Owen Vallis
Owen Vallis
I think were just missing some type annotations. Looks like the static typing tests are failing for tensorflow_similarity/visualization/projector_v2/notebook.py:71
Sounds good, thanks!
Thanks for sharing the results. We're looking at making the storage and indexing layer more flexible and I'll be sure to include this in the changes.
Thanks Gautam, I've assigned the issue to you. Let me know if you have any questions.
Thanks, @sejalapeno, I've assigned the issue to you. Let me know if you have any questions.
For now, let's put it in the top-level example directory. This is similar to what we did for the sampler_io_cookbook, but will focus on demonstrating how to produce multiple augmented...
Hi @Lorenzobattistela, thanks for offering to pick this up! This is still currently open. The "initial" version I worked on is still in this [file](https://github.com/tensorflow/similarity/blob/new_losses/tensorflow_similarity/losses/lifted_struct_loss.py), but I never verified the...
Hi @Lorenzobattistela, I had another look at the paper and I think we can greatly simplify the implementation I shared. Here are a few thoughts: - Looks like the original...
Thanks for raising this, I'll take a look and see if I can repro the error on our side
is tf.data.AUTOTUNE a breaking import in 2.3? The official docs import as tf.data.AUTOTUNE for current versions of TF. See here https://www.tensorflow.org/guide/data_performance#prefetching