CareGPT
1.0.0
中文 ｜ English
影片教學安裝部署線上體驗
⚡特性：
conda create - n llm python = 3.11
conda activate llm
python - m pip install - r requirements . txt # 转为HF格式
python - m transformers . models . llama . convert_llama_weights_to_hf
- - input_dir path_to_llama_weights - - model_size 7 B - - output_dir path_to_llama_model如果您使用自訂資料集，請務必在dataset_info.json檔案中以下列格式提供您的資料集定義。
"数据集名称" : {
"hf_hub_url" : " HuggingFace上的项目地址（若指定，则忽略下列三个参数） " ,
"script_url" : "包含数据加载脚本的本地文件夹名称（若指定，则忽略下列两个参数） " ,
"file_name" : "该目录下数据集文件的名称（若上述参数未指定，则此项必需） " ,
"file_sha1" : "数据集文件的SHA-1哈希值（可选） " ,
"columns" : {
"prompt" : "数据集代表提示词的表头名称（默认：instruction） " ,
"query" : "数据集代表请求的表头名称（默认：input） " ,
"response" : "数据集代表回答的表头名称（默认：output） " ,
"history" : "数据集代表历史对话的表头名称（默认：None） "
}
}其中prompt和response列應是非空的字串。 query列的內容將會和prompt列拼接作為模型輸入。 history列應為一個列表，其中每個元素是一個字串二元組，分別代表使用者請求和模型答案。
.txt格式，一行一個無監督資料。
Machine learning (ML) is a field devoted to understanding and building methods that let machines "learn" – that is, methods that leverage data to improve computer performance on some set of tasks.
Machine learning algorithms build a model based on sample data, known as training data, in order to make predictions or decisions without being explicitly programmed to do so. Machine learning algorithms are used in a wide variety of applications, such as in medicine, email filtering, speech recognition, agriculture, and computer vision, where it is difficult or unfeasible to develop conventional algorithms to perform the needed tasks.[
{
"instruction" : "听起来很不错。人工智能可能在哪些方面面临挑战呢？ " ,
"input" : " " ,
"output" : "人工智能面临的挑战包括数据隐私、安全和道德方面的问题，以及影响就业机会的自动化等问题。 " ,
"history" : [
[ "你好，你能帮我解答一个问题吗？ " , "当然，请问有什么问题？ " ],
[ "我想了解人工智能的未来发展方向，你有什么想法吗？ " , "人工智能在未来的发展方向可能包括更强大的机器学习算法，更先进的自然语言处理技术，以及更加智能的机器人。 " ]
]
}
][
{
"instruction" : "听起来很不错。人工智能可能在哪些方面面临挑战呢？ " ,
"input" : " " ,
"output" : "人工智能面临的挑战包括数据隐私、安全和道德方面的问题，以及影响就业机会的自动化等问题。 " ,
"history" : []
}
][
{
"instruction" : "生成三个与“道歉”意思相同的动词" ,
"input" : " " ,
"output" : [
"承认，表示遗憾，弥补。 " ,
"道歉"
]
}
]看看你的顯示卡是否為NVLINK連接，NVLINK連接才能有效使用accelerate進行平行加速訓練。
nvidia-smi topo -m
accelerate config # configure the environment
accelerate launch src / train_bash . py # arguments (same as above) # LLaMA-2
accelerate launch src / train_bash . py
- - stage sft
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - do_train
- - dataset mm
- - finetuning_type lora
- - quantization_bit 4
- - overwrite_cache
- - output_dir output
- - per_device_train_batch_size 8
- - gradient_accumulation_steps 4
- - lr_scheduler_type cosine
- - logging_steps 10
- - save_steps 1000
- - learning_rate 5e-5
- - num_train_epochs 2.0
- - plot_loss
- - fp16
- - template llama2
- - lora_target q_proj , v_proj
# LLaMA
accelerate launch src / train_bash . py
- - stage sft
- - model_name_or_path . / Llama - 7 b - hf
- - do_train
- - dataset mm , hm
- - finetuning_type lora
- - overwrite_cache
- - output_dir output - 1
- - per_device_train_batch_size 4
- - gradient_accumulation_steps 4
- - lr_scheduler_type cosine
- - logging_steps 10
- - save_steps 2000
- - learning_rate 5e-5
- - num_train_epochs 2.0
- - plot_loss
- - fp16
- - template default
- - lora_target q_proj , v_proj # LLaMA-2, DPO
accelerate launch src / train_bash . py
- - stage dpo
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - do_train
- - dataset rlhf
- - template llama2
- - finetuning_type lora
- - quantization_bit 4
- - lora_target q_proj , v_proj
- - resume_lora_training False
- - checkpoint_dir . / output - 2
- - output_dir output - dpo
- - per_device_train_batch_size 2
- - gradient_accumulation_steps 4
- - lr_scheduler_type cosine
- - logging_steps 10
- - save_steps 1000
- - learning_rate 1e-5
- - num_train_epochs 1.0
- - plot_loss
- - fp16 # LLaMA-2
python src / web_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output
- - finetuning_type lora
- - template llama2
# LLaMA
python src / web_demo . py
- - model_name_or_path . / Llama - 7 b - hf
- - checkpoint_dir output - 1
- - finetuning_type lora
- - template default
# DPO
python src / web_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output - dpo
- - finetuning_type lora
- - template llama2 # LLaMA-2
python src / api_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output
- - finetuning_type lora
- - template llama2
# LLaMA
python src / api_demo . py
- - model_name_or_path . / Llama - 7 b - hf
- - checkpoint_dir output - 1
- - finetuning_type lora
- - template default
# DPO
python src / api_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output - dpo
- - finetuning_type lora
- - template llama2測試API：
curl - X 'POST'
'http://127.0.0.1:8888/v1/chat/completions'
- H 'accept: application/json'
- H 'Content-Type: application/json'
- d ' {
"model" : "string",
"messages": [
{
"role" : "user",
"content": "你好"
}
],
" temperature ": 0 ,
"top_p" : 0 ,
"max_new_tokens" : 0 ,
"stream" : false
}' # LLaMA-2
python src / cli_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output
- - finetuning_type lora
- - template llama2
# LLaMA
python src / cli_demo . py
- - model_name_or_path . / Llama - 7 b - hf
- - checkpoint_dir output - 1
- - finetuning_type lora
- - template default
# DPO
python src / cli_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output - dpo
- - finetuning_type lora
- - template llama2 # LLaMA-2
CUDA_VISIBLE_DEVICES = 0 python src / train_bash . py
- - stage sft
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - do_predict
- - dataset mm
- - template llama2
- - finetuning_type lora
- - checkpoint_dir output
- - output_dir predict_output
- - per_device_eval_batch_size 8
- - max_samples 100
- - predict_with_generate
# LLaMA
CUDA_VISIBLE_DEVICES = 0 python src / train_bash . py
- - stage sft
- - model_name_or_path . / Llama - 7 b - hf
- - do_predict
- - dataset mm
- - template default
- - finetuning_type lora
- - checkpoint_dir output - 1
- - output_dir predict_output
- - per_device_eval_batch_size 8
- - max_samples 100
- - predict_with_generate # LLaMA-2
CUDA_VISIBLE_DEVICES = 0 python src / train_bash . py
- - stage sft
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - do_eval
- - dataset mm
- - template llama2
- - finetuning_type lora
- - checkpoint_dir output
- - output_dir eval_output
- - per_device_eval_batch_size 8
- - max_samples 100
- - predict_with_generate
# LLaMA
CUDA_VISIBLE_DEVICES = 0 python src / train_bash . py
- - stage sft
- - model_name_or_path . / Llama - 7 b - hf
- - do_eval
- - dataset mm
- - template default
- - finetuning_type lora
- - checkpoint_dir output - 1
- - output_dir eval_output
- - per_device_eval_batch_size 8
- - max_samples 100
- - predict_with_generate在4/8-bit評估時，建議使用--per_device_eval_batch_size=1和--max_target_length 128
# LLaMA-2
python src / export_model . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - template llama2
- - finetuning_type lora
- - checkpoint_dir output - 1
- - output_dir output_export
# LLaMA
python src / export_model . py
- - model_name_or_path . / Llama - 7 b - hf
- - template default
- - finetuning_type lora
- - checkpoint_dir output
- - output_dir output_export % cd Gradio
python app . py 
# LLaMA-2
python src / api_demo . py
- - model_name_or_path . / Llama - 2 - 7 b - chat - hf
- - checkpoint_dir output
- - finetuning_type lora
- - template llama2
# LLaMA
python src / api_demo . py
- - model_name_or_path . / Llama - 7 b - hf
- - checkpoint_dir output - 1
- - finetuning_type lora
- - template default设置，修改接口地址為： http://127.0.0.1:8000/ （即你的API介面位址），然後就可以使用了。 
PubMed Central和PubMed Abstracts等。這些寶貴的文本大大豐富了BLOOMZ模型的醫學知識體系，所以許多開源專案都會優先選擇BLOOMZ做醫學微調的底座模型；质量> 数量這個真理，如：Less is More! 上交清源&& 里海| 利用200條數據微調模型，怒超MiniGPT-4 ！ ，超大規模的SFT資料會讓下游任務LLM減弱或失去ICL、CoT等能力；大规模预训练+小规模监督微调=超强的LLM模型；英文10B以下选择Mistral-7B中文, 10B以下选择Yi-6B , 10B以上选择Qwen-14B和Yi-34B ； Important
歡迎大家在ISSUE中補充新的經驗！
11~13方法論來自於130億大語言模型只改變1個權重就會完全喪失語言能力！復旦大學自然語言處理實驗室最新研究.
14方法論來自於How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition
17~25方法論來自LLM Optimization: Layer-wise Optimal Rank Adaptation (LORA) 中文版解讀
| 階段 | 權重介紹 | 下載地址 | 特點 | 底座模型 | 微調方法 | 數據集 |
|---|---|---|---|---|---|---|
| ?監督微調 | 多輪對話資料是基於LLaMA2-7b-Chat訓練而來 | CareLlama2-7b-chat-sft-multi、?CareLlama2-7b-multi | 優秀的多輪對話能力 | LLaMA2-7b-Chat | QLoRA | mm |
| 監督微調 | 豐富高效醫病對話資料基於LLaMA2-7b-Chat訓練而來 | CareLlama2-7b-chat-sft-med | 優秀的患者疾病診斷能力 | LLaMA2-7b-Chat | QLoRA | hm |
| 監 |
