Mobilenet v2 issues

Course Q&A Deep Learning Specialization Convolutional Neural Networks
4

Here’s the template code that was given to you in alpaca_model:

base_model = tf.keras.applications.MobileNetV2(input_shape=None,
                                                   include_top=None, # <== Important!!!!
                                                   weights=base_model_path)

When you filled in the required parts, you also changed the close paren to be the equivalent of this (without your other changes):

base_model = tf.keras.applications.MobileNetV2(input_shape=None,
                                                   include_top=None, # <== Important!!!!
                                                   weights=base_model_path
)

That is legal python syntax, but it confuses the grader for some reason.

Also note that you added a bunch of unnecessary “import” statements. I don’t think those cause any problems, but they are a mistake. If something shows as not imported, that means you forgot to run the earlier cells in the notebook. You need to do that every time you restart or reopen the notebook.

1 Like