def train(self):
train_data = ak.image_dataset_from_directory(self.data_path)
self.model = self._create_model()
logger.info(f"executing a {self.model.__class__.__name__} algorithm...")
logger.info(f"Training started...")
self.model.fit(train_data, **self.training_args)
logger.info("finished training!")
self.save_desc_file()
saved = self.save_model()
if saved:
logger.info(f"model saved successfully")
import autokeras as ak
import tensorflow as tf
import os
from pathlib import Path
HERE = Path(__file__).parent.absolute()
train_data = ak.image_dataset_from_directory(HERE.joinpath(
'traindata', 'digits'),
image_size=(120, 120),
batch_size=64)
clf = ak.ImageClassifier(overwrite=True,
max_trials=1,
project_name='digitsTrainer')
clf.fit(train_data, epochs=5)
print(clf.evaluate(train_data))
clf.export_model().save("digits", save_format="tf")
# TODO: use tesseract to verify images for traindata and train on many more digits
sunflowers/
tulips/
```
We can split the data into training and testing as we load them.
"""
batch_size = 32
img_height = 180
img_width = 180
train_data = ak.image_dataset_from_directory(
data_dir,
# Use 20% data as testing data.
validation_split=0.2,
subset="training",
# Set seed to ensure the same split when loading testing data.
seed=123,
image_size=(img_height, img_width),
batch_size=batch_size,
)
test_data = ak.image_dataset_from_directory(
data_dir,
validation_split=0.2,
subset="validation",
seed=123,
image_size=(img_height, img_width),
batch_size=batch_size,
)
"""
Then we just do one quick demo of AutoKeras to make sure the dataset works.
def predict(self):
trained_model = self.load_model()
pred_data = ak.image_dataset_from_directory(self.data_path)
trained_model.predict(pred_data)
def evaluate(self):
trained_model = self.load_model()
test_data = ak.image_dataset_from_directory(self.data_path)
trained_model.evaluate(test_data)