Upload ms-swift/examples/notebook/qwen2_5-self-cognition/infer.ipynb with huggingface_hub
Browse files
ms-swift/examples/notebook/qwen2_5-self-cognition/infer.ipynb
ADDED
|
@@ -0,0 +1,148 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cells": [
|
| 3 |
+
{
|
| 4 |
+
"cell_type": "markdown",
|
| 5 |
+
"metadata": {},
|
| 6 |
+
"source": [
|
| 7 |
+
"## Inference\n",
|
| 8 |
+
"We have trained a well-trained checkpoint through the `self-cognition-sft.ipynb` tutorial, and here we use `PtEngine` to do the inference on it."
|
| 9 |
+
]
|
| 10 |
+
},
|
| 11 |
+
{
|
| 12 |
+
"cell_type": "code",
|
| 13 |
+
"execution_count": 6,
|
| 14 |
+
"metadata": {},
|
| 15 |
+
"outputs": [],
|
| 16 |
+
"source": [
|
| 17 |
+
"# import some libraries\n",
|
| 18 |
+
"import os\n",
|
| 19 |
+
"os.environ['CUDA_VISIBLE_DEVICES'] = '0'\n",
|
| 20 |
+
"\n",
|
| 21 |
+
"from swift.llm import InferEngine, InferRequest, PtEngine, RequestConfig, get_template"
|
| 22 |
+
]
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"cell_type": "code",
|
| 26 |
+
"execution_count": 7,
|
| 27 |
+
"metadata": {},
|
| 28 |
+
"outputs": [],
|
| 29 |
+
"source": [
|
| 30 |
+
"# Hyperparameters for inference\n",
|
| 31 |
+
"last_model_checkpoint = 'output/checkpoint-xxx'\n",
|
| 32 |
+
"\n",
|
| 33 |
+
"# model\n",
|
| 34 |
+
"model_id_or_path = 'Qwen/Qwen2.5-3B-Instruct' # model_id or model_path\n",
|
| 35 |
+
"system = 'You are a helpful assistant.'\n",
|
| 36 |
+
"infer_backend = 'pt'\n",
|
| 37 |
+
"\n",
|
| 38 |
+
"# generation_config\n",
|
| 39 |
+
"max_new_tokens = 512\n",
|
| 40 |
+
"temperature = 0\n",
|
| 41 |
+
"stream = True"
|
| 42 |
+
]
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"cell_type": "code",
|
| 46 |
+
"execution_count": null,
|
| 47 |
+
"metadata": {},
|
| 48 |
+
"outputs": [],
|
| 49 |
+
"source": [
|
| 50 |
+
"# Get model and template, and load LoRA weights.\n",
|
| 51 |
+
"engine = PtEngine(model_id_or_path, adapters=[last_model_checkpoint])\n",
|
| 52 |
+
"template = get_template(engine.model_meta.template, engine.tokenizer, default_system=system)\n",
|
| 53 |
+
"# You can modify the `default_template` directly here, or pass it in during `engine.infer`.\n",
|
| 54 |
+
"engine.default_template = template"
|
| 55 |
+
]
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"cell_type": "code",
|
| 59 |
+
"execution_count": 11,
|
| 60 |
+
"metadata": {},
|
| 61 |
+
"outputs": [
|
| 62 |
+
{
|
| 63 |
+
"name": "stdout",
|
| 64 |
+
"output_type": "stream",
|
| 65 |
+
"text": [
|
| 66 |
+
"query: who are you?\n",
|
| 67 |
+
"response: I am an artificial intelligence language model named Xiao Huang, developed by ModelScope. I can answer various questions and engage in conversation with humans. If you have any questions or need help, feel free to ask me at any time.\n",
|
| 68 |
+
"--------------------------------------------------\n",
|
| 69 |
+
"query: What should I do if I can't sleep at night?\n",
|
| 70 |
+
"response: If you're having trouble sleeping, there are several things you can try:\n",
|
| 71 |
+
"\n",
|
| 72 |
+
"1. Establish a regular sleep schedule: Try to go to bed and wake up at the same time every day, even on weekends.\n",
|
| 73 |
+
"\n",
|
| 74 |
+
"2. Create a relaxing bedtime routine: Engage in calming activities before bed, such as reading a book or taking a warm bath.\n",
|
| 75 |
+
"\n",
|
| 76 |
+
"3. Make your bedroom conducive to sleep: Keep your bedroom cool, dark, and quiet. Invest in comfortable bedding and pillows.\n",
|
| 77 |
+
"\n",
|
| 78 |
+
"4. Avoid stimulating activities before bed: Avoid using electronic devices, watching TV, or engaging in mentally stimulating activities before bed.\n",
|
| 79 |
+
"\n",
|
| 80 |
+
"5. Exercise regularly: Regular physical activity can help improve your sleep quality, but avoid exercising too close to bedtime.\n",
|
| 81 |
+
"\n",
|
| 82 |
+
"6. Manage stress: Practice relaxation techniques, such as deep breathing, meditation, or yoga, to help manage stress and promote better sleep.\n",
|
| 83 |
+
"\n",
|
| 84 |
+
"7. Limit caffeine and alcohol intake: Both caffeine and alcohol can disrupt sleep patterns, so it's best to limit their consumption, especially in the evening.\n",
|
| 85 |
+
"\n",
|
| 86 |
+
"8. Seek professional help: If you continue to have difficulty sleeping despite trying these strategies, consider seeking help from a healthcare provider or a sleep specialist.\n",
|
| 87 |
+
"--------------------------------------------------\n",
|
| 88 |
+
"query: 你是谁训练的?\n",
|
| 89 |
+
"response: 我是由魔搭团队训练和开发的。\n",
|
| 90 |
+
"--------------------------------------------------\n"
|
| 91 |
+
]
|
| 92 |
+
}
|
| 93 |
+
],
|
| 94 |
+
"source": [
|
| 95 |
+
"query_list = [\n",
|
| 96 |
+
" 'who are you?',\n",
|
| 97 |
+
" \"What should I do if I can't sleep at night?\",\n",
|
| 98 |
+
" '你是谁训练的?',\n",
|
| 99 |
+
"]\n",
|
| 100 |
+
"\n",
|
| 101 |
+
"def infer_stream(engine: InferEngine, infer_request: InferRequest):\n",
|
| 102 |
+
" request_config = RequestConfig(max_tokens=max_new_tokens, temperature=temperature, stream=True)\n",
|
| 103 |
+
" gen_list = engine.infer([infer_request], request_config)\n",
|
| 104 |
+
" query = infer_request.messages[0]['content']\n",
|
| 105 |
+
" print(f'query: {query}\\nresponse: ', end='')\n",
|
| 106 |
+
" for resp in gen_list[0]:\n",
|
| 107 |
+
" if resp is None:\n",
|
| 108 |
+
" continue\n",
|
| 109 |
+
" print(resp.choices[0].delta.content, end='', flush=True)\n",
|
| 110 |
+
" print()\n",
|
| 111 |
+
"\n",
|
| 112 |
+
"def infer(engine: InferEngine, infer_request: InferRequest):\n",
|
| 113 |
+
" request_config = RequestConfig(max_tokens=max_new_tokens, temperature=temperature)\n",
|
| 114 |
+
" resp_list = engine.infer([infer_request], request_config)\n",
|
| 115 |
+
" query = infer_request.messages[0]['content']\n",
|
| 116 |
+
" response = resp_list[0].choices[0].message.content\n",
|
| 117 |
+
" print(f'query: {query}')\n",
|
| 118 |
+
" print(f'response: {response}')\n",
|
| 119 |
+
"\n",
|
| 120 |
+
"infer_func = infer_stream if stream else infer\n",
|
| 121 |
+
"for query in query_list:\n",
|
| 122 |
+
" infer_func(engine, InferRequest(messages=[{'role': 'user', 'content': query}]))\n",
|
| 123 |
+
" print('-' * 50)"
|
| 124 |
+
]
|
| 125 |
+
}
|
| 126 |
+
],
|
| 127 |
+
"metadata": {
|
| 128 |
+
"kernelspec": {
|
| 129 |
+
"display_name": "test_py310",
|
| 130 |
+
"language": "python",
|
| 131 |
+
"name": "python3"
|
| 132 |
+
},
|
| 133 |
+
"language_info": {
|
| 134 |
+
"codemirror_mode": {
|
| 135 |
+
"name": "ipython",
|
| 136 |
+
"version": 3
|
| 137 |
+
},
|
| 138 |
+
"file_extension": ".py",
|
| 139 |
+
"mimetype": "text/x-python",
|
| 140 |
+
"name": "python",
|
| 141 |
+
"nbconvert_exporter": "python",
|
| 142 |
+
"pygments_lexer": "ipython3",
|
| 143 |
+
"version": "3.10.15"
|
| 144 |
+
}
|
| 145 |
+
},
|
| 146 |
+
"nbformat": 4,
|
| 147 |
+
"nbformat_minor": 2
|
| 148 |
+
}
|