File size: 1,160 Bytes
3932b77
 
 
 
 
 
 
 
 
 
929cce7
 
 
3932b77
929cce7
3932b77
 
929cce7
3932b77
929cce7
3932b77
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
deepspeed run_speech_recognition_seq2seq_streaming.py \
	--deepspeed="ds_config.json" \
	--model_name_or_path="openai/whisper-large-v2" \
	--dataset_name="NbAiLab/NCC_S" \
	--language="Norwegian" \
	--train_split_name="train" \
	--eval_split_name="validation" \
	--model_index_name="Whisper Large V2 Norwegian" \
	--max_steps="50000" \
	--output_dir="./" \
	--per_device_train_batch_size="64" \
	--per_device_eval_batch_size="32" \
	--gradient_accumulation_steps="2" \
	--logging_steps="25" \
	--learning_rate="3e-6" \
	--warmup_steps="500" \
	--evaluation_strategy="steps" \
	--eval_steps="100" \
	--save_strategy="steps" \
	--save_steps="100" \
	--generation_max_length="225" \
	--length_column_name="duration" \
	--max_duration_in_seconds="30" \
	--text_column_name="text" \
	--freeze_feature_encoder="False" \
	--report_to="tensorboard" \
	--metric_for_best_model="wer" \
	--greater_is_better="False" \
	--load_best_model_at_end \
	--gradient_checkpointing \
	--fp16 \
	--overwrite_output_dir \
	--do_train \
	--do_eval \
	--predict_with_generate \
	--do_normalize_eval \
	--use_auth_token \
	--push_to_hub \
	--hub_model_id="NbAiLab/whisper-large-v2-nob"