-
Notifications
You must be signed in to change notification settings - Fork 1
/
train.py
128 lines (116 loc) · 3.86 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
import os
import argparse
import math
from typing import NoReturn
import torch
from torch.utils.data.dataloader import DataLoader
from torch.utils.data import Subset
from torch.optim import SGD, lr_scheduler
from torch.hub import load_state_dict_from_url
from dataset import (
get_coco_dataset,
get_number_of_classes,
get_model_categories_metadata,
)
from model import get_fasterrcnn_resnet50_fpn
from engine import train_one_epoch, evaluate
from utils import collate_fn
# ### Global Variables ###
# ## Model ##
TRAINABLE_BACKBONE_LAYERS = 3
# ## Data Fetching ##
BATCH_SIZE = 2
NUM_WORKERS = 2
# ## Optimization ##
LEARNING_RATE = 0.005 * BATCH_SIZE / 4 # Apply linear scaling rule
MOMENTUM = 0.9
WEIGHT_DECAY = 0.0005
STEP_SIZE = 3
GAMMA = 0.1
# ## Training ##
NUM_EPOCHS = 10
RATIO_TRAINING_SPLIT = 0.8
# ## Logging ##
PRINT_FREQUENCY = 10
def train(dataset_path: str) -> NoReturn:
"""Train an object detection model on the given dataset. The
script will store a snapshot of the model after each epoch,
containing the model's weights and the mapping between the
model's output and the categories. Those snapshots can directly
serve as input to the "detect.py" script.
Args:
dataset_path (str): path to the coco training dataset directory.
Returns:
NoReturn: [description]
"""
model_output_path = os.path.join(
"outputs", "models", os.path.basename(dataset_path)
)
os.makedirs(model_output_path, exist_ok=True)
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
print("Running training on device {}".format(device))
train_dataset = get_coco_dataset(dataset_path, train=True)
validation_dataset = get_coco_dataset(dataset_path, train=False)
indices = torch.randperm(len(train_dataset)).tolist()
train_samples = math.ceil(RATIO_TRAINING_SPLIT * len(train_dataset))
train_dataset = Subset(train_dataset, indices[:train_samples])
validation_dataset = Subset(validation_dataset, indices[train_samples:])
training_dataloader = DataLoader(
train_dataset,
batch_size=BATCH_SIZE,
num_workers=NUM_WORKERS,
shuffle=True,
collate_fn=collate_fn,
)
validation_dataloader = DataLoader(
validation_dataset,
batch_size=BATCH_SIZE,
num_workers=NUM_WORKERS,
shuffle=False,
collate_fn=collate_fn,
)
model = get_fasterrcnn_resnet50_fpn(
trainable_backbone_layers=TRAINABLE_BACKBONE_LAYERS,
number_classes=get_number_of_classes(train_dataset.dataset),
)
model.to(device)
# construct an optimizer
params = [p for p in model.parameters() if p.requires_grad]
optimizer = SGD(
params, lr=LEARNING_RATE, momentum=MOMENTUM, weight_decay=WEIGHT_DECAY
)
# and a learning rate scheduler
lr_scheduler_training = lr_scheduler.StepLR(
optimizer, step_size=STEP_SIZE, gamma=GAMMA
)
for epoch in range(NUM_EPOCHS):
train_one_epoch(
model,
optimizer,
training_dataloader,
device,
epoch,
print_freq=PRINT_FREQUENCY,
)
torch.save(
{
"state_dict": model.state_dict(),
"categories": get_model_categories_metadata(train_dataset.dataset),
},
os.path.join(model_output_path, f"epoch_{epoch}.pth"),
)
lr_scheduler_training.step()
# evaluate on the test dataset
evaluate(model, validation_dataloader, device=device)
def main():
parser = argparse.ArgumentParser(description="Train a model on a coco dataset.")
parser.add_argument(
"--dataset-path",
dest="dataset_path",
help="path to your coco dataset directory",
required=True,
)
args = parser.parse_args()
train(args.dataset_path)
if __name__ == "__main__":
main()