u-net
u-net copied to clipboard
InvalidArgumentError (see above for traceback): Incompatible shapes: [1146880] vs. [158720]
I´m trying to run the Network but always running into the Error below:
Caused by op 'mul', defined at:
File "train.py", line 206, in <module>
train_and_predict()
File "train.py", line 153, in train_and_predict
model = get_unet()
File "train.py", line 115, in get_unet
model.compile(optimizer=Adam(lr=1e-5), loss=dice_coef_loss, metrics=[dice_coef])
File "/usr/lib64/python3.4/site-packages/keras/engine/training.py", line 899, in compile
sample_weight, mask)
File "/usr/lib64/python3.4/site-packages/keras/engine/training.py", line 430, in weighted
score_array = fn(y_true, y_pred)
File "train.py", line 30, in dice_coef_loss
return 1.-dice_coef(y_true, y_pred)
File "train.py", line 25, in dice_coef
intersection = K.sum(y_true_f * y_pred_f)
File "/usr/lib/python3.4/site-packages/tensorflow/python/ops/math_ops.py", line 794, in binary_op_wrapper
return func(x, y, name=name)
File "/usr/lib/python3.4/site-packages/tensorflow/python/ops/math_ops.py", line 1015, in _mul_dispatch
return gen_math_ops._mul(x, y, name=name)
File "/usr/lib/python3.4/site-packages/tensorflow/python/ops/gen_math_ops.py", line 1625, in _mul
result = _op_def_lib.apply_op("Mul", x=x, y=y, name=name)
File "/usr/lib/python3.4/site-packages/tensorflow/python/framework/op_def_library.py", line 763, in apply_op
op_def=op_def)
File "/usr/lib/python3.4/site-packages/tensorflow/python/framework/ops.py", line 2327, in create_op
original_op=self._default_original_op, op_def=op_def)
File "/usr/lib/python3.4/site-packages/tensorflow/python/framework/ops.py", line 1226, in __init__
self._traceback = _extract_stack()
InvalidArgumentError (see above for traceback): Incompatible shapes: [1146880] vs. [158720]
[[Node: mul = Mul[T=DT_FLOAT, _device="/job:localhost/replica:0/task:0/gpu:0"](Reshape, Reshape_1)]]
[[Node: mul_5/_29 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/cpu:0", send_device="/job:localhost/replica:0/task:0/gpu:0", send_device_incarnation=1, tensor_name="edge_3199_mul_5", tensor_type=DT_FLOAT, _device="/job:localhost/replica:0/task:0/cpu:0"]()]]
At line 25 it has a problem with calculating intersection = K.sum(y_true_f * y_pred_f
Where is y_true and y_pred defined??
Thanks for the help.
y_true and y_pred are parameters