-
Notifications
You must be signed in to change notification settings - Fork 0
/
training.py
executable file
·172 lines (151 loc) · 7.18 KB
/
training.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
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
@author: winston lin
"""
import torch
import sys
import numpy as np
import os
from utils import cc_coef, DynamicChunkSplitData
from tqdm import tqdm
import torch.optim as optim
import matplotlib.pyplot as plt
from dataloader import MspPodcastDataset
from torch.utils.data.sampler import SubsetRandomSampler
from model import LSTMnet_MeanAtten, LSTMnet_GateAtten, LSTMnet_RnnAtten, LSTMnet_SelfAtten
import argparse
def collate_fn(batch):
data, label = zip(*batch)
# LLDs use 16ms hop size to extract features (16ms*62=0.992sec~=1sec)
chunk_data = DynamicChunkSplitData(data, m=62, C=11, n=1)
label = np.array(label)
return torch.from_numpy(chunk_data), torch.from_numpy(label)
def model_validation(model, valid_loader):
model.eval()
batch_loss_valid_all = []
for _, data_batch in enumerate(tqdm(valid_loader, file=sys.stdout)):
# Input Tensor Data/Targets
input_tensor, input_target = data_batch
input_var = torch.autograd.Variable(input_tensor.cuda()).float()
input_tar = torch.autograd.Variable(input_target.cuda()).float()
# models flow
pred = model(input_var)
# loss calculation
loss = cc_coef(pred, input_tar)
batch_loss_valid_all.append(loss.data.cpu().numpy())
torch.cuda.empty_cache()
return np.mean(batch_loss_valid_all)
###############################################################################
argparse = argparse.ArgumentParser()
argparse.add_argument("-iter", "--iterations", required=True)
argparse.add_argument("-batch", "--batch_size", required=True)
argparse.add_argument("-emo", "--emo_attr", required=True)
argparse.add_argument("-atten", "--atten_type", required=True)
args = vars(argparse.parse_args())
# Parameters
iter_max = int(args['iterations'])
batch_size = int(args['batch_size'])
emo_attr = args['emo_attr']
atten_type = args['atten_type']
shuffle = True
# LSTM-model loading
if atten_type=='NonAtten':
model = LSTMnet_MeanAtten(input_dim=130, hidden_dim=130 , output_dim=1, num_layers=2)
elif atten_type=='GatedVec':
model = LSTMnet_GateAtten(input_dim=130, hidden_dim=130 , output_dim=1, num_layers=2)
elif atten_type=='RnnAttenVec':
model = LSTMnet_RnnAtten(input_dim=130, hidden_dim=130 , output_dim=1, num_layers=2)
elif atten_type=='SelfAttenVec':
model = LSTMnet_SelfAtten(input_dim=130, hidden_dim=130 , output_dim=1, num_layers=2)
model.cuda()
# PATH settings
SAVING_PATH = './Models/'
root_dir = '/media/winston/UTD-MSP/Speech_Datasets/MSP-PODCAST-Publish-1.6/Features/OpenSmile_lld_IS13ComParE/feat_mat/'
label_dir = '/media/winston/UTD-MSP/Speech_Datasets/MSP-PODCAST-Publish-1.6/Labels/labels_concensus.csv'
# creating repo
if not os.path.isdir(SAVING_PATH):
os.makedirs(SAVING_PATH)
# loading datasets
training_dataset = MspPodcastDataset(root_dir, label_dir, split_set='Train', emo_attr=emo_attr)
validation_dataset = MspPodcastDataset(root_dir, label_dir, split_set='Validation', emo_attr=emo_attr)
# shuffle datasets by generating random indices
train_indices = list(range(len(training_dataset)))
valid_indices = list(range(len(validation_dataset)))
if shuffle:
np.random.shuffle(train_indices)
# creating data samplers and loaders:
train_sampler = SubsetRandomSampler(train_indices)
train_loader = torch.utils.data.DataLoader(training_dataset,
batch_size=batch_size,
sampler=train_sampler,
num_workers=12,
pin_memory=True,
collate_fn=collate_fn)
valid_sampler = SubsetRandomSampler(valid_indices)
valid_loader = torch.utils.data.DataLoader(validation_dataset,
batch_size=batch_size,
sampler=valid_sampler,
num_workers=12,
pin_memory=True,
collate_fn=collate_fn)
# create an optimizer for training
optimizer = optim.Adam(model.parameters(), lr=0.0001)
# emotion-recog model training (Iteration-Based)
Iter_trainLoss_All = []
Iter_validLoss_All = []
val_loss_best = 0
iter_count = 0
num_iter_to_valid = 50
while True:
# stopping criteria
if iter_count>=iter_max:
break
for _, data_batch in enumerate(train_loader):
# iter setting & record
model.train()
iter_count += 1
# Input Tensor Data/Targets
input_tensor, input_target = data_batch
input_var = torch.autograd.Variable(input_tensor.cuda()).float()
input_tar = torch.autograd.Variable(input_target.cuda()).float()
# models flow
pred = model(input_var)
# CCC loss for mean target
loss = cc_coef(pred, input_tar)
train_loss = loss.data.cpu().numpy()
Iter_trainLoss_All.append(train_loss)
# compute gradient and do Adam step
optimizer.zero_grad()
loss.backward()
optimizer.step()
# clear GPU memory
torch.cuda.empty_cache()
# do the model validation every XX iterations
if iter_count%num_iter_to_valid==0:
print('validation process')
val_loss = model_validation(model, valid_loader)
Iter_validLoss_All.append(val_loss)
print('Iteration: '+str(iter_count)+' ,Training-loss: '+str(train_loss)+' ,Validation-loss: '+str(val_loss))
print('=================================================================')
# Checkpoint for saving best model based on val-loss
if iter_count/num_iter_to_valid==1:
val_loss_best = val_loss
torch.save(model.state_dict(), os.path.join(SAVING_PATH, 'LSTM_iter'+str(iter_max)+'_batch'+str(batch_size)+'_ChunkSeq2One_'+atten_type+'_'+emo_attr+'.pth.tar'))
print("=> Saving the initial best model (Iteration="+str(iter_count)+")")
else:
if val_loss_best > val_loss:
torch.save(model.state_dict(), os.path.join(SAVING_PATH, 'LSTM_iter'+str(iter_max)+'_batch'+str(batch_size)+'_ChunkSeq2One_'+atten_type+'_'+emo_attr+'.pth.tar'))
print("=> Saving a new best model (Iteration="+str(iter_count)+")")
print("=> Loss reduction from "+str(val_loss_best)+" to "+str(val_loss) )
val_loss_best = val_loss
else:
print("=> Validation Loss did not improve (Iteration="+str(iter_count)+")")
print('=================================================================')
# Drawing Loss Curve for Epoch-based and Batch-based
Iter_trainLoss_All = np.mean(np.array(Iter_trainLoss_All[:len(Iter_validLoss_All)*num_iter_to_valid]).reshape(-1, num_iter_to_valid), axis=1).tolist()
plt.title('Epoch-Loss Curve')
plt.plot(Iter_trainLoss_All,color='blue',linewidth=3)
plt.plot(Iter_validLoss_All,color='red',linewidth=3)
plt.savefig(os.path.join(SAVING_PATH, 'LSTM_iter'+str(iter_max)+'_batch'+str(batch_size)+'_ChunkSeq2One_'+atten_type+'_'+emo_attr+'.png'))
#plt.show()