Add files via upload

competition/科大讯飞AI开发者大赛2024/大模型能力评测中文成语释义与解析_baseline.ipynb

@@ -0,0 +1,171 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "714855ed-9040-44b1-a930-86edf0952277",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "4428241dd396428683892fe51ed5d438",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
+     ]
+    }
+   ],
+   "source": [
+    "from transformers import AutoModelForCausalLM, AutoTokenizer\n",
+    "device = \"cuda\" # the device to load the model onto\n",
+    "\n",
+    "model = AutoModelForCausalLM.from_pretrained(\n",
+    "    \"/home/lyz/hf-models/Qwen/Qwen1.5-4B-Chat/\",\n",
+    "    torch_dtype=\"auto\",\n",
+    "    device_map=\"auto\"\n",
+    ")\n",
+    "tokenizer = AutoTokenizer.from_pretrained(\"/home/lyz/hf-models/Qwen/Qwen1.5-4B-Chat/\")\n",
+    "\n",
+    "prompt = \"hello\"\n",
+    "messages = [\n",
+    "    {\"role\": \"system\", \"content\": \"You are a helpful assistant.\"},\n",
+    "    {\"role\": \"user\", \"content\": prompt}\n",
+    "]\n",
+    "text = tokenizer.apply_chat_template(\n",
+    "    messages,\n",
+    "    tokenize=False,\n",
+    "    add_generation_prompt=True\n",
+    ")\n",
+    "model_inputs = tokenizer([text], return_tensors=\"pt\").to(device)\n",
+    "\n",
+    "generated_ids = model.generate(\n",
+    "    model_inputs.input_ids,\n",
+    "    max_new_tokens=512\n",
+    ")\n",
+    "generated_ids = [\n",
+    "    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)\n",
+    "]\n",
+    "\n",
+    "response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "579d0f7f-a511-4d53-9b6c-a4cd1fcc2b87",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "test = pd.read_csv('./test_input.csv', header=None)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a06baea3-940e-4a14-885b-b8175075efdc",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for test_prompt in test[0].values:\n",
+    "    prompt = f\"列举与下面句子最相关的五个成语。只需要输出五个成语，不需要有其他的输出，写在一行中：{test_prompt}\"\n",
+    "\n",
+    "    words = ['同舟共济'] * 5\n",
+    "    for _ in range(10):\n",
+    "        messages = [\n",
+    "            {\"role\": \"user\", \"content\": prompt},\n",
+    "        ]\n",
+    "        text = tokenizer.apply_chat_template(\n",
+    "            messages,\n",
+    "            tokenize=False,\n",
+    "            add_generation_prompt=True\n",
+    "        )\n",
+    "        model_inputs = tokenizer([text], return_tensors=\"pt\").to(device)\n",
+    "        \n",
+    "        generated_ids = model.generate(\n",
+    "            model_inputs.input_ids,\n",
+    "            max_new_tokens=512\n",
+    "        )\n",
+    "        generated_ids = [\n",
+    "            output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)\n",
+    "        ]\n",
+    "        \n",
+    "        response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]\n",
+    "        response = response.replace('\\n', ' ').replace('、', ' ')\n",
+    "        words = [x for x in response.split() if len(x) == 4 and x.strip() != '']\n",
+    "        if len(words) == 5:\n",
+    "            break\n",
+    "\n",
+    "\n",
+    "    if len(' '.join(words).strip()) != 24:\n",
+    "        words = ['同舟共济'] * 5\n",
+    "\n",
+    "    with open('submit.csv', 'a+') as up:\n",
+    "        up.write(' '.join(words) + '\\n')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "9e852a9a-6aae-4bfa-b18b-030e593b1e77",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "24"
+      ]
+     },
+     "execution_count": 1,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "len('一模一样 如出一辙 千篇一律 大同小异 毫无二致')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "60d6f9b8-8f12-40bc-baca-27198dd4989d",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "py3.11",
+   "language": "python",
+   "name": "py3.11"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}