Hello, I am at the last assignment of the course and I receive an error that I don’t understand when I try to train the Siamese network using the train_model function.
The train_model function definition seems pretty straight forward, so I don’t understand where I could have gone wrong. I receive the following error message:
Epoch 1/2
Tensor("TripletLoss/diag_part:0", shape=(None,), dtype=float32)
Tensor("TripletLoss/diag_part:0", shape=(None,), dtype=float32)
---------------------------------------------------------------------------
UnimplementedError Traceback (most recent call last)
Cell In[26], line 9
3 train_generator = train_dataset.shuffle(len(train_Q1),
4 seed=7,
5 reshuffle_each_iteration=True).batch(batch_size=batch_size)
6 val_generator = val_dataset.shuffle(len(val_Q1),
7 seed=7,
8 reshuffle_each_iteration=True).batch(batch_size=batch_size)
----> 9 model = train_model(Siamese, TripletLoss,text_vectorization,
10 train_generator,
11 val_generator,
12 epochs=epochs,)
Cell In[25], line 29, in train_model(Siamese, TripletLoss, text_vectorizer, train_dataset, val_dataset, d_feature, lr, epochs)
25 model.compile(loss=TripletLoss,
26 optimizer = tf.keras.optimizers.Adam(learning_rate=lr)
27 )
28 # Train the model
---> 29 model.fit(train_dataset,
30 epochs = epochs,
31 validation_data = val_dataset,
32 )
34 ### END CODE HERE ###
36 return model
...
UnimplementedError: Graph execution error:
...
tensor = tf.cast(tensor, dtype=ref_input.dtype)
Node: 'SiameseModel/Cast'
2 root error(s) found.
(0) UNIMPLEMENTED: Cast string to float is not supported
[[{{node SiameseModel/Cast}}]]
(1) CANCELLED: Function was cancelled before it was started
0 successful operations.
0 derived errors ignored. [Op:__inference_train_function_26261]