Closed 54huige closed 5 years ago
Any luck, getting this error?
Hello, you add 3 items in the train folder and 4 items in the test folder. They must the same number of items with the same folder name.
Found 173 images belonging to 3 classes. Found 240 images belonging to 4 classes.
Also, use at least _numexperiments=20 in your train function.
Why did it go wrong? Must there be more images?
Found 173 images belonging to 3 classes. Found 240 images belonging to 4 classes. JSON Mapping for the model classes saved to pets/json\model_class.json Number of experiments (Epochs) : 1 Epoch 1/1 Traceback (most recent call last): File "O:\imageAI\main.py", line 5, in
model_trainer.trainModel(num_objects=173, num_experiments=1, enhance_data=False, batch_size=32, show_network_summary=True)
File "D:\software\python\lib\site-packages\imageai\Prediction\Custom__init__.py", line 249, in trainModel
validation_steps=int(num_test / batch_size), callbacks=[checkpoint, lr_scheduler])
File "D:\software\python\lib\site-packages\tensorflow\python\keras_impl\keras\engine\training.py", line 1598, in fit_generator
initial_epoch=initial_epoch)
File "D:\software\python\lib\site-packages\tensorflow\python\keras_impl\keras\engine\training_generator.py", line 191, in fit_generator
x, y, sample_weight=sample_weight, class_weight=class_weight)
File "D:\software\python\lib\site-packages\tensorflow\python\keras_impl\keras\engine\training.py", line 1378, in train_on_batch
class_weight=class_weight)
File "D:\software\python\lib\site-packages\tensorflow\python\keras_impl\keras\engine\training.py", line 804, in _standardize_user_data
exception_prefix='target')
File "D:\software\python\lib\site-packages\tensorflow\python\keras_impl\keras\engine\training_utils.py", line 192, in standardize_input_data
' but got array with shape ' + str(data_shape))
ValueError: Error when checking target: expected activation_50 to have shape (173,) but got array with shape (3,)