-
Notifications
You must be signed in to change notification settings - Fork 0
/
train.py
146 lines (118 loc) · 3.67 KB
/
train.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
import time
import torch
from torch.functional import split
import torchvision.transforms as transforms
import torch.optim as optim
import torchvision.transforms.functional as FT
from tqdm import tqdm
from torch.utils.data import DataLoader
from model import Yolov1
from dataset import VOCDataset
from utils import (
intersection_over_union,
mean_average_precision,
non_max_suppression,
mean_average_precision,
cellboxes_to_boxes,
get_bboxes,
plot_image,
save_checkpoint,
load_checkpoint
)
from loss import YoloLoss
seed = 123
torch.manual_seed(seed)
# Hyperparameters ect.
LEARNING_RATE = 2e-5
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
BATCH_SIZE = 16
WEIGHT_DECAY = 0
EPOCHS = 100
NUM_WORKERS = 2
PIN_MEMORY = True
LOAD_MODEL = True
LOAD_MODEL_FILE = "overfit.pth.tar"
IMG_DIR = "data/images"
LABEL_DIR = "data/labels"
class Compose(object):
def __init__(self, transforms):
self.transforms = transforms
def __call__(self, img, bboxes):
for t in self.transforms:
img, bboxes = t(img), bboxes
return img, bboxes
transform = Compose([transforms.Resize((448, 448)), transforms.ToTensor()])
def train_fn(train_loader, model, optimizer, loss_fn):
loop = tqdm(train_loader, leave=True)
mean_loss = []
for batch_idx, (x, y) in enumerate(loop):
x, y = x.to(DEVICE), y.to(DEVICE)
out = model(x)
loss = loss_fn(out, y)
mean_loss.append(loss.item())
optimizer.zero_grad()
loss.backward()
optimizer.step()
# Update the progress bar
loop.set_postfix(loss=loss.item())
print(f"Mean loss was {sum(mean_loss)/len(mean_loss)}")
def main():
model = Yolov1(split_size=7, num_boxes=2, num_classes=20).to(DEVICE)
optimaizer = optim.Adam(
model.parameters(), lr=LEARNING_RATE, weight_decay=WEIGHT_DECAY
)
loss_fn = YoloLoss()
if LOAD_MODEL:
load_checkpoint(torch.load(LOAD_MODEL_FILE), model, optimaizer)
train_dataset = VOCDataset(
"data/8examples.csv",
transform=transform,
img_dir=IMG_DIR,
label_dir=LABEL_DIR
)
test_dataset = VOCDataset(
"data/test.csv",
transform=transform,
img_dir=IMG_DIR,
label_dir=LABEL_DIR
)
train_loader = DataLoader(
dataset=train_dataset,
batch_size=BATCH_SIZE,
num_workers=NUM_WORKERS,
pin_memory=PIN_MEMORY,
shuffle=True,
drop_last=False
)
test_loader = DataLoader(
dataset=test_dataset,
batch_size=BATCH_SIZE,
num_workers=NUM_WORKERS,
pin_memory=PIN_MEMORY,
shuffle=True,
drop_last=True
)
for epoch in range(EPOCHS):
for x, y in train_loader:
x = x.to(DEVICE)
for idx in range(8):
bboxes = cellboxes_to_boxes(model(x))
bboxes = non_max_suppression( bboxes[idx], iou_threshold=0.5, threshold=0.4)
plot_image(x[idx].permute(1, 2, 0).to("cpu"), bboxes)
pred_boxes, target_boxes = get_bboxes(
train_loader, model, iou_threshold=0.5, threshold=0.4
)
mean_avg_prec = mean_average_precision(
pred_boxes, target_boxes, iou_threshold=0.5, box_format="midpoint"
)
print(f"Train mAP: {mean_avg_prec}")
if mean_avg_prec > 0.9:
check_point = {
"state_dict": model.state_dict(),
"optimizer": optimaizer.state_dict()
}
save_checkpoint(check_point, filename=LOAD_MODEL_FILE)
time.sleep(10)
train_fn(train_loader, model, optimaizer, loss_fn)
if __name__ == "__main__":
main()