batch_size=FLAGS.batch_size, class_mode='categorical') test_data = image_data_generator.flow_from_directory( '/home/meizu/WORK/code/YF_baidu_ML/dataset/flowers/flower_photos/test', target_size=(FLAGS.input_image_size, FLAGS.input_image_size), batch_size=FLAGS.batch_size, class_mode='categorical') # compile the model (should be done *after* setting layers to non-trainable) model.compile(optimizer='rmsprop', loss='categorical_crossentropy') train_data_count = len(train_data.filenames) test_data_count = len(test_data.filenames) model.fit_generator(train_data, steps_per_epoch=(train_data_count // FLAGS.batch_size + 1), epochs=1, verbose=1, validation_data=test_data, validation_steps=(test_data_count // FLAGS.batch_size + 1)) # a = model_inception_v3.predict_generator(g, len(g.filenames)) # print '' # a = model_inception_v3.predict_generator(g, steps=len(g.filenames)) # # # extract features # for label_name, label_lists in flowers_dataset.items(): # for category in ['training', 'testing', 'validation']: # category_list = label_lists[category] # for index, image_path in enumerate(category_list): # if not gfile.Exists(image_path): # tf.logging.fatal('File does not exist %s', image_path) # img = misc.imresize(misc.imread(image_path), [299, 299]).astype(np.float32)
'\n', ' Returns:\n', ' Numpy array(s) of predictions.\n', '\n', ' Raises:\n', ' ValueError: In case of mismatch between the provided\n', " input data and the model's expectations,\n", ' or in case a stateful model receives a number of samples\n', ' that is not a multiple of the batch size.\n', """ num_epochs = 2 for epoch in range(num_epochs): model.fit_generator(generator=val_batches, steps_per_epoch=3, epochs=1) """ ([' def fit_generator(self,\n', ' generator,\n', ' steps_per_epoch,\n', ' epochs=1,\n', ' verbose=1,\n', ' callbacks=None,\n', ' validation_data=None,\n', ' validation_steps=None,\n', ' class_weight=None,\n', ' max_q_size=10,\n', ' workers=1,\n', ' pickle_safe=False,\n', ' initial_epoch=0):\n',