https-deeplearning-ai / tensorflow-2-public Goto Github PK

Here you create 2d tensors but it is not really necessary or wise imho.

It would be nice to know why you preferred rank-2 tensors there.

If you have Rank 1 tensors (vectors) use cons xs = tf.tensor1d(array) or in this case a very handy function is const xs = tf.range(0,n) for creating dummy data.

The C2 new folder rename has a space at the end causing failure to clone on Windows systems

Error:
error: unable to create file C2_Device-based-TF-lite /README.md: No such file or directory
fatal: cannot create directory at ‘C2_Device-based-TF-lite /W1’: No such file or directory
warning: Clone succeeded, but checkout failed.

Most resources file is deprecated such aaptOptions, jvm extensions files, org.tensorflow:tensorflow-lite:0.0.0-nightly & more, the app needed to rewrite to match current build requirement

For shuffling to have an effect dataset.shuffle(100) should be replaced by dataset = dataset.shuffle(100)

Link:

Pre optimized mobileNet - Please find the link to the Pre optimized mobileNet link here.

Not working when I access.