Imahn Shekhzadeh

Status today, with torchtext version 0.15.2+cpu it is still not possible to download the IWSLT datasets. ``` requests.exceptions.HTTPError: 404 Client Error: Not Found for url: https://drive.google.com/uc?id=12ycYSzLIG253AFN35Y6qoyf9wtkOjakp This exception is thrown...

I have the same issue. @bintoro If you ask me, this project should be archived, since it's not in active development and does not work for the newest versions of...

> [...] and the data are sampled using a TasksetDataset First of all note that there are only the classes `Taskset` and `TaskDataset`, with the latter being deprecated in favor...

In general, it is much easier to read code when it is properly formatted in the text msg, take a look here: https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/creating-and-highlighting-code-blocks :slightly_smiling_face: You might be using a wrong...

> However, I'm receiving the same adapt_loss and loss at every timestep. This is weird, indeed you should receive different loss values. Maybe you can `proposal` by `learner()` everywhere and...

> @ImahnShekhzadeh can you rename this file train_cifar10_ddp.py please? and re-add the previous file? Thanks Done

Hi @Fabian-Unruh, Yes, you are right that this happens because of numerical imprecisions. I suspect that you are using `float32` as `dtype`, which is the default in PyTorch? When I...