issue of conversion about tensorflow HOT 4 CLOSED

i had the same error!

I found a soultion using the keras legacy optimizer like this :

tf.keras.optimizers.legacy.SGD

python: 3.8.18
tensorflow: 2.13.0

from tensorflow.

Hi @snehajhaveri01,

Please fill the template and share the reproducible code.

Thank You

from tensorflow.

i had the same error!

I found a soultion using the keras legacy optimizer like this :

tf.keras.optimizers.legacy.SGD

python: 3.8.18 tensorflow: 2.13.0

i solved it the same way , thank you !

from tensorflow.

3.8.18

i had the same error!
I found a soultion using the keras legacy optimizer like this :
tf.keras.optimizers.legacy.SGD
python: 3.8.18 tensorflow: 2.13.0

i solved it the same way , thank you !

how'd you solve it? i could not maintain python version in colab. I tried it on my machine. I have .h5 model but it is not able to load and showing

TypeError: Error when deserializing class 'InputLayer' using config={'batch_shape': [None, 50], 'dtype': 'float32', 'sparse': False, 'name': 'input_layer_10'}.

Exception encountered: Unrecognized keyword arguments: ['batch_shape']

I tried making python version of 3.8.18 and when installed tensorflow==2.13.0. It caused some problems with numpy and when I resolved it. now it is still not resolving the issue.

@maximilianofried @snehajhaveri01

from tensorflow.