examples
examples copied to clipboard
pytorch
A set of examples around pytorch in Vision, Text, Reinforcement Learning, etc.
the random seed has to be set in the main_worker, not in the `def main()`. I found that although the seed is set in `def main()`, each process in distributed...
My machine: 2 machines with different ips and 2 available Gpus on each machine When I use the multigpu_torchrun.py example, when I pass these two directives: `torchrun --nproc_per_node=2 --nnodes=2 --node_rank=0...
Hi, I'm using the tutorial [https://github.com/pytorch/tutorials/blob/master/intermediate_source/ddp_tutorial.rst](url) for DDP train,using 4 gpus in myself code, reference Basic Use Case. But when I finished the modification, it was stuck during run the...
Your issue may already be reported! Please search on the [issue tracker](https://github.com/pytorch/examples/issues) before creating one. ## Context I am running ``` python neural_style/neural_style.py train --dataset "path" --style-image "image-path" --save-model-dir "path"...
## Context * Pytorch version: libtorch-macos-2.0.1.zip * Operating System and version: macos 13.4.1 (c) ## Your Environment * Installed using source? [yes/no]: no * Are you planning to deploy it...
## 📚 Documentation @HamidShojanazeri, I'm following your [FSDP example](https://github.com/pytorch/examples/tree/main/distributed/FSDP) and swapped in a bigger model, `google/flan-t5-xxl`, and am a little unclear on what happens when the script starts up. I'm...
the absolute position embedding used in [examples](https://github.com/pytorch/examples/tree/main)/[vision_transformer](https://github.com/pytorch/examples/tree/main/vision_transformer)/main.py seemed to be **incorrect**: ```python # Positional embedding self.pos_embedding = nn.Parameter(torch.randn(self.batch_size, 1, self.latent_size)).to(self.device) ``` which should look like this ```python self.pos_embedding = nn.Parameter(torch.randn(1,...
Hi, thanks for the great tutorial on language modeling. A question on [masking the input](https://github.com/pytorch/examples/blob/7f7c222b355abd19ba03a7d4ba90f1092973cdbc/word_language_model/model.py#L128): Why do we mask the input in the encoder layer? I'm aware that the mask...
Fixes #816 Resume no longer overwrites learning rate specific on command line. (My linter also removed 2 extraneous spaces in a `dict`, I can revert this if desired 🙂)
