Hello!
I wonder if the ungraded lab is broken. The first runnable cell gives error:
ERROR: Could not find a version that satisfies the requirement tensorflow==2.4.3 (from versions: 2.8.0rc0, 2.8.0rc1, 2.8.0, 2.8.1, 2.8.2, 2.8.3, 2.8.4, 2.9.0rc0, 2.9.0rc1, 2.9.0rc2, 2.9.0, 2.9.1, 2.9.2, 2.9.3, 2.10.0rc0, 2.10.0rc1, 2.10.0rc2, 2.10.0rc3, 2.10.0, 2.10.1, 2.11.0rc0, 2.11.0rc1, 2.11.0rc2, 2.11.0, 2.11.1, 2.12.0rc0, 2.12.0rc1, 2.12.0, 2.12.1, 2.13.0rc0, 2.13.0rc1, 2.13.0rc2, 2.13.0, 2.14.0rc0)
ERROR: No matching distribution found for tensorflow==2.4.3
After that, it is pretty much downhill with logs such as “keras is no longer supported, please use tf.keras instead.
Your TensorFlow version is newer than 2.4.0 and so graph support has been removed in eager mode and some static graphs may not be supported. See PR #1483 for discussion.” when creating the deepexplainer.
Later, computing the shap values behaves erraticly. Sometimes it logs “tf.keras.backend.set_learning_phase is deprecated and will be removed after 2020-10-11. To update it, simply pass a True/False value to the training argument of the __call__ method of your layer or model.”, sometimes it throws an error, and most of the time, the plots of the shap values do not look like something I think they should. One time the highlights can only seen on one example, another time they are only on the diagonal, and very rarely can they be seen on multiple categories for a single input. This is especially noteworthy as the text suggests to take a look on the coat and see if also the shirt and the pullover are highlighted. The attached image below is the best one I got, and yes, there are some faint highlights on the shirt when the coat is inputted, but none on the pullover as far as I can see.
Am I just not getting something, or does the initial error break the whole thing?
