VIME_Training.ipynb

[edangx100]

Hi Antons,

When i run this notebook, i have this error at Cell 22:

• ValueError: Input 0 of layer "sequential_3" is incompatible with the layer: expected shape=(None, 110), found shape=(None, 5, 110)

May I check with you how to resolve this error?

Cell 22 :

vime_s = tf.keras.models.load_model('./vime_self')

# Sequential model to produce embeddings
encoding_model = Sequential(
    [
        Input( num_features),
        vime_s.encoder
    ]
)

dense_model = Sequential(
    [
        Input(num_features),
        Dense(num_features, activation="relu"),
    
    ]
)

# Create corrupted sample
samples = X_val_unlabelled.sample(10)
sample_corrupted = generate_k_corrupted(
    x=samples,
    k=5,
    p_m=0.4
)

val_encoding = encoding_model.predict(sample_corrupted, batch_size=batch_size)
random_encoding = dense_model.predict(sample_corrupted, batch_size=batch_size)

[j-h-CoDe]

i have same issue with upper comment.
how can i solve it?

[nitzantal]

Same here