rcasero/keras

Deep Learning for humans

keras's Issues

tensorflow.python.framework.errors_impl.InvalidArgumentError: Incompatible shape

In commit f1c5c1d, running the following code with python 3.6

import os
os.environ['KERAS_BACKEND'] = 'tensorflow'
import keras
import keras.backend as K
import numpy as np

from keras.models import Model, Sequential
from keras.layers import Activation, Conv2D, Input
from keras.layers.normalization import BatchNormalization

from keras.utils import multi_gpu_model

# remove warning "Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA"
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'  # Just disables the warning, doesn't enable AVX/FMA

# image_data_format = 'channels_first'
image_data_format = 'channels_last'

K.set_image_data_format(image_data_format)

# simulate input images
if image_data_format == 'channels_first':
    im = np.zeros(shape=(10, 3, 64, 64), dtype='uint8')

    # simulate network output
    out = 2 * np.ones(shape=(10, 1, 64, 64), dtype='float32')
    aux_out = 5 * np.ones(shape=(10, 1, 22, 22), dtype='float32')
    # simulate training weights for network output
    weight = np.ones(shape=(10, 1, 64, 64), dtype='float32')
    aux_weight = np.ones(shape=(10, 1, 22, 22), dtype='float32')

    # simulate validation data
    im_validation = 3 * np.ones(shape=(5, 3, 64, 64), dtype='uint8')
    out_validation = 4 * np.ones(shape=(5, 1, 64, 64), dtype='float32')

elif image_data_format == 'channels_last':
    im = np.zeros(shape=(10, 64, 64, 3), dtype='uint8')

    # simulate network output
    out = 2 * np.ones(shape=(10, 64, 64, 1), dtype='float32')
    aux_out = 5 * np.ones(shape=(10, 22, 22, 1), dtype='float32')
    # simulate training weights for network output
    weight = np.ones(shape=(10, 64, 64, 1), dtype='float32')
    aux_weight = np.ones(shape=(10, 22, 22, 1), dtype='float32')

    # simulate validation data
    im_validation = 3 * np.ones(shape=(5, 64, 64, 3), dtype='uint8')
    out_validation = 4 * np.ones(shape=(5, 64, 64, 1), dtype='float32')

else:
    raise ValueError('Unrecognised position for channels')


validation_data = (im_validation, out_validation)

# optimizer
optimizer = keras.optimizers.SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True)

'''simple CNN with one output
'''

# create network model
model = Sequential()
model.add(Conv2D(input_shape=im.shape[1:],
                 filters=32, kernel_size=(3, 3), strides=1, padding='same'))
model.add(BatchNormalization(axis=3))
model.add(Activation('relu'))
model.add(Conv2D(filters=1, kernel_size=(1, 1), strides=1, padding='same'))

'''string format (sample_weights_mode='element')
'''

# compile model
model.compile(loss='mae', optimizer=optimizer, metrics=['accuracy'], sample_weight_mode='element')

# train with validation_data
model.fit(im, out, sample_weight=weight, validation_data=validation_data, batch_size=3, epochs=3)

produces the following error

Traceback (most recent call last):
  File "<input>", line 79, in <module>
  File "/home/rcasero/.conda/envs/cytometer_tensorflow/lib/python3.6/site-packages/keras/engine/training.py", line 1070, in fit
    validation_steps=validation_steps)
  File "/home/rcasero/.conda/envs/cytometer_tensorflow/lib/python3.6/site-packages/keras/engine/training_arrays.py", line 199, in fit_loop
    outs = f(ins_batch)
  File "/home/rcasero/.conda/envs/cytometer_tensorflow/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 2661, in __call__
    return self._call(inputs)
  File "/home/rcasero/.conda/envs/cytometer_tensorflow/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py", line 2631, in _call
    fetched = self._callable_fn(*array_vals)
  File "/home/rcasero/.conda/envs/cytometer_tensorflow/lib/python3.6/site-packages/tensorflow/python/client/session.py", line 1454, in __call__
    self._session._session, self._handle, args, status, None)
  File "/home/rcasero/.conda/envs/cytometer_tensorflow/lib/python3.6/site-packages/tensorflow/python/framework/errors_impl.py", line 519, in __exit__
    c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: Incompatible shapes: [3,64,64] vs. [3,64,64,1]
	 [[Node: loss/conv2d_2_loss/mul = Mul[T=DT_FLOAT, _class=["loc:@training/SGD/gradients/loss/conv2d_2_loss/mul_grad/Sum"], _device="/job:localhost/replica:0/task:0/device:GPU:0"](loss/conv2d_2_loss/Mean, _arg_conv2d_2_sample_weights_0_2/_47)]]
	 [[Node: loss/mul/_99 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge_518_loss/mul", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"]()]]