File size: 1,106 Bytes
8168661 112f924 8168661 844c8c5 8168661 112f924 8168661 83407cc 8168661 83407cc 8168661 112f924 8168661 112f924 8168661 66fa49f 8168661 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 |
# Whisper Finetuning script for the NST dataset
# This is a test script for XLA on TPU
#PT_XLA_DEBUG=1
python xla_spawn.py --num_cores=4 run_whisper_finetuning.py\
--model_name_or_path="openai/whisper-small" \
--output_dir="../whisper-NST-TPU-test4" \
--overwrite_output_dir=True \
--language="Norwegian" \
--task="transcribe" \
--dataset_name="NbAiLab/NST" \
--dataset_config="no-close" \
--do_train=True \
--do_eval=False \
--audio_column_name="audio" \
--text_column_name="text" \
--per_device_train_batch_size=16 \
--per_device_train_batch_size=16 \
--learning_rate=2e-5 \
--warmup_steps=0 \
--max_steps=20 \
--gradient_checkpointing=True \
--gradient_accumulation_steps=1 \
--group_by_length=False \
--evaluation_strategy="steps" \
--save_steps=20 \
--eval_steps=1 \
--max_eval_samples=2 \
--logging_steps=100 \
--load_best_model_at_end=False \
--metric_for_best_model="wer" \
--greater_is_better=False \
--report_to="tensorboard" \
--predict_with_generate=False \
--generation_max_length=225 \
--print_training_arguments=True \
--xla=True \
--push_to_hub=True
|