-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtrain3.py
179 lines (148 loc) · 7.17 KB
/
train3.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
#MobileViT
import torch
from torchvision import transforms
from torch.utils.data import DataLoader
import torch.nn as nn
import wandb
import os
from datetime import datetime
import math
from tqdm import tqdm
from transformers import SegformerForSemanticSegmentation, MobileViTForSemanticSegmentation, MobileViTFeatureExtractor
from model import *
from loss import *
from dataset import *
from positionalembedding import *
np.random.seed(0)
torch.manual_seed(0)
wandb.init(project='inpaint', name='mobilevit-testing')
train_filepath = '/data/cornucopia/jsb212/seg-dataset/images_arshot'
val_filepath = '/data/cornucopia/jsb212/seg-dataset/eval-inpaint'
device = 'cuda' if torch.cuda.is_available() else 'cpu'
batch_size = 32
num_epochs = 250
lr = 0.001
pos_embedding = 64 #int or False
normalise = False
height = width = 256
beta = 0 #style loss hyperparam
now = datetime.now()
now_str = './outputs/' + now.strftime('%Y-%m-%d_%H-%M')
os.makedirs(now_str, exist_ok=True)
os.makedirs(f'{now_str}/imgs_epochs_train', exist_ok=True)
os.makedirs(f'{now_str}/imgs_epochs_val', exist_ok=True)
# Define the loss function
loss_fn = WeightedLoss([VGGLoss(),
nn.MSELoss(),
TVLoss(p=1)],
[1, 30, 10]).to(device) #1, 40, 10
styleloss = StyleLoss(device=device)
img_transforms_lst = [
transforms.Resize((height, width)),
transforms.ToTensor()
]
if normalise:
img_transforms_lst.append(transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]))
invNormalise = InvNormalise(device=device)
img_transform = transforms.Compose(img_transforms_lst)
feature_extractor = MobileViTFeatureExtractor.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
#Load data
train_data = ImgMaskDataset(train_filepath, img_transform)
val_data = ImgMaskDataset(val_filepath, img_transform)
num_samples = len(train_data.imgs) + len(val_data)
train_loader = DataLoader(train_data, batch_size=batch_size, shuffle=True, drop_last=True)
val_loader = DataLoader(val_data, batch_size=batch_size, shuffle=True, drop_last=True)
#Initialise model
model = MobileViTForSemanticSegmentation.from_pretrained("apple/deeplabv3-mobilevit-xx-small")
model.segmentation_head.classifier = nn.Sequential(nn.ConvTranspose2d(256, 128, kernel_size=4, stride=2, padding=1, output_padding=0),
nn.LeakyReLU(),
nn.ConvTranspose2d(128, 64, kernel_size=4, stride=2, padding=1, output_padding=0),
nn.LeakyReLU(),
nn.ConvTranspose2d(64, 32, kernel_size=4, stride=2, padding=1, output_padding=0),
nn.LeakyReLU(),
nn.ConvTranspose2d(32, 3, kernel_size=4, stride=2, padding=1, output_padding=0),
nn.Tanh())
# model.decode_head.classifier = InpaintingHead2() #segformer
if pos_embedding:
pe = positionalencoding2d(pos_embedding, height, width)
pe = pe.to(device)
model.mobilevit.conv_stem.convolution = nn.Conv2d(in_channels=3+pos_embedding, out_channels=16,
kernel_size=3, stride=2, padding=1)
if torch.cuda.device_count() > 1:
print(torch.cuda.device_count(), "GPUs")
model = nn.DataParallel(model)
model.to(device)
model_params = get_state_dict(model)
# Define the optimizer
optimizer = torch.optim.Adam(model.parameters(), lr=lr)
scheduler = torch.optim.lr_scheduler.ExponentialLR(optimizer, gamma=0.9995)
num_iters = math.floor(num_samples/batch_size) * num_epochs
# Train the model
loss_val = 10000
with tqdm(total=num_iters) as pbar:
for epoch in range(num_epochs):
# Training loop
model.train()
for i, (images, masks) in enumerate(train_loader):
images, masks = images.to(device), masks.to(device)
if normalise:
imagesT = invNormalise(images)
masksX = masks.unsqueeze(1).repeat(1, 3, 1, 1)
images_masked = images * masksX
if pos_embedding:
peX = pe.repeat(images_masked.shape[0], 1, 1, 1)
images_masked = torch.cat((images_masked, peX), 1)
images_masked = images_masked.to(device)
optimizer.zero_grad()
outputs = model(images_masked).logits
if normalise:
outputs = invNormalise(outputs)
images = imagesT
output_comp = images * masksX + outputs * (1-masksX) #composite output - combine nonmasked area of gt with masked area of generation
loss = loss_fn(outputs, images)
#loss += beta*(styleloss(outputs, images) + styleloss(output_comp, images))
loss.backward()
optimizer.step()
pbar.update(1)
torch.cuda.empty_cache()
if epoch % 10 == 0:
save_array_as_img(array=outputs[0], filepath=f'{now_str}/imgs_epochs_train/epoch_{epoch}.png')
# Validation loop
model.eval()
with torch.no_grad():
for i, (images, masks) in enumerate(val_loader):
images, masks = images.to(device), masks.to(device)
if normalise:
imagesT = invNormalise(images)
masksX = masks.unsqueeze(1).repeat(1, 3, 1, 1)
images_masked = images * masksX
if pos_embedding:
peX = pe.repeat(images_masked.shape[0], 1, 1, 1)
images_masked = torch.cat((images_masked, peX), 1)
images_masked = images_masked.to(device)
outputs = model(images_masked).logits
if normalise:
outputs = invNormalise(outputs)
images = imagesT
output_comp = images * masksX + outputs * (1-masksX) #composite output - combine nonmasked area of gt with masked area of generation
val_loss = loss_fn(outputs, images)
#val_loss += beta*(styleloss(outputs, images) + styleloss(output_comp, images))
pbar.update(1)
torch.cuda.empty_cache()
if epoch % 10 == 0:
save_array_as_img(array=outputs[0], filepath=f'{now_str}/imgs_epochs_val/epoch_{epoch}.png')
current_lr = scheduler.get_last_lr()[0]
print("Epoch {}: Train Loss: {}, Val Loss: {}".format(epoch+1, loss.item(), val_loss.item()))
wandb.log({'Train Loss': loss.item(), 'Val Loss': val_loss.item(), 'Learning Rate': current_lr})
if loss.item() < loss_val:
model_params = get_state_dict(model)
scheduler.step()
for i, output in enumerate(outputs):
save_array_as_img(array=output, filepath=f'{now_str}/output_{i}.png')
input = images[i]
save_array_as_img(array=input, filepath=f'{now_str}/input_{i}.png')
mask = masks[i]
save_array_as_img(array=mask, filepath=f'{now_str}/mask_{i}.png')
final_output = input * mask + output * (1-mask)
save_array_as_img(array=final_output, filepath=f'{now_str}/final_img_{i}.png')
torch.save(model_params, f'{now_str}/mobilevit.pth')