ValueError: Error when checking target: expected dense_20 to have shape (59,) but got array with shape (1,)

[kskaran94]

Thanks so much for coming here to raise an issue. Please take a moment to 'check' the below boxes:

• output of shape for x and y e.g. (212,12)
• Talos params dictionary
• The Keras model wired for Talos
• Description of extra variables in the model

You can provide the code in pastebin / gist or any other format you like.

---

1. (11541, 59) and (11541, 1)

2. p = {'lr': (0.5, 5, 10), 'batch_size': (2, 30, 10), 'epochs': [150], 'optimizer': [Adam, Nadam, RMSprop]}

3. 
   def autoencoder(x_train, y_train, x_val, y_val, params):
     input_dim = X_train_0.shape[1] #num of predictor variables, 
     encoding_dim = 32
     hidden_dim = int(encoding_dim / 2)
     learning_rate = 1e-3
   
     input_layer = Input(shape=(input_dim, ))
     encoder = Dense(encoding_dim, activation="relu")(input_layer)
     encoder = Dense(hidden_dim, activation="relu")(encoder)
     decoder = Dense(hidden_dim, activation="relu")(encoder)
     decoder = Dense(encoding_dim, activation="relu")(decoder)
     decoder = Dense(input_dim, activation="linear")(decoder)
     autoencoder = Model(inputs=input_layer, outputs=decoder)
   
     autoencoder.compile(loss='mean_squared_error',
                 # here we add a regulizer normalization function from Talos
                 optimizer=params['optimizer'](lr=lr_normalizer(params['lr'],params['optimizer'])),
                 metrics=['acc'])
   
     history = autoencoder.fit(x_train, y_train, 
                       #validation_data=[x_val, y_val],
                       batch_size=params['batch_size'],
                       epochs=params['epochs'],
                       verbose=0)
   
     return history, autoencoder

[mikkokotila]

Have you tried to run your model without Talos? The kind of error you are getting is usually a result of error in the Keras model data inputs.

[kskaran94]

My Keras Model is working fine

[mikkokotila]

Can you share your Scan() command and the full trace for the issue.

[mikkokotila]

Closing this for inactivity. Feel free to reopen if still relevant.