-
Notifications
You must be signed in to change notification settings - Fork 0
/
training.py
54 lines (42 loc) · 1.35 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
"""
AUTHOR: Jopapy19
Date: 2021/03-13 UM907 32
"""
import os
import utils.config as config
import tensorflow as tf
import utils.data_management as dm
import time
from utils import model
from PIL import Image
def get_unique_model_name(specific_name="VGG16_model"):
model_fileName = time.strftime(f"{specific_name}_at_%Y%m%d_%H%M%S.h5")
os.makedirs(config.TRAINED_MODEL_DIR, exist_ok=True)
model_file_path = os.path.join(config.TRAINED_MODEL_DIR, model_fileName)
return model_file_path
def train():
my_model = model.custom_model()
callbacks = model.callbacks()
train_generator, valid_generator = dm.train_valid_generator()
"""
STEPS FOR EPOCHS
train_generator.samples = 256
batch_size = 16
Fit and save the model
"""
steps_per_epoch = train_generator.samples // train_generator.batch_size
validation_steps = valid_generator.samples // valid_generator.batch_size
my_model.fit(
train_generator,
validation_data=valid_generator,
epochs=config.EPOCHS,
steps_per_epoch=steps_per_epoch,
validation_steps=validation_steps,
callbacks=callbacks
)
#Saving our model
model_file_path = get_unique_model_name()
my_model.save(model_file_path)
print(f"Saving model at\n ==>{model_file_path}")
if __name__ == "__main__":
train()