Input to reshape is a tensor with 102400 values, but the requested shape requires a multiple of 165888

[Pablo_Gonzalez_Uzeta]

Befor using sparse_categorical_crossentropy i used to be getting :

ValueError: Shapes (None, 1) and (None, 26) are incompatible

Afterwards i am getting :

Node: ‘sequential_3/flatten_3/Reshape’
Input to reshape is a tensor with 102400 values, but the requested shape requires a multiple of 165888
[[{{node sequential_3/flatten_3/Reshape}}]] [Op:__inference_train_function_2600]

Evrything seems right, any ideas ?

[balaji.ambresh]

The number of output neurons should not be 1 since this is a multi class classification problem. So, the 1st feedback: ValueError: Shapes (None, 1) and (None, 26) are incompatible is correct.

As far as the 2nd error is concerned, please confirm that the input_shape parameter is set correctly.

[Pablo_Gonzalez_Uzeta]

Thank you, you were right.