From 774ed4a027a43e705699f78f5549148062c15173 Mon Sep 17 00:00:00 2001 From: flozi00 Date: Tue, 4 Jan 2022 12:59:20 +0100 Subject: [PATCH] Fix Code block (#14983) --- examples/pytorch/speech-pretraining/README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/examples/pytorch/speech-pretraining/README.md b/examples/pytorch/speech-pretraining/README.md index ecbbce9f2b..3c15bcc3f7 100644 --- a/examples/pytorch/speech-pretraining/README.md +++ b/examples/pytorch/speech-pretraining/README.md @@ -88,6 +88,7 @@ The results of this run can be seen [here](https://wandb.ai/patrickvonplaten/wav To pre-train `"base-sized"` Wav2Vec2 model, *e.g.* [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on [librispeech_asr](https://huggingface.co/datasets/librispeech_asr), the following command can be run: +```bash accelerate launch run_wav2vec2_pretraining_no_trainer.py \ --dataset_name=librispeech_asr \ --dataset_config_names clean clean other \ @@ -109,6 +110,7 @@ accelerate launch run_wav2vec2_pretraining_no_trainer.py \ --adam_beta2="0.98" \ --adam_epsilon="1e-06" \ --gradient_checkpointing \ +``` The experiment was run on 8 GPU V100 (16 GB RAM each) for 4 days. In case you have more than 8 GPUs available for a higher effective `batch_size`,