{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "97211288-7031-49f0-9916-844e0f1c7128",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "from transformers import BertTokenizer, BertModel\n",
    "from collections import OrderedDict\n",
    "import torch\n",
    "import random\n",
    "import pandas as pd\n",
    "from transformers import AutoModelForSequenceClassification,DataCollatorWithPadding\n",
    "from transformers import TrainingArguments, Trainer\n",
    "from sklearn.metrics import accuracy_score, precision_recall_fscore_support\n",
    "import sys, os\n",
    "from transformers import AutoModelForSequenceClassification, AutoTokenizer\n",
    "from transformers import pipeline\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "68632e9b-1933-4f62-a32e-d111dc2964bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/cache/comfyui/InstantID_wozai/cat_classification\n"
     ]
    }
   ],
   "source": [
    "home_path = os.getcwd()\n",
    "print(home_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "005d6545-4e86-496b-a61b-246c45a596f1",
   "metadata": {},
   "outputs": [],
   "source": [
    "model_path = f\"{home_path}/cate_calssification_en_thai_checkpoint_150320\"\n",
    "assert os.path.isdir(model_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "c80d4a18-be09-477c-b352-7fd382775523",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = AutoModelForSequenceClassification.from_pretrained(model_path)\n",
    "tokenizer = AutoTokenizer.from_pretrained(model_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "f6a3e36a-5284-4913-9b53-9e1ea5a05692",
   "metadata": {},
   "outputs": [],
   "source": [
    "val_df = pd.read_csv(f\"your_data.csv\")\n",
    "# 保存到文件\n",
    "with open(f\"{home_path}/label_name.json\", \"r\") as f:\n",
    "    cat_dic = json.load(f)\n",
    "label_name_dict = cat_dic[\"label_name_dict\"]\n",
    "name_label_dict = cat_dic[\"name_label_dict\"]\n",
    "classifier = pipeline(\"text-classification\", model=model, tokenizer=tokenizer, device=\"cuda:0\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "01613942-c304-4980-92bf-05b3da07fdae",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re\n",
    "def get_label(name):\n",
    "    label = classifier(name)[0][\"label\"].split('_')[1]\n",
    "    return label_name_dict[label]\n",
    "def is_english(text):\n",
    "    # 检查是否只包含英文字母和常见的标点符号、空格\n",
    "    try:\n",
    "        return bool(re.fullmatch(r'[A-Za-z\\s.,!?\\'\\\";:-]*', text[:4]))\n",
    "    except:\n",
    "        return False\n",
    "val_df[\"is_en\"] = val_df.pri_cate_name.apply(lambda x: is_english(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "9cd99694-d8c6-4a6e-a592-6444d7226bd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "en_df = val_df[val_df.is_en]\n",
    "th_df = val_df[~val_df.is_en]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "5f317115-278b-421d-b42c-c56d91064417",
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict_df(df):\n",
    "    df[\"predict_label\"] = df[\"product_name\"].apply(lambda x: get_label(x))\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8e673c24-aece-44a7-aff1-fa5de49cc8e2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# 预测数据的label\n",
    "en_df = predict_df(en_df)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "vllm_xtts",
   "language": "python",
   "name": "myenv"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.15"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}