{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "UsageError: Line magic function `%huggingface-cli` not found.\n"
     ]
    }
   ],
   "source": [
    "%huggingface-cli repo create whisper-small-ko"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "echo 'python run_speech_recognition_seq2seq_streaming.py \\\n",
    "\t--model_name_or_path=\"openai/whisper-small\" \\\n",
    "\t--dataset_name=\"kresnik/zeroth_korean\" \\\n",
    "\t--dataset_config_name=\"ko\" \\\n",
    "\t--language=\"korean\" \\\n",
    "\t--train_split_name=\"train+validation\" \\\n",
    "\t--eval_split_name=\"test\" \\\n",
    "\t--model_index_name=\"Whisper Small Korean\" \\\n",
    "\t--max_steps=\"4000\" \\\n",
    "\t--output_dir=\"./\" \\\n",
    "\t--per_device_train_batch_size=\"64\" \\\n",
    "\t--per_device_eval_batch_size=\"32\" \\\n",
    "\t--logging_steps=\"25\" \\\n",
    "\t--learning_rate=\"1e-5\" \\\n",
    "\t--warmup_steps=\"500\" \\\n",
    "\t--evaluation_strategy=\"steps\" \\\n",
    "\t--eval_steps=\"1000\" \\\n",
    "\t--save_strategy=\"steps\" \\\n",
    "\t--save_steps=\"1000\" \\\n",
    "\t--generation_max_length=\"225\" \\\n",
    "\t--length_column_name=\"input_length\" \\\n",
    "\t--max_duration_in_seconds=\"30\" \\\n",
    "\t--text_column_name=\"sentence\" \\\n",
    "\t--freeze_feature_encoder=\"False\" \\\n",
    "\t--report_to=\"tensorboard\" \\\n",
    "\t--metric_for_best_model=\"wer\" \\\n",
    "\t--greater_is_better=\"False\" \\\n",
    "\t--load_best_model_at_end \\\n",
    "\t--gradient_checkpointing \\\n",
    "\t--fp16 \\\n",
    "\t--overwrite_output_dir \\\n",
    "\t--do_train \\\n",
    "\t--do_eval \\\n",
    "\t--predict_with_generate \\\n",
    "\t--do_normalize_eval \\\n",
    "\t--streaming \\\n",
    "\t--use_auth_token \\\n",
    "\t--push_to_hub' >> run.sh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "stt",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}