-
Notifications
You must be signed in to change notification settings - Fork 4
/
train_model.py
250 lines (213 loc) · 9.28 KB
/
train_model.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
# -*- coding: utf-8 -*-
import click
import logging
from pathlib import Path
from dotenv import find_dotenv, load_dotenv
import pandas as pd
import matplotlib.pyplot as plt
from datetime import datetime
import torch
import torch.nn as nn
from torch.utils.tensorboard import SummaryWriter
from src.data.make_dataloader import create_dataloader, show_oracle_character
import os
import sys
# Ignore warnings
import warnings
warnings.filterwarnings("ignore")
# plt.ion() # interactive mode
sys.path.append(os.path.dirname(os.path.realpath(__file__)))
print(os.path.dirname(os.path.realpath(__file__)))
def show_loader_batch(dataloader):
for i_batch, sample_batched in enumerate(dataloader):
print(i_batch, sample_batched['image'].size(),
sample_batched['label'].size())
plt.figure(figsize=(10, 10))
show_oracle_character(sample_batched)
plt.axis('off')
plt.ioff()
plt.show()
if i_batch == 0:
break
def test_loader_function(root_dir, csv_file):
training_loader, dataset = create_dataloader(csv_file,
root_dir,
batch_size=16,
rescale_size=45,
randomcrop_size=40)
validation_loader, dataset = create_dataloader(csv_file,
root_dir,
batch_size=16,
rescale_size=45,
randomcrop_size=40,
datatype='validation')
show_loader_batch(training_loader)
show_loader_batch(validation_loader)
# defining the model architecture
class Net(nn.Module):
def __init__(self, class_number):
super(Net, self).__init__()
self.cnn_layers = nn.Sequential(
# Defining a 2D convolution layer
nn.Conv2d(1, 32, kernel_size=5, stride=2, padding=2),
nn.BatchNorm2d(32),
nn.ReLU(inplace=True),
nn.MaxPool2d(kernel_size=2, stride=2), # a. half-of-size
# Defining another 2D convolution layer
nn.Conv2d(32, 64, kernel_size=3, stride=1, padding=1),
nn.BatchNorm2d(64),
nn.ReLU(inplace=True),
nn.MaxPool2d(kernel_size=2, stride=2), # a. half-of-size
)
self.linear_layers = nn.Sequential(
nn.Linear(64 * 5 * 5, class_number) # 793
)
# Defining the forward pass
def forward(self, x):
x = self.cnn_layers(x)
x = x.view(x.size(0), -1)
x = self.linear_layers(x)
return x
def run_main(root_dir, csv_file, label_name_file, output_dir):
# root_dir='./data/raw/image/'
# csv_file="./data/processed/image_name_label.csv"
# label_name_file="./data/processed/label_name.csv"
# output_dir="./models"
training_loader, dataset = create_dataloader(csv_file,
root_dir,
batch_size=16,
rescale_size=45,
randomcrop_size=40)
validation_loader, dataset = create_dataloader(csv_file,
root_dir,
batch_size=16,
rescale_size=45,
randomcrop_size=40,
datatype='validation')
# get total class number
label2name_frame = pd.read_csv(label_name_file)
class_number = len(label2name_frame)
# defining the model
model = Net(class_number)
# defining the loss function
loss_fn = nn.CrossEntropyLoss()
# defining the optimizer
optimizer = torch.optim.Adam(model.parameters(), lr=0.01)
# checking if GPU is available
if torch.cuda.is_available():
model = model.cuda()
timestamp = datetime.now().strftime('%Y%m%d_%H%M%S')
writer = SummaryWriter(output_dir + '/runs/obs_{}'.format(timestamp))
epoch_number = 0
EPOCHS = 200
best_vloss = 10.
for epoch in range(EPOCHS):
print('EPOCH {}:'.format(epoch_number + 1))
# Make sure gradient tracking is on, and do a pass over the data
model.train(True)
running_loss = 0.
last_loss = 0.
# Here, we use enumerate(training_loader) instead of
# iter(training_loader) so that we can track the batch
# index and do some intra-epoch reporting
for i, data in enumerate(training_loader):
# Every data instance is an input + label pair
inputs = data['image'].float()
labels = data['label'].long()
# inputs = Variable(inputs)
# Zero your gradients for every batch!
optimizer.zero_grad()
# Make predictions for this batch
outputs = model(inputs)
# Compute the loss and its gradients
loss = loss_fn(outputs, labels)
loss.backward()
# Adjust learning weights
optimizer.step()
# Gather data and report
running_loss += loss.item()
# 1602/16 = 100 reports on the loss for every 25 batches.
if i % 100 == 99:
last_loss = running_loss / 100 # loss per batch
print(' batch {} loss: {}'.format(i + 1, last_loss))
tb_x = epoch * len(training_loader) + i + 1
writer.add_scalar('Loss/train', last_loss, tb_x)
running_loss = 0.
writer.add_images('mage_batch', inputs, epoch)
avg_loss = last_loss
# We don't need gradients on to do reporting
model.train(False)
running_vloss = 0.0
for i, vdata in enumerate(validation_loader):
vinputs = vdata['image'].float()
vlabels = vdata['label'].long()
voutputs = model(vinputs)
vloss = loss_fn(voutputs, vlabels)
running_vloss += vloss
avg_vloss = running_vloss / (i + 1)
print('LOSS train {} valid {}'.format(avg_loss, avg_vloss))
# Log the running loss averaged per batch
# for both training and validation
writer.add_scalars('Training vs. Validation Loss', {
'Training': avg_loss,
'Validation': avg_vloss
}, epoch_number + 1)
writer.flush()
# Track best performance, and save the model's state
if avg_vloss < best_vloss:
best_vloss = avg_vloss
# model_path = 'model_{}_{}'.format(timestamp, epoch_number)
model_path = Path(output_dir) / 'model_best.pt'
torch.save(model.state_dict(), str(model_path))
writer.add_graph(model, vinputs)
writer.flush()
epoch_number += 1
# Check the Final best model performance
correct_count, all_count = 0, 0
for data in validation_loader:
images = data['image'].float()
labels = data['label'].long()
for i in range(len(labels)):
# img = images[i].view(1, 1, 28, 28)
img = images[i, :, :, :]
img = torch.unsqueeze(img, dim=0)
with torch.no_grad():
logps = model(img)
# ps = torch.exp(logps)
# probab = list(ps.cpu()[0])
pred_label = logps.argmax(1).item()
true_label = labels.cpu()[i]
if (true_label == pred_label):
correct_count += 1
all_count += 1
print("Number Of Images =", all_count)
print("\nModel Accuracy =", (correct_count / all_count))
return True
@click.command()
@click.argument('input_image_filepath', type=click.Path(exists=True))
@click.argument('input_label_filepath', type=click.Path())
@click.argument('input_label_name_filepath', type=click.Path())
@click.argument('output_model_filepath', type=click.Path())
def main(input_image_filepath, input_label_filepath, input_label_name_filepath,
output_model_filepath):
""" Runs data processing scripts to turn raw data from (../raw) into
cleaned data ready to be analyzed (saved in ../processed).
"""
logger = logging.getLogger(__name__)
logger.info('image_dir: {}'.format(input_image_filepath))
logger.info('image_labe_dir: {}'.format(input_label_filepath))
logger.info('image_label_name_dir: {}'.format(input_label_name_filepath))
logger.info('output_model_filepath: {}'.format(output_model_filepath))
# check the image
# test_loader_function(input_image_filepath, input_label_filepath) # PASS
run_main(input_image_filepath, input_label_filepath,
input_label_name_filepath, output_model_filepath)
if __name__ == '__main__':
log_fmt = '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
logging.basicConfig(level=logging.INFO, format=log_fmt)
# not used in this stub but often useful for finding various files
project_dir = Path(__file__).resolve().parents[2]
# find .env automagically by walking up directories until it's found, then
# load up the .env entries as environment variables
load_dotenv(find_dotenv())
main()