Welcome to Machine-Learning Discussions!

[haifaksh]

Discussed in #1

^{Originally posted by haifaksh October 13, 2022}
Hello,

How can I run run fairest-train on GPU for Mac M1 max ?

I tried the below code on Jupyter notebook but when I check the activity monitor, it is only using CPU and GPU

torch.device = "mps" !fairseq-train /Users/ha/data/preprocessed --max-epoch 7 \ --encoder-normalize-before --decoder-normalize-before \ --arch transformer --layernorm-embedding \ --task translation_multi_simple_epoch \ --sampling-method "temperature" \ --sampling-temperature 1.5 \ --encoder-langtok "src" \ --decoder-langtok \ --lang-pairs "$LANGPAIRS" \ --criterion label_smoothed_cross_entropy --label-smoothing 0.2 \ --optimizer adam --adam-eps 1e-06 --adam-betas '(0.9, 0.98)' \ --lr-scheduler inverse_sqrt --lr 3e-04 --warmup-updates 2500 --max-update 40000 \ --dropout 0.3 --attention-dropout 0.1 --weight-decay 0.0 \ --max-tokens 1024 --max-tokens-valid 1024 --update-freq 2 \ --save-interval 1 --save-interval-updates 5000 --keep-interval-updates 10 --no-epoch-checkpoints \ --seed 222 --log-format simple --log-interval 50 --ddp-backend=legacy_ddp