{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d67610d1-a6d9-420f-90d7-248d46b31697",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "loading file tokenizer.model from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/tokenizer.model\n",
      "loading file added_tokens.json from cache at None\n",
      "loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/special_tokens_map.json\n",
      "loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/tokenizer_config.json\n",
      "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/config.json\n",
      "loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/config.json\n",
      "Model config BaichuanConfig {\n",
      "  \"_from_model_config\": true,\n",
      "  \"_name_or_path\": \"baichuan-inc/Baichuan-13B-Base\",\n",
      "  \"architectures\": [\n",
      "    \"BaichuanForCausalLM\"\n",
      "  ],\n",
      "  \"auto_map\": {\n",
      "    \"AutoConfig\": \"baichuan-inc/Baichuan-13B-Base--configuration_baichuan.BaichuanConfig\",\n",
      "    \"AutoModelForCausalLM\": \"baichuan-inc/Baichuan-13B-Base--modeling_baichuan.BaichuanForCausalLM\"\n",
      "  },\n",
      "  \"bos_token_id\": 1,\n",
      "  \"eos_token_id\": 2,\n",
      "  \"gradient_checkpointing\": [\n",
      "    false\n",
      "  ],\n",
      "  \"hidden_act\": \"silu\",\n",
      "  \"hidden_size\": 5120,\n",
      "  \"initializer_range\": 0.02,\n",
      "  \"intermediate_size\": 13696,\n",
      "  \"model_max_length\": 4096,\n",
      "  \"model_type\": \"baichuan\",\n",
      "  \"num_attention_heads\": 40,\n",
      "  \"num_hidden_layers\": 40,\n",
      "  \"pad_token_id\": 0,\n",
      "  \"rms_norm_eps\": 1e-06,\n",
      "  \"tie_word_embeddings\": false,\n",
      "  \"torch_dtype\": \"bfloat16\",\n",
      "  \"transformers_version\": \"4.32.1\",\n",
      "  \"use_cache\": true,\n",
      "  \"vocab_size\": 64000\n",
      "}\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2023-08-31 19:08:00,343] [INFO] [real_accelerator.py:158:get_accelerator] Setting ds_accelerator to cuda (auto detect)\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "loading weights file pytorch_model.bin from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/pytorch_model.bin.index.json\n",
      "Generate config GenerationConfig {\n",
      "  \"_from_model_config\": true,\n",
      "  \"bos_token_id\": 1,\n",
      "  \"eos_token_id\": 2,\n",
      "  \"pad_token_id\": 0,\n",
      "  \"transformers_version\": \"4.32.1\"\n",
      "}\n",
      "\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "87e88732661e46cebbc8cc1ab0f93a77",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Loading checkpoint shards:   0%|          | 0/3 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "All model checkpoint weights were used when initializing BaichuanForCausalLM.\n",
      "\n",
      "All the weights of BaichuanForCausalLM were initialized from the model checkpoint at baichuan-inc/Baichuan-13B-Base.\n",
      "If your task is similar to the task the model of the checkpoint was trained on, you can already use BaichuanForCausalLM for predictions without further training.\n",
      "loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--baichuan-inc--Baichuan-13B-Base/snapshots/0ef0739c7bdd34df954003ef76d80f3dabca2ff9/generation_config.json\n",
      "Generate config GenerationConfig {\n",
      "  \"_from_model_config\": true,\n",
      "  \"bos_token_id\": 1,\n",
      "  \"eos_token_id\": 2,\n",
      "  \"pad_token_id\": 0,\n",
      "  \"transformers_version\": \"4.32.1\"\n",
      "}\n",
      "\n",
      "Generate config GenerationConfig {\n",
      "  \"_from_model_config\": true,\n",
      "  \"bos_token_id\": 1,\n",
      "  \"eos_token_id\": 2,\n",
      "  \"pad_token_id\": 0,\n",
      "  \"transformers_version\": \"4.32.1\"\n",
      "}\n",
      "\n",
      "/root/miniconda3/lib/python3.8/site-packages/transformers/generation/configuration_utils.py:367: UserWarning: `do_sample` is set to `False`. However, `top_p` is set to `0.92` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `top_p`.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "begin to gen\n"
     ]
    }
   ],
   "source": [
    "import transformers\n",
    "from transformers import AutoModelForCausalLM, AutoTokenizer\n",
    "\n",
    "transformers.logging.set_verbosity_info()\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"baichuan-inc/Baichuan-13B-Base\", trust_remote_code=True)\n",
    "model = AutoModelForCausalLM.from_pretrained(\"baichuan-inc/Baichuan-13B-Base\", device_map=\"auto\", trust_remote_code=True)\n",
    "inputs = tokenizer('make lyric:想带给你一句话', return_tensors='pt')\n",
    "inputs = inputs.to('cuda:0')\n",
    "print(\"begin to gen\")\n",
    "pred = model.generate(**inputs, max_new_tokens=64,repetition_penalty=1.1, top_p=0.92)\n",
    "print(tokenizer.decode(pred.cpu()[0], skip_special_tokens=True))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "91c204d0-b141-4dea-a468-8f74b1361782",
   "metadata": {},
   "outputs": [],
   "source": [
    "from peft import PeftModel\n",
    "lora_model = PeftModel.from_pretrained(model, \"./autodl-tmp/LLaMA-Efficient-Tuning/pppfuck\")\n",
    "lora_pred = lora_model.generate(**inputs, max_new_tokens=128,repetition_penalty=1.1)\n",
    "print(tokenizer.decode(lora_pred.cpu()[0], skip_special_tokens=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "84885258-cc93-4e08-a394-d2c5283a3fb3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.8.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}