Error: Layer weight shape (3, 3, 3, 64) not compatible with provided weight shape (64, 3, 3, 3)
khawla....@gmail.com
When
I tried to load VGG16 weights with this function :
load_model_weights(model,
weights_path):
f = h5py.File(weights_path)
nb_layers =
len(f.attrs["layer_names"])
for k in range(nb_layers):
if k >= len(model.layers):
# without the last
(fully-connected) layers in the savefile
break
g = f['layer_{}'.format(k)]
weights =
[g['param_{}'.format(p)] for p in range(g.attrs['nb_params'])]
model.layers[k].set_weights(weights)
model.layers[k].trainable = False
f.close()
I
am having this error :
ValueError:
Layer weight shape (3, 3, 3, 64) not compatible with provided weight
shape (64, 3, 3, 3)
Some
users suggested to use K.set_image_dim_ordering('tf') but I am
having this error:
Traceback
(most recent call last):
File "<input>", line 1,
in <module>
AttributeError:
module 'keras.backend' has no attribute 'set_image_dim_ordering'
Other
users suggested :
from
keras.utils.conv_utils import convert_kernel
reshaped_weights
= convert_kernel(weights)
model.layers[k].set_weights(reshaped_weights)
But
I got this error :
Traceback
(most recent call last):
File "<input>", line 1,
in <module>
File "<input>", line
11, in load_model_weights
File
".../lib/python3.6/site-packages/keras/utils/conv_utils.py",
line 78, in convert_kernel
raise ValueError('Invalid kernel
shape:', kernel.shape)
ValueError:
('Invalid kernel shape:', (0,))
keras version: '2.3.1'
Tensorflow version: 1.15.0
Any one can help me?