You now can train a really good image classification model on 25k images in less than 2 hours, entirely for free, on a Colab GPU. And it's dead easy. What a time to be alive I remember when you had to write your own CUDA and C++ and you needed a hardware budget
-
-
I thought i tried TF 2.2.0 (the default from Colab right now), and it didn’t work. I will double check.
-
I found smthg perplexing about using my own http://tf.data .Dataset pipeline, but the rest pretty much the same as your notebook (on binary classification). I posted this Q and its resolution:https://stackoverflow.com/questions/62585490/as-list-is-not-defined-on-an-unknown-tensorshape-on-y-t-rank-leny-t-shape-a …
End of conversation
New conversation -
-
-
Thanks. I used “%pip install tf-nightly==2.3.0.dev20200620”, and it seems to be the last good build for this issue.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Update: it appears 2.5.0-dev20200626, this issue is resolved.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.