stale[bot]
commented
3 years ago This issue has been automatically marked as stale. If this issue is still affecting you, please leave any comment (for example, "bump"), and we'll keep it open. We are sorry that we haven't been able to prioritize it yet. If you have any new additional information, please include it with your comment!
❓ Questions and Help
Before asking:
What is your question?
Our machine is set with 8 GPUs. When I run the training script according to the IWSLT'14 German to English translation tutorial (fairseq/examples/translation) and display GPUs with command-line
nvidia-smi, I could identify that duplicate processes (but the usage is 0) are created as many as the number of GPUs on our machine (i.g. 8 processes per each GPU).I would like to know whether this is due to pytorch or GPU-related code in fairseq.
Thank you! :)
Code
``` fairseq-train \ data-bin/iwslt14.tokenized.de-en \ --arch transformer_iwslt_de_en --share-decoder-input-output-embed \ --optimizer adam --adam-betas '(0.9, 0.98)' --clip-norm 0.0 \ --lr 5e-4 --lr-scheduler inverse_sqrt --warmup-updates 4000 \ --dropout 0.3 --weight-decay 0.0001 \ --criterion label_smoothed_cross_entropy --label-smoothing 0.1 \ --max-tokens 4096 \ --eval-bleu \ --eval-bleu-args '{"beam": 5, "max_len_a": 1.2, "max_len_b": 10}' \ --eval-bleu-detok moses \ --eval-bleu-remove-bpe \ --eval-bleu-print-samples \ --best-checkpoint-metric bleu --maximize-best-checkpoint-metric ``` #### What have you tried? #### What's your environment? - fairseq Version (e.g., 1.0 or master): master - PyTorch Version (e.g., 1.0): 1.7.1 (install using pyenv) - OS (e.g., Linux): Linux 18.04 - How you installed fairseq (`pip`, source): pip install --editable ./ - Build command you used (if compiling from source): - Python version: 1.7.1 - CUDA/cuDNN version: 11.0/8.03 - GPU models and configuration: RTX2080Ti x 8ea - Any other relevant information: