560, 0.0744, 0.0987],
[ 0.0997, 0.0042, -0.0353, -0.0677, 0.0536],
[ 0.0791, 0.0338, -0.0949, -0.0367, 0.0982]],
[[ 0.0951, 0.0567, 0.1046, -0.0704, -0.0307],
[ 0.0991, 0.0478, 0.0778, 0.0336, 0.0927],
[ 0.0694, 0.0703, 0.1084, 0.1068, 0.0833],
[-0.0444, 0.0951, 0.0369, -0.1047, 0.0575],
[ 0.0866, 0.1128, 0.0547, -0.0558, 0.0164]],
[[ 0.0562, -0.1008, -0.0253, 0.0131, 0.0880],
[-0.0422, -0.0437, 0.0181, -0.0247, -0.0727],
[-0.0072, 0.0113, 0.0032, 0.0571, 0.0431],
[-0.0011, -0.0778, 0.0533, 0.0441, 0.0252],
[-0.0267, 0.0294, -0.0218, -0.1044, -0.0390]]],
[[[ 0.0848, 0.1063, -0.0568, -0.0719, 0.0082],
[-0.0797, 0.0945, -0.0328, -0.0290, -0.0759],
[ 0.0794, -0.0985, -0.0748, -0.0041, 0.0019],
[-0.0375, -0.0602, -0.0819, -0.1079, 0.0773],
[ 0.0944, -0.0515, -0.0621, 0.0124, 0.0627]],
[[ 0.0582, -0.1018, 0.0449, -0.0540, 0.0171],
[-0.0551, 0.1112, -0.0233, -0.1066, 0.0431],
[ 0.0936, -0.1010, -0.0952, -0.0989, -0.0224],
[-0.0374, -0.0663, -0.1145, -0.0551, -0.0156],
[-0.0178, -0.0025, -0.0656, -0.0878, 0.0368]],
[[ 0.0199, -0.0584, 0.0842, -0.1144, -0.0983],
[ 0.0355, 0.0959, -0.0010, -0.0964, 0.0487],
[-0.0616, -0.0055, 0.0847, 0.0304, -0.0560],
[ 0.0960, -0.1062, 0.1005, -0.0658, -0.0791],
[ 0.0265, -0.0321, -0.0826, -0.1112, -0.0275]]],
[[[-0.1136, -0.0062, -0.0536, -0.0138, -0.1113],
[ 0.0939, 0.0356, -0.0073, 0.0578, 0.0617],
[ 0.0470, 0.0967, -0.0801, 0.1070, 0.0272],
[ 0.0662, 0.0654, 0.0242, 0.1028, -0.0372],
[ 0.0699, 0.0887, 0.1111, 0.0259, -0.0816]],
[[ 0.0348, 0.0764, 0.0776, -0.0557, 0.0993],
[ 0.0266, -0.0372, -0.0937, 0.0853, -0.0370],
[ 0.0783, -0.0296, 0.0722, -0.1091, -0.1092],
[ 0.1108, 0.1095, -0.0578, 0.0673, 0.0882],
[ 0.0088, -0.0430, 0.0211, 0.1035, -0.0614]],
[[ 0.0309, 0.0961, -0.0820, 0.0557, 0.1115],
[ 0.0167, 0.0800, 0.0058, 0.1004, -0.0041],
[-0.0032, -0.0768, 0.0156, 0.0554, 0.0967],
[-0.0623, -0.0133, 0.0419, 0.0855, -0.0093],
[-0.0782, 0.0276, -0.0715, -0.0109, 0.0090]]],
[[[-0.0351, 0.0579, -0.0033, 0.0753, 0.0762],
[ 0.0846, 0.0981, -0.0195, -0.0937, 0.0385],
[-0.0608, -0.0624, 0.0757, 0.0834, -0.0371],
[ 0.0948, -0.0822, -0.1009, 0.1132, -0.1114],
[ 0.0245, 0.0276, 0.0120, -0.0866, 0.0686]],
[[ 0.0277, 0.0420, 0.0820, 0.0180, -0.0904],
[-0.0340, 0.0228, 0.1026, 0.0177, 0.0052],
[-0.0608, -0.0869, -0.1016, 0.0253, 0.1009],
[-0.0366, -0.1122, 0.0936, -0.0223, -0.0959],
[-0.0824, 0.1027, 0.0007, -0.0691, -0.0679]],
[[-0.0165, -0.0798, 0.0052, -0.0003, -0.1124],
[-0.0151, 0.0668, 0.1082, -0.0442, -0.1077],
[-0.0055, -0.0172, 0.0627, 0.0263, -0.0847],
[ 0.0741, 0.1126, 0.0694, -0.0641, -0.0290],
[-0.0695, -0.0944, 0.0760, 0.1055, 0.1067]]],
[[[-0.0462, -0.0732, -0.0339, -0.0569, 0.0344],
[ 0.0217, -0.0960, -0.1135, -0.0820, 0.0353],
[ 0.0226, 0.0066, 0.0191, 0.0522, 0.0986],
[-0.0946, 0.0257, -0.0353, 0.0866, -0.0039],
[ 0.0354, -0.0436, -0.0857, 0.0711, 0.0421]],
[[-0.0715, 0.0448, -0.0478, 0.0755, -0.0578],
[ 0.1127, -0.0526, -0.1046, 0.0710, -0.0189],
[-0.0079, 0.1007, -0.0464, -0.0967, 0.0137],
[ 0.0587, -0.0915, 0.0212, 0.0212, 0.0259],
[ 0.0572, 0.0283, 0.0013, 0.1119, -0.0051]],
[[ 0.0268, 0.0909, -0.0809, -0.0398, 0.1101],
[ 0.1020, 0.0993, 0.0176, -0.0125, -0.0770],
[ 0.0786, -0.1061, 0.0761, -0.0004, 0.0296],
[-0.0290, -0.0337, 0.0569, -0.0355, -0.0754],
[ 0.0760, -0.0028, 0.0969, -0.0407, 0.0886]]],
[[[ 0.0132, -0.0045, -0.0614, 0.0071, -0.1035],
[ 0.0208, 0.0518, -0.1080, -0.0129, -0.0524],
[-0.0328, 0.0093, -0.0668, 0.0273, 0.0307],
[ 0.0311, 0.0798, 0.0596, -0.0256, -0.0979],
[-0.0542, -0.0553, -0.0613, -0.0782, -0.0579]],
[[-0.0875, -0.0210, 0.0966, -0.0768, -0.0592],
[ 0.0897, 0.0828, 0.0343, 0.0133, -0.0669],
[-0.0986, -0.0866, -0.1075, 0.0613, 0.0486],
[-0.1153, 0.1053, 0.0163, -0.0813, -0.0268],
[-0.0267, 0.1103, -0.0289, -0.0212, -0.0427]],
[[-0.0471, -0.0681, 0.0008, 0.0685, -0.0726],
[-0.0161, -0.0378, 0.0142, -0.0448, -0.0542],
[-0.0594, -0.1057, -0.0388, 0.0282, -0.0045],
[-0.0942, 0.1069, -0.0601, -0.1092, 0.0658],
[ 0.0058, -0.0673, -0.0139, 0.0544, 0.0266]]]])
conv1.bias
tensor([-0.0518, 0.0757, -0.0106, 0.0676, -0.0081, 0.0287])
conv2.weight
tensor([[[[ 3.7345e-02, 3.6917e-02, 7.1036e-04, -1.8674e-02, -2.0425e-02],
[ 6.0888e-02, 3.9115e-02, 6.0703e-02, -4.9426e-02, 2.0759e-03],
[ 4.4816e-02, 2.4605e-02, 7.5611e-02, 7.0319e-02, -4.8377e-03],
[-5.8067e-02, 2.4481e-02, -7.9937e-02, -6.9166e-02, 4.6737e-02],
[-1.4972e-02, -7.3878e-02, -7.6790e-02, -1.9032e-02, -1.9925e-02]],
[[ 2.2720e-02, -2.5971e-02, -6.3477e-02, 1.7530e-02, 2.5905e-02],
[-6.4675e-02, -6.6284e-02, -8.0954e-02, 3.6396e-02, -6.3034e-02],
[-1.8645e-03, 4.3630e-02, -2.6132e-02, -7.6036e-02, 7.8512e-02],
[ 4.3217e-02, -3.1471e-02, -9.8552e-03, 3.9861e-02, -2.3391e-02],
[ 2.8649e-02, 5.9830e-02, 2.7637e-02, 1.8659e-02, -5.6407e-02]],
[[-4.5381e-02, 5.4505e-03, 4.0909e-03, 5.6494e-02, -5.9140e-02],
[ 2.3393e-02, -3.3347e-02, -6.6707e-02, 1.9643e-02, -2.0795e-02],
[-7.7018e-02, 3.0887e-02, -1.8568e-02, 6.2216e-02, 1.9621e-02],
[ 3.4469e-02, -8.1499e-03, 3.7170e-03, -4.2050e-02, -1.1584e-02],
[ 1.5811e-02, 7.1654e-02, -6.3917e-02, 2.9590e-02, 2.9235e-02]],
[[ 3.9611e-02, -2.3142e-02, 3.7166e-02, -3.2922e-02, -7.0504e-02],
[-5.1643e-03, 7.2051e-02, 4.7072e-02, 4.3785e-02, 3.0908e-02],
[ 4.5853e-02, -5.6516e-02, -7.1385e-02, 4.2695e-02, 2.6823e-02],
[-3.3464e-03, -5.1895e-02, -5.9452e-02, -1.4120e-02, -4.2594e-02],
[-5.9216e-03, -7.9317e-02, 3.2249e-02, 2.3419e-02, 1.3252e-02]],
[[ 1.1583e-02, 7.0880e-02, 1.9811e-02, 8.0881e-02, -4.4006e-02],
[ 3.5248e-02, -2.2885e-02, 2.7275e-02, 4.8847e-02, -2.0026e-02],
[ 3.0439e-02, 7.2296e-02, 7.4953e-02, 4.2624e-02, 3.7546e-02],
[ 6.0431e-02, -7.8233e-02, -2.1510e-02, 7.9771e-02, 2.0746e-02],
[-6.6449e-02, 8.1094e-02, 1.7321e-03, 9.2274e-03, 4.6537e-02]],
[[-4.3946e-02, 3.4462e-02, 4.2152e-02, -3.2618e-02, -3.1919e-02],
[ 3.7446e-02, -1.6665e-02, -4.6682e-02, -4.2671e-02, 3.8549e-02],
[ 2.0830e-02, 3.1240e-02, 1.4524e-02, 6.9791e-02, 6.4641e-02],
[-5.1961e-02, 2.9224e-02, -8.1169e-02, 5.0892e-02, 7.3306e-02],
[-5.3219e-02, 4.4223e-02, 1.6751e-02, 1.0149e-02, 2.7877e-02]]],
[[[ 4.0394e-02, -7.5706e-02, 4.0839e-02, 4.1527e-02, 7.5264e-02],
[ 5.0450e-02, 2.5265e-02, 5.6446e-02, -9.1349e-03, -2.0079e-02],
[ 4.0942e-03, 1.9982e-02, 7.8088e-02, 1.3781e-02, 5.9151e-02],
[ 1.5582e-03, -7.1641e-02, -3.9163e-02, 2.0775e-02, 6.9971e-04],
[ 1.4160e-02, -2.1367e-02, 7.2005e-02, 3.0974e-02, 5.5173e-02]],
[[ 2.6252e-02, 2.1418e-02, -7.9839e-03, -6.4778e-02, -1.7517e-02],
[ 3.0320e-02, -7.2231e-03, -6.9592e-02, -2.1982e-02, -7.9040e-02],
[-6.8035e-02, -4.2568e-02, 6.6144e-03, 3.2699e-02, 4.0103e-03],
[-7.6126e-02, 6.6878e-02, 2.0500e-02, 7.3015e-02, 4.4364e-02],
[ 6.4838e-02, -6.1012e-02, -3.7262e-02, -2.4985e-02, 4.3699e-02]],
[[ 2.9414e-02, -6.4046e-02, -7.9652e-02, 5.7636e-02, 1.8768e-02],
[ 1.3999e-02, 6.1029e-04, -2.5749e-02, 5.8838e-02, -7.4751e-02],
[ 6.9579e-02, 4.0549e-02, 2.6570e-02, 1.9762e-02, 6.2969e-02],
[-3.0153e-02, 4.1174e-02, -1.4508e-02, -7.5177e-02, 6.0151e-02],
[ 3.5434e-02, -3.4538e-02, 9.2284e-04, -6.9311e-02, 5.5276e-02]],
[[-1.1414e-02, 5.8701e-02, -8.5818e-03, 5.3662e-02, -6.6120e-02],
[-6.6017e-02, -7.2396e-02, 3.9557e-02, -3.2917e-02, 2.3871e-02],
[ 3.6651e-02, -5.7215e-02, 7.5414e-02, -2.8045e-02, -5.4822e-03],
[ 6.6979e-03, 4.3628e-02, -7.9500e-02, 1.6581e-02, -2.0443e-03],
[ 4.8128e-02, -2.5281e-03, 6.3143e-02, 5.5622e-02, 5.0823e-02]],
[[ 2.6724e-02, -4.4536e-02, 3.8301e-02, -1.6428e-02, 7.6386e-03],
[ 6.9414e-02, 6.4933e-02, -6.0955e-02, 4.2710e-02, 6.7609e-02],
[ 3.2155e-02, -3.9059e-02, -8.1267e-02, 6.0943e-02, 7.8761e-02],
[ 9.2058e-03, -6.8716e-02, 5.9567e-02, -4.7019e-02, 4.3247e-02],
[-8.0727e-02, -7.7734e-02, 6.4857e-02, 1.4899e-03, 4.7743e-02]],
[[ 4.5334e-02, 1.9591e-02, 4.8368e-02, -7.7771e-02, -2.2575e-02],
[ 2.3952e-02, -7.0499e-02, -6.3645e-02, -1.8650e-02, -2.2908e-02],
[ 4.0042e-02, 5.1475e-02, -4.9496e-02, -7.1370e-02, -7.3195e-02],
[ 7.5049e-02, -5.0792e-02, 7.1668e-02, 2.3936e-02, 7.4133e-02],
[ 7.1541e-03, 3.1869e-02, -7.6662e-02, -1.1101e-02, -6.0137e-02]]],
[[[-3.1704e-02, -6.3101e-02, 5.3376e-02, -7.5087e-02, 7.4585e-02],
[ 3.3090e-03, -1.2829e-02, 4.2248e-03, -3.8829e-02, 2.2651e-02],
[-3.0851e-02, 6.0922e-02, 6.3309e-02, -1.0301e-02, -3.4664e-04],
[ 3.7358e-03, 5.7330e-02, 7.6840e-02, 4.9586e-02, -5.9538e-02],
[ 3.8326e-02, -6.2273e-02, 3.8199e-02, 9.8926e-03, -6.3158e-02]],
[[ 3.4346e-02, -3.0442e-02, 9.5122e-03, 5.9389e-03, 6.4301e-02],
[ 3.7045e-02, -1.4779e-02, 5.3899e-02, 1.2880e-02, 3.2471e-02],
[-4.9418e-02, -7.3648e-02, -1.2379e-02, 6.0696e-02, -4.2189e-03],
[-5.9245e-03, -2.1985e-02, 3.4781e-02, 6.4471e-02, -6.9246e-02],
[-5.3104e-02, 4.7656e-02, -4.7472e-02, 5.0731e-02, 6.2965e-02]],
[[-7.3520e-02, -2.9047e-02, 2.4189e-02, 9.0452e-03, 3.1695e-02],
[-5.4734e-02, 2.8641e-02, -6.1692e-02, -2.3195e-02, 1.1079e-02],
[ 5.4220e-02, -7.6615e-02, 5.6999e-02, 2.3648e-03, -4.8024e-02],
[ 2.4688e-02, 2.6944e-02, -5.6788e-02, 4.2183e-02, -2.4088e-02],
[ 5.4214e-03, -1.7575e-02, 6.7586e-02, 5.1225e-02, 2.3619e-02]],
[[-2.5996e-03, -7.9755e-02, 6.8051e-02, 7.2919e-02, 7.9731e-02],
[-5.1041e-02, -2.0511e-02, 4.2875e-02, -3.9107e-02, 7.6118e-02],
[ 9.0300e-03, 3.5622e-02, 1.0749e-02, -1.6854e-02, -5.9495e-02],
[ 3.1847e-02, 7.7569e-02, 3.8437e-02, -3.2095e-02, 8.4368e-03],
[ 2.2344e-02, 5.6201e-02, 4.6436e-02, -6.0905e-02, -6.2908e-02]],
[[ 1.1728e-02, -2.6272e-03, -3.3887e-02, -4.3492e-02, 1.4497e-02],
[ 3.9038e-02, -6.3376e-02, -2.3013e-02, -4.0424e-02, 1.4657e-02],
[ 4.9225e-03, -3.3368e-02, 1.8460e-02, -3.6485e-02, -6.4865e-02],
[-6.4731e-05, -6.8108e-02, -1.3410e-02, -3.4235e-02, -6.4206e-02],
[-1.7974e-02, 3.3237e-02, 5.5293e-02, -3.7493e-02, 2.4666e-02]],
[[ 6.7512e-02, -8.9689e-03, -2.2933e-02, 1.6010e-02, 6.9638e-02],
[-3.7105e-02, 6.7179e-02, -1.0214e-02, -4.8984e-02, -4.1328e-02],
[-8.6841e-03, -1.7692e-02, -1.4424e-02, 7.3771e-02, 1.2215e-02],
[-7.9845e-02, 6.7047e-02, 3.1997e-02, -5.1304e-02, 6.0482e-02],
[-1.4893e-02, -2.7676e-02, 5.4934e-02, 6.3990e-02, 2.7662e-02]]],
...,
[[[ 1.0199e-02, -5.2322e-02, -3.7206e-02, -1.4544e-02, -4.6219e-02],
[-1.2879e-02, 4.7828e-02, 7.8953e-02, -1.3771e-02, 2.3383e-02],
[-2.0076e-02, -2.9035e-02, 5.4802e-02, 3.0305e-02, -2.6655e-02],
[ 4.6331e-02, 6.4619e-02, 3.6669e-02, -5.7059e-02, -1.2394e-02],
[-4.7441e-02, 3.9712e-02, -6.8904e-02, -7.4229e-02, -5.4093e-03]],
[[-3.7258e-02, 8.1076e-02, -7.6125e-02, -2.9282e-02, 1.0107e-02],
[ 2.3341e-02, 2.7966e-02, 4.9232e-02, 4.2246e-02, -2.2956e-02],
[ 6.1198e-02, 5.5776e-02, -6.1231e-02, -7.3973e-02, -1.8083e-02],
[ 2.4045e-02, 5.3863e-02, 3.1777e-02, -3.6017e-02, 7.9472e-02],
[-6.9948e-02, 6.3699e-02, 4.3250e-02, 4.6033e-02, 8.9336e-04]],
[[ 2.5100e-02, 5.6652e-03, -3.4297e-02, -4.4208e-02, -3.4073e-03],
[ 1.6430e-02, -5.2721e-02, 3.6158e-02, 7.7301e-02, 4.7632e-03],
[ 7.6485e-02, -4.3372e-03, 4.2569e-02, -3.6177e-02, -2.1329e-03],
[-8.4518e-03, -9.9161e-03, -2.1800e-02, -6.9410e-02, 5.5372e-02],
[ 7.8497e-02, 6.9532e-02, -1.9760e-03, -4.9298e-02, -3.0007e-02]],
[[ 5.7926e-02, -1.7255e-02, 5.0323e-02, 7.3531e-03, 7.3418e-02],
[ 5.8616e-02, 3.4145e-02, 6.8296e-02, -8.6052e-03, -1.9639e-02],
[-7.3903e-03, 3.6604e-02, -6.6443e-02, 4.2403e-02, -1.6865e-02],
[-4.4089e-02, -3.2403e-02, -3.5313e-02, 6.2766e-02, 4.4323e-02],
[-7.7840e-02, -5.1307e-02, 3.1470e-02, 2.4083e-02, -3.1171e-02]],
[[-7.2114e-02, 4.7854e-02, -3.7292e-02, 3.6088e-02, -1.2283e-02],
[ 2.4217e-02, 4.2792e-02, 4.0946e-02, -4.8699e-02, -2.0729e-02],
[ 4.8880e-02, -8.1478e-02, -7.3582e-02, -2.0240e-02, 5.9753e-02],
[ 3.2898e-02, -1.7339e-02, -5.6911e-02, 6.5161e-02, 1.7858e-02],
[ 3.9480e-02, -3.2163e-02, 3.1307e-02, -5.5870e-02, -1.6457e-02]],
[[ 2.2531e-02, -8.0621e-02, 6.8040e-02, -5.3297e-02, 6.8076e-02],
[-2.8019e-02, -4.2122e-02, 9.4667e-03, 1.1930e-02, 5.3379e-02],
[ 7.0529e-03, -7.5513e-02, -5.8059e-03, -2.2628e-02, -3.1337e-02],
[ 4.6129e-02, 1.0021e-02, 1.3801e-02, 5.4890e-02, 8.2099e-03],
[-3.4219e-02, -3.5771e-02, 4.9384e-02, 4.2971e-03, 1.0185e-02]]],
[[[ 1.5722e-02, -1.1761e-02, -5.5889e-02, -4.0064e-02, 4.6955e-02],
[ 3.8323e-02, 1.4971e-02, 4.7523e-02, -7.3487e-02, -6.6887e-02],
[ 5.4269e-02, 6.5048e-03, -2.5124e-02, 7.0294e-02, 2.8026e-02],
[-7.3427e-02, -6.3417e-02, -6.9835e-02, 5.1313e-02, -3.0339e-02],
[-3.7274e-02, -6.8501e-02, 1.3327e-02, 7.0011e-03, 7.3245e-02]],
[[ 3.7582e-02, 8.0081e-02, -4.8837e-02, 7.3556e-03, 6.0067e-02],
[-3.2691e-02, -4.5259e-02, -6.4128e-02, -5.0167e-02, -2.3580e-02],
[-5.0516e-02, -7.3288e-02, -5.1057e-02, -4.8879e-02, 1.4947e-02],
[ 5.5937e-02, 2.8608e-02, -5.1257e-02, -9.8808e-03, 7.7368e-03],
[-6.1867e-03, 5.5593e-02, -6.4040e-02, -1.9817e-02, 4.8827e-02]],
[[ 7.0229e-02, 3.4918e-02, 3.4894e-02, -3.7108e-02, 5.1772e-02],
[-8.0578e-02, -1.9076e-02, -5.2325e-02, -3.1958e-02, -1.4594e-02],
[-6.5627e-02, -4.4570e-02, 7.8603e-02, -1.8659e-02, -4.3064e-02],
[-5.8323e-02, 2.8146e-02, 2.7508e-02, -2.0674e-03, -2.3774e-02],
[ 5.3830e-02, 3.1012e-03, -1.3121e-02, 4.2424e-02, 1.3136e-03]],
[[ 6.2128e-02, -8.1062e-02, 5.5346e-03, 4.9415e-02, 1.9914e-02],
[ 4.5311e-02, 4.5756e-02, -1.6908e-02, 1.4688e-02, 3.0094e-02],
[-4.5764e-03, -3.1909e-02, 6.0014e-02, -6.2276e-02, -3.4248e-02],
[ 3.9257e-02, 1.9695e-02, -8.7933e-03, 7.9642e-02, -3.3138e-02],
[-1.9051e-02, 4.8750e-02, -4.9173e-03, 7.9091e-02, 1.4798e-02]],
[[-5.5164e-02, 3.3881e-02, -3.4420e-04, 6.3908e-02, 7.2785e-02],
[ 3.9900e-02, 6.5517e-03, -8.0552e-02, -1.5329e-02, -5.7844e-02],
[-6.8076e-03, 5.0193e-02, 7.4728e-02, -2.9026e-02, 2.0508e-02],
[ 6.2877e-02, 8.0455e-02, 5.1410e-02, -3.8253e-02, -6.9830e-02],
[-1.4001e-02, -6.4940e-02, 4.3407e-03, -2.0574e-03, -6.3191e-02]],
[[ 1.8252e-02, 8.8192e-03, 1.6308e-02, 8.0085e-02, 5.4516e-02],
[ 3.2235e-02, -1.4626e-02, -5.4716e-02, 7.6015e-02, -5.5979e-02],
[ 3.0786e-02, 3.1961e-02, -1.8162e-02, 6.9846e-02, 3.8360e-02],
[ 7.4391e-02, 1.7810e-02, 5.8638e-02, -2.3435e-02, -5.3262e-02],
[ 7.7326e-02, -2.3349e-02, -7.6374e-02, -1.5019e-02, -4.1779e-02]]],
[[[-4.0523e-02, -1.3673e-02, 1.7464e-02, -8.0872e-02, 1.3032e-02],
[-5.7647e-02, 7.6154e-02, 2.9756e-02, 6.1005e-02, -6.2796e-02],
[-1.8865e-02, -3.1772e-02, -3.0434e-02, -6.4698e-02, -7.2306e-02],
[-1.3506e-02, 7.4316e-02, 3.0436e-02, -1.4313e-02, 7.2392e-02],
[-3.5933e-02, 1.1374e-02, -1.8395e-02, -3.0843e-02, 5.5321e-02]],
[[ 1.4394e-02, 3.6295e-02, 4.9757e-02, -4.6856e-03, 9.4505e-03],
[-5.9689e-02, 5.7720e-02, 7.4232e-02, 7.1351e-02, 3.7225e-02],
[ 3.1824e-02, -3.5381e-02, -3.4969e-02, 8.2402e-03, -2.4470e-02],
[-7.7093e-02, 4.4498e-02, 4.7004e-02, -7.2011e-02, -3.8237e-02],
[-6.7163e-02, -1.7111e-02, 5.9214e-02, 7.7507e-02, -4.4913e-02]],
[[-8.0340e-02, 1.9304e-02, 1.0000e-03, 4.0900e-03, -6.6380e-02],
[-6.6810e-02, -6.3729e-03, 2.0123e-02, -4.3774e-02, -3.4981e-02],
[-6.8541e-02, 1.6689e-02, -7.1792e-02, 1.5246e-02, -6.8854e-04],
[ 4.3843e-02, 2.6870e-02, -3.8101e-04, -7.8826e-02, 6.5015e-02],
[-2.0440e-02, 7.9386e-02, -3.8219e-03, -2.4580e-02, 4.0634e-02]],
[[ 5.5544e-02, -3.7836e-02, -7.9338e-02, 7.0619e-02, -3.3446e-02],
[-3.2902e-03, -4.8389e-02, 4.7453e-02, 5.8127e-02, -4.8315e-03],
[-5.6379e-02, -5.8292e-02, -3.9011e-02, -4.0794e-03, 4.6537e-02],
[ 8.0529e-02, 6.2578e-02, 2.7042e-02, -7.3989e-02, 8.9902e-03],
[-3.2401e-02, -6.5050e-02, 6.9688e-02, -6.6803e-02, 3.6991e-02]],
[[-2.2901e-02, 4.9715e-02, -5.3222e-02, -7.6049e-02, -6.6969e-02],
[ 9.8178e-03, -6.9489e-02, -3.7907e-02, 3.8849e-02, -3.6555e-02],
[-4.1870e-02, 1.7604e-02, 1.3747e-02, 1.2642e-02, -2.0011e-02],
[-6.3959e-02, 6.5200e-02, -3.1475e-02, -5.4463e-02, -7.0515e-02],
[-3.3264e-02, 5.0797e-04, 4.7155e-02, 5.2026e-02, -6.2295e-02]],
[[-7.0623e-02, -3.8786e-02, -1.6983e-02, -4.0680e-02, -3.5289e-03],
[ 6.6456e-02, -5.6733e-02, -3.2044e-02, -8.0967e-02, 6.8003e-02],
[ 7.7967e-02, -1.8929e-02, 4.1282e-02, -4.3998e-02, 7.8460e-02],
[ 4.9829e-02, -3.3562e-02, -7.5281e-02, 5.1357e-02, 2.8261e-02],
[ 6.7955e-02, -3.4526e-02, -4.7810e-02, 7.8384e-02, -1.2523e-02]]]])
conv2.bias
tensor([-0.0807, -0.0406, -0.0171, 0.0340, 0.0777, 0.0155, 0.0788, -0.0089,
-0.0693, -0.0276, 0.0578, 0.0165, 0.0525, -0.0673, -0.0171, 0.0804])
fc1.weight
tensor([[-0.0032, -0.0052, -0.0235, ..., -0.0029, -0.0208, 0.0015],
[-0.0460, 0.0396, 0.0050, ..., 0.0194, 0.0410, 0.0174],
[-0.0391, -0.0378, -0.0265, ..., 0.0376, 0.0157, 0.0217],
...,
[-0.0304, -0.0122, 0.0140, ..., -0.0028, 0.0002, -0.0292],
[ 0.0346, 0.0187, 0.0185, ..., -0.0256, 0.0434, -0.0359],
[-0.0116, 0.0302, 0.0412, ..., 0.0163, -0.0479, -0.0163]])
fc1.bias
tensor([-0.0453, 0.0006, 0.0223, 0.0062, -0.0188, 0.0133, 0.0158, -0.0061,
0.0145, 0.0035, -0.0145, -0.0451, 0.0352, 0.0302, -0.0086, 0.0333,
0.0025, 0.0102, 0.0128, -0.0002, -0.0021, 0.0439, 0.0468, 0.0120,
-0.0089, 0.0115, -0.0260, 0.0178, 0.0217, 0.0456, -0.0116, 0.0224,
-0.0207, -0.0060, -0.0386, 0.0405, -0.0180, -0.0116, -0.0009, 0.0239,
-0.0203, -0.0064, 0.0406, -0.0173, 0.0378, 0.0180, 0.0195, 0.0145,
-0.0435, 0.0157, 0.0363, -0.0170, 0.0064, -0.0107, 0.0341, -0.0227,
0.0179, -0.0221, -0.0264, -0.0018, -0.0104, -0.0481, -0.0343, -0.0245,
-0.0190, -0.0425, 0.0329, -0.0488, 0.0279, 0.0034, -0.0491, -0.0422,
0.0237, 0.0211, -0.0260, -0.0380, 0.0010, -0.0025, -0.0367, 0.0386,
-0.0090, 0.0157, -0.0292, -0.0302, 0.0423, -0.0476, -0.0401, 0.0195,
0.0315, 0.0285, 0.0456, 0.0297, 0.0091, -0.0291, 0.0301, -0.0231,
-0.0055, 0.0359, 0.0164, -0.0297, 0.0243, -0.0429, 0.0311, 0.0072,
-0.0400, -0.0473, 0.0470, -0.0068, -0.0160, 0.0092, -0.0327, -0.0040,
0.0245, 0.0178, -0.0249, -0.0146, 0.0401, 0.0019, 0.0216, -0.0497])
fc2.weight
tensor([[ 0.0327, -0.0668, 0.0384, ..., -0.0704, 0.0015, -0.0560],
[-0.0657, -0.0632, 0.0364, ..., 0.0721, 0.0100, 0.0768],
[ 0.0130, 0.0838, -0.0002, ..., 0.0178, 0.0226, -0.0903],
...,
[ 0.0313, -0.0549, 0.0724, ..., 0.0874, -0.0439, -0.0327],
[ 0.0371, -0.0318, 0.0875, ..., 0.0631, -0.0813, 0.0321],
[-0.0071, 0.0583, 0.0542, ..., -0.0285, 0.0750, 0.0572]])
fc2.bias
tensor([ 0.0134, -0.0475, 0.0059, 0.0127, 0.0480, -0.0361, 0.0095, 0.0558,
0.0322, 0.0769, 0.0570, -0.0611, -0.0341, -0.0588, -0.0095, 0.0799,
-0.0563, 0.0562, -0.0833, -0.0296, 0.0359, -0.0193, -0.0529, -0.0428,
-0.0597, -0.0521, 0.0202, -0.0128, -0.0195, 0.0684, -0.0416, 0.0795,
0.0608, 0.0296, -0.0304, -0.0381, -0.0834, 0.0112, 0.0332, 0.0466,
-0.0837, -0.0375, 0.0573, -0.0380, 0.0802, -0.0668, 0.0514, -0.0070,
0.0033, 0.0871, -0.0178, 0.0522, -0.0571, 0.0472, 0.0561, 0.0215,
0.0012, -0.0419, -0.0055, -0.0174, -0.0204, -0.0619, -0.0016, -0.0784,
-0.0819, -0.0562, -0.0807, 0.0765, 0.0316, 0.0423, -0.0063, -0.0210,
0.0315, 0.0806, 0.0109, -0.0175, 0.0705, -0.0466, 0.0097, 0.0098,
0.0798, 0.0796, -0.0783, -0.0730])
fc3.weight
tensor([[-0.0351, 0.0782, 0.0364, -0.0547, 0.0099, -0.0188, -0.0582, -0.0056,
-0.1074, 0.0905, -0.0629, 0.1084, -0.0456, -0.0382, 0.0472, -0.0674,
-0.0414, -0.0523, -0.0009, -0.0946, -0.0463, 0.0898, 0.0631, -0.0910,
0.0091, -0.0383, -0.0311, 0.0978, 0.0204, -0.1040, -0.0793, -0.1019,
0.0631, -0.0136, -0.0283, -0.0260, -0.1050, 0.0029, -0.1016, -0.0253,
-0.0350, 0.0862, 0.0013, 0.0482, -0.0284, 0.0073, 0.0053, 0.0703,
0.0078, -0.0487, 0.0899, -0.0934, 0.0385, -0.0764, 0.0070, -0.0486,
0.0197, 0.0372, -0.0763, 0.0477, -0.1010, -0.0224, -0.0491, -0.0548,
-0.0461, 0.0629, 0.0691, 0.0064, -0.0637, -0.0727, 0.0074, -0.0961,
-0.0429, 0.0362, 0.0140, 0.0659, -0.0640, 0.0336, 0.1020, -0.0572,
0.0779, 0.0741, -0.0588, -0.0856],
[ 0.0477, -0.0715, 0.0228, 0.0505, 0.0311, -0.0720, 0.0702, 0.0573,
0.0241, 0.0151, -0.1036, -0.0023, 0.0145, 0.0096, -0.0538, 0.0805,
0.0951, 0.1079, -0.0928, 0.0317, -0.0034, -0.0983, 0.0781, 0.0109,
-0.0001, -0.0881, 0.0208, 0.0635, 0.0434, 0.1080, -0.0625, -0.0924,
0.0947, 0.0438, 0.0596, -0.0324, -0.0555, -0.0729, 0.0614, -0.0416,
-0.0770, -0.0748, -0.0193, -0.0991, -0.0900, 0.0201, 0.0715, -0.0605,
0.0033, -0.0418, 0.0473, -0.1069, 0.0095, -0.0540, 0.0695, 0.1042,
0.0245, 0.0715, 0.1073, 0.0643, -0.0956, -0.0843, 0.0391, 0.0033,
0.0787, -0.1019, -0.0056, -0.0830, 0.0327, 0.0031, -0.0049, -0.0971,
0.0549, -0.0827, -0.0933, -0.0676, -0.0909, -0.0925, 0.0300, 0.0154,
0.0706, -0.0158, -0.0479, -0.0853],
[-0.0502, -0.0625, 0.0947, -0.0247, 0.0150, -0.0101, 0.0695, 0.0064,
0.0390, -0.0735, -0.0347, 0.0255, -0.0964, 0.0016, -0.0568, 0.1050,
-0.0972, 0.0176, 0.0756, -0.0781, 0.0529, -0.0751, 0.0823, 0.0953,
0.0651, -0.0740, -0.0460, 0.0549, -0.0199, -0.0245, -0.1058, 0.0528,
-0.1070, 0.0086, 0.0665, -0.0895, 0.0798, -0.0902, 0.1072, 0.0173,
0.0257, 0.0481, -0.0289, 0.0291, -0.1057, -0.0429, -0.1033, -0.0716,
-0.0690, 0.0628, 0.0718, -0.1040, -0.1042, 0.0831, 0.0844, 0.1073,
0.0846, 0.0477, 0.0619, 0.0479, -0.0872, -0.0647, 0.0556, 0.0539,
-0.0421, -0.0697, 0.0625, 0.0688, 0.0778, 0.0239, -0.0431, 0.0142,
0.0107, 0.0207, 0.0443, -0.0164, 0.0764, -0.0577, -0.0278, -0.0256,
0.0416, -0.0133, -0.0185, 0.0236],
[ 0.1058, 0.0798, 0.0002, 0.0795, 0.1071, -0.0654, 0.0125, 0.0800,
-0.0396, -0.0300, 0.0885, 0.0236, -0.0025, -0.0938, -0.0494, 0.0230,
0.0239, -0.0673, -0.0466, 0.0425, 0.0175, 0.0395, 0.1038, 0.0651,
-0.0052, 0.0480, -0.0311, -0.0081, 0.0582, 0.0096, 0.0950, 0.0721,
-0.0648, 0.0789, -0.0936, 0.0235, -0.0654, -0.0109, 0.0745, 0.0503,
0.0175, 0.0719, 0.1069, -0.0279, -0.0692, -0.0972, 0.0624, -0.0348,
0.0703, 0.0675, 0.0817, -0.0586, 0.0760, -0.0307, -0.0205, -0.0042,
0.0939, 0.0528, -0.0242, -0.0632, -0.0184, -0.0781, 0.0133, 0.0246,
0.0962, -0.0454, 0.0117, -0.0783, 0.0378, 0.0393, 0.0874, 0.0969,
-0.1045, 0.0167, -0.0041, -0.0450, -0.0738, -0.0291, -0.0480, 0.1086,
0.0284, -0.0981, 0.0933, 0.0159],
[-0.1059, 0.0177, -0.0834, 0.0280, -0.0464, 0.0020, 0.0380, 0.1060,
0.0865, -0.0857, -0.0820, -0.0471, 0.0236, -0.0603, 0.0667, 0.0142,
0.0905, -0.0847, -0.0254, -0.0391, -0.0173, -0.1018, -0.0318, 0.0223,
-0.0592, 0.0776, 0.0137, -0.0570, -0.0139, 0.0646, -0.0869, -0.0624,
-0.0882, 0.0348, 0.0161, -0.0336, 0.0755, 0.1023, -0.1000, -0.0267,
-0.1080, 0.0422, -0.0045, 0.0512, -0.0861, -0.0749, 0.1039, -0.0755,
0.0404, 0.0974, -0.0386, -0.0557, 0.0769, -0.0065, 0.0711, 0.0475,
0.0997, 0.0045, 0.0312, -0.0187, -0.1067, -0.0217, 0.0200, -0.0657,
0.0338, 0.0409, 0.0502, 0.0765, -0.0575, 0.0739, -0.0023, -0.0798,
0.0923, 0.0901, -0.0271, 0.0934, 0.0941, -0.0731, 0.0918, 0.0247,
-0.0655, -0.0685, -0.0875, 0.0479],
[-0.0871, 0.0840, -0.0604, -0.0193, -0.0877, -0.0734, 0.0127, -0.0357,
0.0891, -0.0521, -0.0469, -0.0877, -0.0775, 0.0123, -0.0194, -0.1078,
0.0492, -0.1044, -0.0714, 0.0606, 0.0053, -0.0227, 0.0231, -0.0439,
-0.0554, 0.0112, -0.0717, 0.0829, -0.0140, 0.0908, -0.0258, 0.0669,
-0.0290, -0.0555, 0.1082, -0.0481, -0.0175, -0.0107, -0.0297, -0.0472,
-0.0606, -0.0301, -0.0846, -0.0839, 0.0077, 0.0267, 0.0926, -0.0119,
-0.0195, 0.0077, 0.0336, 0.0981, 0.0040, -0.0447, -0.0131, -0.1037,
-0.1037, 0.0380, 0.0343, 0.1038, 0.0378, -0.0402, -0.0581, -0.0523,
0.0604, 0.0402, -0.0838, -0.0718, -0.0488, 0.0242, 0.0727, -0.0429,
0.0225, -0.0209, 0.0964, 0.0810, -0.0704, 0.0720, -0.0120, 0.0704,
-0.0779, 0.0169, -0.0885, 0.0133],
[ 0.0809, -0.0432, -0.0358, 0.0023, -0.0809, -0.0058, -0.0977, 0.0731,
0.0576, -0.0883, 0.0758, 0.0168, -0.0107, -0.0284, 0.0640, 0.0921,
-0.1068, -0.0934, 0.0853, 0.0940, -0.0152, 0.0899, 0.0431, 0.0047,
-0.0164, -0.0339, 0.0470, 0.0021, -0.0960, 0.1058, 0.0646, -0.0491,
-0.1037, 0.0242, 0.0083, 0.0221, -0.0810, -0.0367, -0.0478, -0.0849,
-0.0836, 0.0972, -0.0792, -0.1085, -0.0072, -0.0512, -0.0573, -0.0067,
0.0944, 0.0517, 0.0925, 0.0974, 0.0239, 0.1011, -0.1045, -0.0396,
0.0565, -0.0717, -0.0250, -0.0896, 0.0717, 0.0155, -0.0818, -0.0084,
-0.0022, -0.1046, -0.0607, -0.0657, -0.0968, 0.0873, -0.0598, -0.0593,
0.0558, -0.1064, 0.0574, 0.1044, -0.0233, -0.0505, 0.1041, 0.1085,
-0.0538, -0.0595, 0.0956, 0.0604],
[ 0.1015, -0.0722, 0.0488, 0.0101, 0.0202, -0.0593, 0.0623, -0.0953,
0.0665, 0.1086, -0.1021, 0.0690, 0.0366, -0.0012, 0.0754, 0.0651,
0.0005, 0.0642, 0.0149, 0.0153, -0.0449, -0.0533, -0.0103, -0.0387,
-0.0602, 0.1004, 0.0283, 0.0476, 0.0596, 0.0504, 0.0366, -0.0626,
-0.0361, -0.0263, -0.1079, 0.0652, -0.1043, 0.0658, -0.1018, 0.0300,
-0.0766, -0.0371, -0.0634, -0.0091, -0.0625, 0.0402, -0.0438, -0.0524,
-0.0687, 0.0636, 0.0641, 0.0817, -0.1090, 0.0409, -0.0558, 0.1062,
0.0414, -0.0629, 0.0131, -0.0563, 0.0378, 0.0774, -0.0750, 0.0777,
0.0578, -0.0936, 0.0269, -0.0396, -0.0964, -0.0894, -0.0125, -0.0236,
0.0014, -0.0819, -0.0124, 0.0247, 0.1023, -0.0884, 0.0624, -0.0268,
0.0912, 0.0312, -0.0989, -0.0456],
[-0.0516, -0.0043, -0.0102, 0.0299, 0.0665, -0.0009, 0.1026, -0.0169,
0.0006, 0.0715, 0.1036, 0.0087, 0.0707, -0.0396, -0.0476, -0.0896,
0.0608, -0.0180, -0.0367, -0.0653, 0.0630, 0.1080, 0.0807, -0.0254,
-0.0631, -0.0055, 0.0611, 0.0971, 0.0037, 0.0689, 0.0861, 0.0770,
0.0932, 0.0045, -0.0973, -0.0262, -0.0513, -0.0835, -0.0036, 0.0355,
-0.0524, -0.0042, 0.0043, 0.0252, 0.0437, -0.0898, -0.0324, 0.0382,
0.0322, 0.0678, 0.0735, 0.0450, -0.0094, -0.0311, 0.0541, -0.0942,
0.0945, 0.0062, 0.0225, 0.0156, -0.0396, 0.0228, 0.0718, -0.0720,
-0.1054, 0.0230, 0.0869, 0.0996, 0.0456, 0.0548, 0.0861, 0.0787,
0.0159, 0.0857, -0.0790, -0.0252, 0.0639, 0.0059, 0.0513, 0.0966,
-0.0251, -0.0521, -0.0341, -0.0758],
[-0.0203, -0.0479, -0.0209, -0.0015, -0.0537, -0.0816, -0.0397, 0.0425,
-0.0706, -0.0772, 0.0870, 0.0587, -0.0714, 0.0805, -0.0104, 0.0030,
-0.0112, -0.0341, 0.0635, -0.0202, 0.0839, -0.0273, 0.0969, -0.1013,
-0.0227, 0.0159, -0.0036, 0.0428, -0.0474, 0.0933, -0.0499, 0.0124,
-0.0043, 0.0435, 0.0785, -0.0419, 0.0231, 0.1041, -0.0128, -0.0573,
0.0456, 0.0235, 0.0565, -0.0868, -0.1073, -0.0534, 0.0074, -0.0648,
-0.0178, -0.0929, 0.1065, 0.0231, -0.0270, -0.0733, 0.0656, 0.0857,
-0.0941, 0.0018, -0.0426, -0.0115, 0.0978, -0.0780, -0.0622, 0.0838,
-0.0606, -0.0661, -0.0331, 0.0457, -0.0763, -0.0903, -0.0381, -0.0198,
-0.0369, -0.0135, -0.1033, 0.1031, 0.0206, 0.0382, -0.1005, -0.0188,
-0.0711, -0.0649, 0.0517, -0.0060]])
fc3.bias
tensor([ 0.0853, -0.1027, -0.0893, -0.0900, 0.0732, 0.0718, -0.0705, 0.0435,
0.0994, 0.0521])
Process finished with exit code 0
