How to get predictions from deepFM saved model using libsering.serialization.tf_saved()

[BhaveshBhansali]

Hello,

I have saved deepFM model using libsering.serialization.tf_saved(model). I load the model using:
loaded_model = tf.saved_model.load('model')

The model predict signature looks like:
_SignatureMap({'predict': <ConcreteFunction pruned(dense_values, item_indices, sparse_indices, user_indices) at 0x56E44DA90>})

However, when I am trying to call model using following, it gives me error:

restored_output_tensor = test_model.signatures['predict'](feed_dict) # where feed_dict is dictionary of features.
`TypeError: pruned(dense_values, item_indices, sparse_indices, user_indices) takes 0 positional arguments, got 1.`

Please find screenshot of the same.

[massquantity]

libsering.serialization.tf_saved is not compatible with tf.saved_model.load. If you want to save and load a model, use Save/Load API.