Unrecognized optimizer and loss function in TensorFlow

Question

I'm encountering an issue while trying to compile a Keras model in TensorFlow 2.15.0 using a custom optimizer and loss function, in google colab.

Here's the code snippet that works fine

model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])

However, when I attempt to configure the optimizer and loss separately as follows:

# Configure the optimizer, loss, and model
optimizer = tf.keras.optimizers.Adam(learning_rate=0.01)
loss = tf.keras.losses.CategoricalCrossentropy(from_logits=False)
print(optimizer)
print(loss)

model.compile(optimizer=optimizer, loss=loss, metrics=['accuracy'])

I encounter the following error:

ValueError: Could not interpret optimizer identifier: <keras.src.optimizers.adam.Adam object at 0x7f925c90ace0>

It seems like TensorFlow is having trouble interpreting the custom optimizer object. Here are the details:

TensorFlow version: 2.15.0
Custom optimizer: <keras.src.optimizers.adam.Adam object at 0x7f925c90ace0>
Custom loss function: <keras.src.losses.CategoricalCrossentropy object at 0x7f9301977c10>

Answer 1

This what resolved the issue.

1. Check Installed Packages

    !pip list | grep keras
   

   This returned unexpected "tf_keras"

    keras                            2.15.0
    tf_keras                         2.15.1
   

2. Uninstall "tf_keras"

    !pip uninstall tf_keras
   

3. Restart Runtime (Ctrl+M)

4. Verify Installed Packages.

    !pip list | grep keras
   

   Should return only keras

    keras                            2.15.0