MarkusSpanring

IWSLT datasets are not properly unpacked from `tgz` file

5

## 🐛 Bug IWSLT datasets are not properly unpacked from the downloaded `tgz` file when using `torchtext.datasets.IWSLT*`. When I unpack the nested `tgz` files by hand everything works as expected....

BAR1 memory of GPU is not released when main process is killed.

19

## 🐛 Bug When the main process receives a `SIGABRT`, `SIGTERM` or `SIGKILL` signal while iterating through a `DataLoader` the `BAR1 memory` of the GPU is not released. This happens...

As the title already suggests. I was looking into the latest examples of PyTorch with Ray Tune [here](https://docs.ray.io/en/latest/tune/examples/tune-pytorch-cifar.html#tune-pytorch-cifar-ref) and found that resuming from a checkpoint and reporting to tune is...

Deterministic mode is not set on remote worker

1

Fix related to #213 This PR should also fix an unreachable code segment introduced in my previous PR #208

## 🐛 Bug Deterministic mode is not set on all workers when `Trainer` is set to `deterministic=True`. ### To Reproduce The script is divided in two parts. `test.py` and `model.py`...

GPU memory not cleaned properly when using multiple workers in DataLoader

5

Hi all! When I try to use multiple workers in the `DataLoader` by specifying `num_workers` some of the processes stay alive after the run and occupy GPU memory. For my...