Bobby V

hi, did you figure it out? It would be great if you could share some insights.

May you please include some photos to articulate your points?

Seems like the Upsampling2D is the standard upscale technique across models in this repository. For example, in the Unet architecture, Upsampling2D is employed instead of Transposed convolution. You may refer...

Can you confirm the following steps? 1/ The colab is running, and the port open messages are generated on Colab? 2/ The bash command on MacOS completed successfully? Can you...

I don't think it is a Catalina related issue, the problem is observed with other os as well. You are really close to the end result... https://github.com/alievk/avatarify/issues/193

I can confirm that this is a good enhancement: Infact, just change the "requirements_client.txt" value, the bash script will automatically install the indicated version, 3.4.5.20

> Hi, @Paulex-Git! > What version of Mac OSX do you have? If Catalina [this link](https://github.com/AlexEMG/DeepLabCut/issues/608) can help. Upgrading to Catalina will resolve the issue. Though it could be a...

If you run the code with standard Hparam, you will get >20 Bleu score python -m nmt.nmt \ --src=vi --tgt=en \ --hparams_path=nmt/standard_hparams/iwslt15.json \ --vocab_prefix=/tmp/nmt_data/vocab \ --train_prefix=/tmp/nmt_data/train \ --dev_prefix=/tmp/nmt_data/tst2012 \ --test_prefix=/tmp/nmt_data/tst2013...