forked from swathikirans/ego-rnn
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmain-run-flow.py
187 lines (157 loc) · 8.21 KB
/
main-run-flow.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
from __future__ import print_function, division
from flow_resnet import *
from spatial_transforms import (Compose, ToTensor, CenterCrop, Scale, Normalize, MultiScaleCornerCrop,
RandomHorizontalFlip)
from tensorboardX import SummaryWriter
import torch.nn as nn
from torch.autograd import Variable
from makeDatasetFlow import *
import argparse
import sys
def main_run(dataset, trainDir, valDir, outDir, stackSize, trainBatchSize, valBatchSize, numEpochs, lr1,
decay_factor, decay_step):
if dataset == 'gtea61':
num_classes = 61
elif dataset == 'gtea71':
num_classes = 71
elif dataset == 'gtea_gaze':
num_classes = 44
elif dataset == 'egtea':
num_classes = 106
else:
print('Dataset not found')
sys.exit()
min_accuracy = 0
model_folder = os.path.join('./', outDir, dataset, 'flow') # Dir for saving models and log files
# Create the dir
if os.path.exists(model_folder):
print('Dir {} exists!'.format(model_folder))
sys.exit()
os.makedirs(model_folder)
# Log files
writer = SummaryWriter(model_folder)
train_log_loss = open((model_folder + '/train_log_loss.txt'), 'w')
train_log_acc = open((model_folder + '/train_log_acc.txt'), 'w')
val_log_loss = open((model_folder + '/val_log_loss.txt'), 'w')
val_log_acc = open((model_folder + '/val_log_acc.txt'), 'w')
# Data loader
normalize = Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
spatial_transform = Compose([Scale(256), RandomHorizontalFlip(), MultiScaleCornerCrop([1, 0.875, 0.75, 0.65625], 224),
ToTensor(), normalize])
vid_seq_train = makeDataset(trainDir, spatial_transform=spatial_transform, sequence=False,
stackSize=stackSize, fmt='.jpg')
train_loader = torch.utils.data.DataLoader(vid_seq_train, batch_size=trainBatchSize,
shuffle=True, sampler=None, num_workers=4, pin_memory=True)
if valDir is not None:
vid_seq_val = makeDataset(valDir, spatial_transform=Compose([Scale(256), CenterCrop(224), ToTensor(), normalize]),
sequence=False, stackSize=stackSize, fmt='.jpg', phase='Test')
val_loader = torch.utils.data.DataLoader(vid_seq_val, batch_size=valBatchSize,
shuffle=False, num_workers=2, pin_memory=True)
valInstances = vid_seq_val.__len__()
trainInstances = vid_seq_train.__len__()
print('Number of samples in the dataset: training = {} | validation = {}'.format(trainInstances, valInstances))
model = flow_resnet34(True, channels=2*stackSize, num_classes=num_classes)
model.train(True)
train_params = list(model.parameters())
model.cuda()
loss_fn = nn.CrossEntropyLoss()
optimizer_fn = torch.optim.SGD(train_params, lr=lr1, momentum=0.9, weight_decay=5e-4)
optim_scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer_fn, milestones=decay_step, gamma=decay_factor)
train_iter = 0
for epoch in range(numEpochs):
optim_scheduler.step()
epoch_loss = 0
numCorrTrain = 0
trainSamples = 0
iterPerEpoch = 0
model.train(True)
writer.add_scalar('lr', optimizer_fn.param_groups[0]['lr'], epoch+1)
for i, (inputs, targets) in enumerate(train_loader):
train_iter += 1
iterPerEpoch += 1
optimizer_fn.zero_grad()
inputVariable = Variable(inputs.cuda())
labelVariable = Variable(targets.cuda())
trainSamples += inputs.size(0)
output_label, _ = model(inputVariable)
loss = loss_fn(output_label, labelVariable)
loss.backward()
optimizer_fn.step()
_, predicted = torch.max(output_label.data, 1)
numCorrTrain += (predicted == targets.cuda()).sum()
epoch_loss += loss.data[0]
avg_loss = epoch_loss/iterPerEpoch
trainAccuracy = (numCorrTrain / trainSamples) * 100
print('Train: Epoch = {} | Loss = {} | Accuracy = {}'.format(epoch + 1, avg_loss, trainAccuracy))
writer.add_scalar('train/epoch_loss', avg_loss, epoch+1)
writer.add_scalar('train/accuracy', trainAccuracy, epoch+1)
train_log_loss.write('Training loss after {} epoch = {}\n'.format(epoch+1, avg_loss))
train_log_acc.write('Training accuracy after {} epoch = {}\n'.format(epoch+1, trainAccuracy))
if valDir is not None:
if (epoch+1) % 1 == 0:
model.train(False)
val_loss_epoch = 0
val_iter = 0
val_samples = 0
numCorr = 0
for j, (inputs, targets) in enumerate(val_loader):
val_iter += 1
val_samples += inputs.size(0)
inputVariable = Variable(inputs.cuda(), volatile=True)
labelVariable = Variable(targets.cuda(async=True), volatile=True)
output_label, _ = model(inputVariable)
val_loss = loss_fn(output_label, labelVariable)
val_loss_epoch += val_loss.data[0]
_, predicted = torch.max(output_label.data, 1)
numCorr += (predicted == targets.cuda()).sum()
val_accuracy = (numCorr / val_samples) * 100
avg_val_loss = val_loss_epoch / val_iter
print('Validation: Epoch = {} | Loss = {} | Accuracy = {}'.format(epoch + 1, avg_val_loss, val_accuracy))
writer.add_scalar('val/epoch_loss', avg_val_loss, epoch + 1)
writer.add_scalar('val/accuracy', val_accuracy, epoch + 1)
val_log_loss.write('Val Loss after {} epochs = {}\n'.format(epoch + 1, avg_val_loss))
val_log_acc.write('Val Accuracy after {} epochs = {}%\n'.format(epoch + 1, val_accuracy))
if val_accuracy > min_accuracy:
save_path_model = (model_folder + '/model_flow_state_dict.pth')
torch.save(model.state_dict(), save_path_model)
min_accuracy = val_accuracy
else:
if (epoch+1) % 10 == 0:
save_path_model = (model_folder + '/model_flow_state_dict_epoch' + str(epoch+1) + '.pth')
torch.save(model.state_dict(), save_path_model)
train_log_loss.close()
train_log_acc.close()
val_log_acc.close()
val_log_loss.close()
writer.export_scalars_to_json(model_folder + "/all_scalars.json")
writer.close()
def __main__():
parser = argparse.ArgumentParser()
parser.add_argument('--dataset', type=str, default='gtea61', help='Dataset')
parser.add_argument('--trainDatasetDir', type=str, default='./dataset/gtea_warped_flow_61/split2/train',
help='Train set directory')
parser.add_argument('--valDatasetDir', type=str, default=None,
help='Validation set directory')
parser.add_argument('--outDir', type=str, default='experiments', help='Directory to save results')
parser.add_argument('--stackSize', type=int, default=5, help='Length of sequence')
parser.add_argument('--trainBatchSize', type=int, default=32, help='Training batch size')
parser.add_argument('--valBatchSize', type=int, default=32, help='Validation batch size')
parser.add_argument('--numEpochs', type=int, default=750, help='Number of epochs')
parser.add_argument('--lr', type=float, default=1e-2, help='Learning rate')
parser.add_argument('--stepSize', type=float, default=[150, 300, 500], nargs="+", help='Learning rate decay step')
parser.add_argument('--decayRate', type=float, default=0.5, help='Learning rate decay rate')
args = parser.parse_args()
dataset = args.dataset
trainDatasetDir = args.trainDatasetDir
valDatasetDir = args.valDatasetDir
outDir = args.outDir
stackSize = args.stackSize
trainBatchSize = args.trainBatchSize
valBatchSize = args.valBatchSize
numEpochs = args.numEpochs
lr1 = args.lr
stepSize = args.stepSize
decayRate = args.decayRate
main_run(dataset, trainDatasetDir, valDatasetDir, outDir, stackSize, trainBatchSize, valBatchSize, numEpochs, lr1,
decayRate, stepSize)
__main__()