Spaces:
Sleeping
Sleeping
File size: 7,659 Bytes
6eb2110 |
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 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 |
{
"cells": [
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-25T14:59:03.066893917Z",
"start_time": "2023-11-25T14:59:02.924638197Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/home/z/miniconda3/envs/llama/lib/python3.10/site-packages/transformers/models/auto/auto_factory.py:472: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.\n",
" warnings.warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6745f1964cda44068721c6c8b5f91eee",
"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": [
"/home/z/miniconda3/envs/llama/lib/python3.10/site-packages/transformers/utils/hub.py:374: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.\n",
" warnings.warn(\n"
]
}
],
"source": [
"import torch\n",
"from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig\n",
"\n",
"# Define the base model ID\n",
"base_model_id = \"meta-llama/Llama-2-13b-hf\"\n",
"\n",
"# Create a BitsAndBytesConfig object with the corrected settings\n",
"quantization_config = BitsAndBytesConfig(\n",
" load_in_4bit=True,\n",
" bnb_4bit_use_double_quant=True,\n",
" bnb_4bit_quant_type=\"nf4\",\n",
" bnb_4bit_compute_dtype=torch.bfloat16,\n",
" load_in_8bit_fp32_cpu_offload=True # Set as suggested in the error\n",
")\n",
"\n",
"# Load the base model with the updated quantization configuration\n",
"# Adjust 'device_map' based on your system's GPU configuration\n",
"base_model = AutoModelForCausalLM.from_pretrained(\n",
" base_model_id, \n",
" quantization_config=quantization_config,\n",
" trust_remote_code=True,\n",
" use_auth_token=True\n",
")\n",
"\n",
"# Load the tokenizer\n",
"tokenizer = AutoTokenizer.from_pretrained(base_model_id, add_bos_token=True, trust_remote_code=True)\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "_BxOhAiqyRgp"
},
"source": [
"Now load the QLoRA adapter from the appropriate checkpoint directory, i.e. the best performing model checkpoint:"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"ExecuteTime": {
"end_time": "2023-11-25T14:59:12.830783738Z",
"start_time": "2023-11-25T14:59:12.826615170Z"
},
"id": "GwsiqhWuyRgp"
},
"outputs": [],
"source": [
"from peft import PeftModel\n",
"\n",
"ft_model = PeftModel.from_pretrained(base_model, \"checkpoint-2800\")"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"from datasets import load_dataset\n",
"\n",
" \n",
"eval_dataset = load_dataset('json', data_files='/home/z/Music/LLAMA/llama/IPG/datasets/new_test_data.json', split='train')\n",
"\n",
"\n",
"def formatting_func(example):\n",
" text = f\"### The job description: {example['text']}\\n ### The skills: \"\n",
" return text\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"\n",
"\n",
"def run_finetune_model(model_id):\n",
"\n",
" example = eval_dataset.filter(lambda x: x['id'] == model_id)[0]\n",
" formatted_text = formatting_func(example)\n",
" \n",
" #print(formatted_text)\n",
" model_input = tokenizer(formatted_text, return_tensors=\"pt\").to(\"cuda\")\n",
"\n",
"\n",
" ft_model.eval()\n",
" with torch.no_grad():\n",
" output_tokens = ft_model.generate(**model_input, max_new_tokens=200)[0]\n",
" generated_text = tokenizer.decode(output_tokens, skip_special_tokens=True)\n",
" \n",
" print(generated_text)\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"### The job description: German BD Manager\n",
"Job Description：\n",
"1、Represent the company to develop new partners for energy storage system；\n",
"2、Maintain good relationship and help partners to develop/grow the business；\n",
"3、Formulate a strategy and target for the market exploration so as to achieve good performance；\n",
"4、Pay attention and collect information for the latest development/tendency in the industry as well as getting feedback/insight to R&D；\n",
"5、Advice and assist the company to build a strong local team including but not limited to after sale service, technical support, sales and marketing.\n",
" \n",
"Job Requirements：\n",
"1、Fluent in English and German；\n",
"2、5+ years of experience in the industry of Energy Storage System, a good education background will be preferential；\n",
"3、Strong execution and result-oriented, attach importance to details and critical thinking as well as desire to progress/evolve；\n",
"4、Open-minded and teamwork, great skills in communication.\n",
" ### The skills: ['programming', 'simulation', 'communication', 'excel', 'word', 'powerpoint', 'marketing', 'c++', 'matlab', 'html', 'data analysis', 'powerpoint', 'communication', 'project management', 'excel', 'microsoft office', 'tableau', 'powerpoint', 'word', 'microsoft office', 'communication', 'python', 'excel', 'microsoft office', 'c++', 'python', 'data analysis', 'python', 'html', 'data analysis', 'communication', 'microsoft office', 'java', 'powerpoint']\n",
" ### The qualifications: \n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n",
"\n"
]
}
],
"source": [
"run_finetune_model(\"19010\")\n"
]
}
],
"metadata": {
"accelerator": "GPU",
"colab": {
"gpuType": "T4",
"provenance": []
},
"gpuClass": "standard",
"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": "3.10.13"
}
},
"nbformat": 4,
"nbformat_minor": 4
}
|