##官網(wǎng)：https://github.com/unslothai/unsloth

#01 創(chuàng)建項目，并設(shè)置python虛擬環(huán)境，python3.11版本

#02 安裝 unsloth（cpu版本）
brew install llvm（Homebrew clang version 19.1.7）
echo 'export PATH="/opt/homebrew/opt/llvm/bin:$PATH"' >> ~/.zshrc
source ~/.zshrc

pip install torch
pip install numpy
pip install"unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"



#03 版本檢查
python -c "import torch; print(torch.__version__)"
2.6.0

#04 引用
from unsloth import FastLanguageModel

#01 linux 服務(wù)建議使用docker


#02 拉取鏡像
docker pull modelscope-registry.us-west-1.cr.aliyuncs.com/modelscope-repo/modelscope:ubuntu22.04-py310-torch2.3.1-1.22.2

#03 啟動

#01 環(huán)境是python3.11

#02 項目目錄
Unsloth-DeepSeek-R1-8b/
├── data/                    # 存放訓(xùn)練數(shù)據(jù)、驗證數(shù)據(jù)等
│   ├── raw/                 # 原始數(shù)據(jù)
│   └── processed/           # 預(yù)處理后的數(shù)據(jù)
│
├── models/                  # 存放模型文件
│   ├── checkpoints/         # 存儲訓(xùn)練過程中的模型檢查點
│   └── final_model/         # 存儲最終微調(diào)后的模型
│
├── scripts/                 # 存放訓(xùn)練腳本、數(shù)據(jù)處理腳本等
│   ├── train.py             # 訓(xùn)練腳本
│   ├── data_preprocessing.py# 數(shù)據(jù)預(yù)處理腳本
│   └── evaluate.py          # 評估腳本
│
├── logs/                    # 存放訓(xùn)練日志文件
│   └── training_logs.txt    # 訓(xùn)練過程中的日志
│
├── wandb/                   # 存放 wandb 相關(guān)的配置和記錄
│   └── wandb_config.py      # wandb 配置文件
│
├── environment/             # 環(huán)境配置文件
│   ├── requirements.txt     # 項目的 Python 依賴
│   └── environment.yml      # 如果使用 Conda，可以創(chuàng)建一個環(huán)境配置文件
│
├── main.py                  # 主運行文件，啟動訓(xùn)練或其他任務(wù)
└── README.md                # 項目的描述文件，包含如何使用和運行的說明


#03 創(chuàng)建目錄
# 創(chuàng)建子目錄
mkdir -p data/raw
mkdir -p data/processed
mkdir -p models/checkpoints
mkdir -p models/final_model
mkdir -p scripts
mkdir -p logs
mkdir -p wandb
mkdir -p environment

# 創(chuàng)建文件
touch scripts/train.py
touch scripts/data_preprocessing.py
touch scripts/evaluate.py
touch logs/training_logs.txt
touch wandb/wandb_config.py
touch environment/requirements.txt
touch environment/environment.yml
touch main.py
touch README.md

#03 安裝即可
pip install torch==2.6.0 transformers datasets

#03 更新證書(后續(xù)如果有pip網(wǎng)站使用https 會驗證該證書)
/Applications/Python\ 3.11/Install\ Certificates.command

#01 安裝 peft
pip install peft

#01 aipkey (本人谷歌郵箱)


#02 命令
pip install wandb
wandb login

#02  運行文件
import wandb  # 導(dǎo)入 wandb 庫，用于跟蹤和可視化實驗
import random  # 導(dǎo)入 random 庫，用于生成隨機數(shù)

# 開始一個新的 wandb 運行來跟蹤當前腳本
wandb.init(
    # 設(shè)置 wandb 項目，所有與該運行相關(guān)的數(shù)據(jù)將被記錄到這個項目中
    project="my-awesome-project",  # 項目名稱，你可以在 wandb 儀表盤中看到這個項目
    
    # 追蹤超參數(shù)和運行的元數(shù)據(jù)
    config={
        "learning_rate": 0.02,  # 設(shè)置學(xué)習(xí)率
        "architecture": "CNN",  # 模型架構(gòu)（這里是卷積神經(jīng)網(wǎng)絡(luò)）
        "dataset": "CIFAR-100",  # 使用的數(shù)據(jù)集（這里是 CIFAR-100 數(shù)據(jù)集）
        "epochs": 10,  # 訓(xùn)練的輪數(shù)
    }
)

# 模擬訓(xùn)練過程
epochs = 10# 總訓(xùn)練輪數(shù)
offset = random.random() / 5# 生成一個小的隨機偏移量，用于模擬訓(xùn)練過程中一些不確定性

# 開始訓(xùn)練循環(huán)，模擬 2 到 10 輪的訓(xùn)練過程
for epoch inrange(2, epochs):  # 從第二輪開始，到第 10 輪結(jié)束
    # 模擬準確率的變化，隨著 epoch 的增加，準確率逐漸提升
    acc = 1 - 2 ** -epoch - random.random() / epoch - offset
    
    # 模擬損失的變化，隨著 epoch 的增加，損失逐漸減少
    loss = 2 ** -epoch + random.random() / epoch + offset

    # 使用 wandb 記錄每一輪的準確率（acc）和損失值（loss）
    wandb.log({"acc": acc, "loss": loss})

# [可選] 結(jié)束 wandb 運行，確保數(shù)據(jù)被正確上傳并完成記錄
wandb.finish()

#01 安裝modelscope 
pip install modelscope

#02 下載模型文件
mkdir -p ./models/DeepSeek-R1-Distill-Llama-8B
mkdir -p ./models/DeepSeek-R1-Distill-Qwen-1.5B
mkdir -p ./models/DeepSeek-R1-Distill-Qwen-7B

modelscope download --model deepseek-ai/DeepSeek-R1-Distill-Llama-8B --local_dir ./models/DeepSeek-R1-Distill-Llama-8B

modelscope download --model deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B --local_dir ./models/DeepSeek-R1-Distill-Qwen-1.5B

modelscope download --model deepseek-ai/DeepSeek-R1-Distill-Qwen-7B --local_dir ./models/DeepSeek-R1-Distill-Qwen-7B



modelscope download --model deepseek-ai/DeepSeek-R1-Distill-Llama-8B --local_dir ./DeepSeek-R1-Distill-Llama-8B

"""


訓(xùn)練前詢問問題：
  皮質(zhì)醇增多癥患者在血漿ACTH明顯升高且大劑量地塞米松抑制試驗陽性的情況下，應(yīng)考慮哪種疾病？
  
訓(xùn)練后再次詢問：


scripts/test_inference.py

"""


import os
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

# 獲取當前腳本的路徑
current_dir = os.path.dirname(__file__)

# 拼接模型和分詞器路徑
model_dir = os.path.join(current_dir, '..', 'models', 'DeepSeek-R1-Distill-Qwen-1.5B')

# 打印路徑確認
print(f"Model path: {model_dir}")

# 確保模型和分詞器的路徑存在
ifnot os.path.exists(model_dir):
    raise ValueError(f"Model directory does not exist at {model_dir}")
else:
    print("Model directory exists, proceeding with loading.")

# 加載模型和分詞器
print("Loading model and tokenizer...")
model = AutoModelForCausalLM.from_pretrained(model_dir)
tokenizer = AutoTokenizer.from_pretrained(model_dir)

# 打印模型和分詞器的配置信息
print(f"Model config: {model.config}")
print(f"Tokenizer config: {tokenizer}")

# 輸入中文文本
input_text = "皮質(zhì)醇增多癥患者在血漿ACTH明顯升高且大劑量地塞米松抑制試驗陽性的情況下，應(yīng)考慮哪種疾病？"
print(f"User input: {input_text}")

# 結(jié)構(gòu)化的 prompt
prompt_style_chat = f"""請寫出一個恰當?shù)幕卮饋硗瓿僧斍皩υ捜蝿?wù)。

### Instruction:
你是一名助人為樂的助手。

### Question:
{input_text}

### Response:
<think>"""

# 使用分詞器處理輸入文本
inputs = tokenizer(prompt_style_chat, return_tensors="pt", padding=True, truncation=True, max_length=512)

# 打印 tokenized 輸入
print(f"Tokenized input: {inputs}")

# 打印輸入形狀
print(f"Input shape: {inputs['input_ids'].shape}")

# 打印模型的最大長度
print(f"Model max length: {model.config.max_position_embeddings}")

# 將模型移至正確的設(shè)備（使用 GPU 如果可用）
device = "cuda"if torch.cuda.is_available() else"cpu"
model.to(device)

# 打印設(shè)備信息
print(f"Using device: {device}")

# 打印分詞器的 pad_token_id
pad_token_id = tokenizer.pad_token_id if tokenizer.pad_token_id isnotNoneelse model.config.pad_token_id
print(f"Using pad_token_id: {pad_token_id}")

# 生成模型輸出
print("Generating response...")
# 使用 max_new_tokens 來控制生成長度
with torch.no_grad():  # 禁用梯度計算，節(jié)省內(nèi)存
    try:
        print("Calling model.generate()...")
        outputs = model.generate(
            inputs['input_ids'].to(device),
            attention_mask=inputs['attention_mask'].to(device),
            max_new_tokens=1200,  # 設(shè)置最大生成的 token 數(shù)量
            temperature=1.0,
            top_p=0.9,
            pad_token_id=pad_token_id
        )

        print("Model.generate() completed.")
    except Exception as e:
        print(f"Error generating response: {e}")
        raise

# 打印生成的輸出 ID 和它們的形狀
print(f"Generated output IDs: {outputs}")
print(f"Shape of generated output: {outputs.shape}")

# 解碼生成的輸出文本
try:
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    print(f"Generated response: {response}")
except Exception as e:
    print(f"Error decoding output: {e}")

User input:皮質(zhì)醇增多癥患者在血漿ACTH明顯升高且大劑量地塞米松抑制試驗陽性的情況下，應(yīng)考慮哪種疾病？
Tokenized input: {'input_ids':tensor([[151646,  14880, 112672,  46944, 112449, 111423,  36407,  60548,  67949,
         105051,  88802,   3407,  14374,  29051,    510,  56568, 110124,  99262,
         103247,  99350,   9370, 110498,   3407,  14374,  15846,    510,  99888,
          99178, 103032, 107284,  99769, 101924,  18493,  99389, 101498,   6823,
             39, 100687, 109061, 100136,  26288, 114786,  29490, 101202,  72261,
         100180, 106555, 102360, 112758, 104248,   3837,  50511, 101118, 113195,
         101160,  26850,  14374,   5949,    510, 151648]]), 'attention_mask':tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
         1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
         1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]])}
Input shape:torch.Size([1,60])
Model max length:131072
Using device:cpu
Using pad_token_id:151643
Generatingresponse...
Callingmodel.generate()...
Model.generate()completed.

Generated response:請寫出一個恰當?shù)幕卮饋硗瓿僧斍皩υ捜蝿?wù)。

### Instruction:
你是一名助人為樂的助手。

### Question:
皮質(zhì)醇增多癥患者在血漿ACTH明顯升高且大劑量地塞米松抑制試驗陽性的情況下，應(yīng)考慮哪種疾病？

### Response:
<think>
好的，我現(xiàn)在需要仔細分析這個問題并給出一個合適的回答。首先，問題描述的是皮質(zhì)醇增多癥（PHT）患者在血漿ACTH明顯升高且大劑量地塞米松抑制試驗（SSDS）顯示陽性的情況下，應(yīng)考慮哪種疾病。

首先，我記得皮質(zhì)醇增多癥是由于皮質(zhì)醇分泌異常導(dǎo)致，通常由代謝紊亂或神經(jīng)退行性疾病引起，比如皮質(zhì)醇過激釋放癥、皮質(zhì)醇過激釋放性代謝綜合征等。通常，患者可能表現(xiàn)出皮質(zhì)醇水平升高，血漿ACTH顯著升高，這符合題意的第一個條件。

接下來，第二個條件是SSDS試驗陽性。SSDS試驗主要用于檢測皮質(zhì)醇釋放的細胞因子，比如PD-L1，這些因子在疾病早期有顯著的表觀變化。皮質(zhì)醇增多癥患者的皮質(zhì)醇釋放確實受阻，導(dǎo)致細胞因子釋放減少，這在SSDS中會被檢測出來，所以這種情況屬于皮質(zhì)醇增多癥。

綜合這兩個條件，患者的血漿ACTH升高和SSDS陽性，符合皮質(zhì)醇增多癥的特征。因此，這種情況下應(yīng)考慮的是皮質(zhì)醇增多癥。

我需要確保我沒有遺漏其他可能導(dǎo)致SSDS試驗陽性的情況。比如，是否有一些其他類型的疾病，比如胰島素素合成障礙或胰島素缺乏，也會影響皮質(zhì)醇釋放？不過，這些更可能是胰島素素合成障礙，而不是直接由皮質(zhì)醇釋放受阻引起的。皮質(zhì)醇增多癥通常是由于皮質(zhì)醇釋放異常，因此SSDS陽性更直接與皮質(zhì)醇釋放受阻相關(guān)。

此外，ACTH升高可能與皮質(zhì)醇增多癥不同，而更可能是由于激素分泌過量或其他激素調(diào)節(jié)問題。因此，ACTH升高的信號應(yīng)該更多指向皮質(zhì)醇增多癥。

綜上所述，這種情況下應(yīng)該考慮的疾病是皮質(zhì)醇增多癥。
</think>

應(yīng)考慮皮質(zhì)醇增多癥（PantoprazolidonePhenomenon）。

因為：

1.血漿ACTH顯著升高，符合皮質(zhì)醇增多癥的特征。
2.SSDS試驗陽性，表明皮質(zhì)醇釋放受阻，屬于皮質(zhì)醇增多癥的表現(xiàn)。

#01 我們使用COT格式 醫(yī)學(xué)領(lǐng)域 medical-o1-reasoning-SFT 數(shù)據(jù)集
https://huggingface.co/datasets/FreedomIntelligence/medical-o1-reasoning-SFT

#02 b本地導(dǎo)入方式（）
from datasets import load_dataset
ds = load_dataset("FreedomIntelligence/medical-o1-reasoning-SFT", "zh")

#01 使用modelscope 數(shù)據(jù)集 官網(wǎng)地址
https://www.modelscope.cn/datasets/YIRONGCHEN/PsyDTCorpus/files

#02 下載完整數(shù)據(jù)集repo
modelscope download --dataset YIRONGCHEN/PsyDTCorpus --local_dir ./dir


#03 下載單個文件到指定本地文件夾（以下載README.md到當前路徑下“dir”目錄為例）
modelscope download --dataset YIRONGCHEN/PsyDTCorpus README.md --local_dir ./dir

#01 用于對medical-o1-reasoning-SFT數(shù)據(jù)集進行修改，Complex_CoT列和Response列進行拼接，并加上文本結(jié)束標記：
defformatting_prompts_func(examples, EOS_TOKEN):
    """
    格式化數(shù)據(jù)集中的每個示例，使其符合訓(xùn)練的要求。

    Args:
        examples (dict): 數(shù)據(jù)集中的輸入示例
        EOS_TOKEN (str): 結(jié)束符

    Returns:
        dict: 格式化后的文本數(shù)據(jù)
    """
    train_prompt_style = """Below is an instruction that describes a task, paired with an input that provides further context. 
    Write a response that appropriately completes the request. 
    Before answering, think carefully about the question and create a step-by-step chain of thoughts to ensure a logical and accurate response.

    ### Instruction:
    You are a medical expert with advanced knowledge in clinical reasoning, diagnostics, and treatment planning. 
    Please answer the following medical question. 

    ### Question:
    {}

    ### Response:
    <think>
    {}
    </think>
    {}"""

    inputs = examples["Question"]
    cots = examples["Complex_CoT"]
    outputs = examples["Response"]
    texts = []
    forinput, cot, output inzip(inputs, cots, outputs):
        text = train_prompt_style.format(input, cot, output) + EOS_TOKEN
        texts.append(text)
    return {"text": texts}



"""

問題（{}） 被嵌套到 ### Question: 下面，替換掉 {}。
推理過程（{}） 被嵌套到 <think></think> 標簽內(nèi)，替換掉第二個 {}。
答案（{}） 被嵌套到模板的最后，替換掉第三個 {}。
具體替換流程：
{} 第一個位置將會被每個樣本中的問題（examples["Question"]）替換。
{} 第二個位置將會被每個樣本中的推理過程（examples["Complex_CoT"]）替換。
{} 第三個位置將會被每個樣本中的答案（examples["Response"]）替換。
例如，如果輸入數(shù)據(jù)如下：

問題（Question）: "What is the cause of fever?"
推理過程（Complex_CoT）: "Fever is usually caused by an infection or inflammation. We need to identify the source."
答案（Response）: "The most common causes of fever are bacterial or viral infections."

"""

{
    "Question": [
        "What is the cause of headache?",
        "How do you treat a cold?"
    ],
    "Complex_CoT": [
        "The causes of headaches are numerous, including tension, dehydration, or sinus issues.",
        "Treating a cold typically involves rest, fluids, and over-the-counter medications for symptoms."
    ],
    "Response": [
        "A headache can be caused by stress, lack of sleep, or a sinus infection.",
        "For a cold, hydration and rest are key. Medications like ibuprofen can help with symptoms."
    ]
}

{
    "text": [
        """Below is an instruction that describes a task, paired with an input that provides further context. 
        Write a response that appropriately completes the request. 
        Before answering, think carefully about the question and create a step-by-step chain of thoughts to ensure a logical and accurate response.

        ### Instruction:
        You are a medical expert with advanced knowledge in clinical reasoning, diagnostics, and treatment planning. 
        Please answer the following medical question. 

        ### Question:
        What is the cause of headache?

        ### Response:
        <think>
        The causes of headaches are numerous, including tension, dehydration, or sinus issues.
        </think>
        A headache can be caused by stress, lack of sleep, or a sinus infection. <|endoftext|>
        """,
        """Below is an instruction that describes a task, paired with an input that provides further context. 
        Write a response that appropriately completes the request. 
        Before answering, think carefully about the question and create a step-by-step chain of thoughts to ensure a logical and accurate response.

        ### Instruction:
        You are a medical expert with advanced knowledge in clinical reasoning, diagnostics, and treatment planning. 
        Please answer the following medical question. 

        ### Question:
        How do you treat a cold?

        ### Response:
        <think>
        Treating a cold typically involves rest, fluids, and over-the-counter medications for symptoms.
        </think>
        For a cold, hydration and rest are key. Medications like ibuprofen can help with symptoms. <|endoftext|>
        """
    ]
}

import os
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments
from datasets import load_dataset
from peft import get_peft_model, LoraConfig
from trl import SFTTrainer  # 使用 SFTTrainer
import wandb
from config import setting

# 設(shè)置環(huán)境變量，禁用tokenizer的并行化
os.environ["TOKENIZERS_PARALLELISM"] = "false"


# 登錄wandb
defsetup_wandb():
    """
    登錄到wandb以便記錄訓(xùn)練過程中的日志和指標。
    """
    wandb.login()


# 設(shè)置路徑
defset_paths():
    """
    設(shè)置項目根目錄、模型路徑、數(shù)據(jù)集路徑和最終模型保存路徑。

    Returns:
        model_dir (str): 模型文件路徑
        dataset_path (str): 數(shù)據(jù)集路徑
        final_model_dir (str): 微調(diào)后模型的保存路徑
    """
    root_dir = setting.root_dir  # 項目根路徑
    model_dir = os.path.join(root_dir, 'models', 'DeepSeek-R1-Distill-Qwen-1.5B')  # 模型文件路徑
    dataset_path = os.path.join(root_dir, 'data', 'medical-o1-reasoning-SFT')  # 數(shù)據(jù)集路徑
    final_model_dir = os.path.join(root_dir, 'models', 'final_model')  # 高效微調(diào)后模型保存路徑
    print(f'設(shè)置模型路徑：{model_dir} | 數(shù)據(jù)集位置：{dataset_path}')
    return model_dir, dataset_path, final_model_dir


# 加載模型和分詞器
defload_model_and_tokenizer(model_dir):
    """
    加載預(yù)訓(xùn)練模型和對應(yīng)的分詞器，并獲取結(jié)束符（EOS_TOKEN）。

    Args:
        model_dir (str): 模型的文件路徑

    Returns:
        model (AutoModelForCausalLM): 加載的模型
        tokenizer (AutoTokenizer): 加載的分詞器
        EOS_TOKEN (str): 模型的結(jié)束符（如果沒有，使用默認值）
    """
    print("加載分詞器：Loading model and tokenizer...")
    model = AutoModelForCausalLM.from_pretrained(model_dir)
    tokenizer = AutoTokenizer.from_pretrained(model_dir)

    EOS_TOKEN = tokenizer.eos_token
    if EOS_TOKEN isNone:
        EOS_TOKEN = "<|endoftext|>"

    print(f'結(jié)束符：{EOS_TOKEN}')
    return model, tokenizer, EOS_TOKEN


# 格式化訓(xùn)練數(shù)據(jù)
defformatting_prompts_func(examples, EOS_TOKEN):
    """
    格式化數(shù)據(jù)集中的每個示例，使其符合訓(xùn)練的要求。

    Args:
        examples (dict): 數(shù)據(jù)集中的輸入示例
        EOS_TOKEN (str): 結(jié)束符

    Returns:
        dict: 格式化后的文本數(shù)據(jù)
    """
    train_prompt_style = """Below is an instruction that describes a task, paired with an input that provides further context. 
    Write a response that appropriately completes the request. 
    Before answering, think carefully about the question and create a step-by-step chain of thoughts to ensure a logical and accurate response.

    ### Instruction:
    You are a medical expert with advanced knowledge in clinical reasoning, diagnostics, and treatment planning. 
    Please answer the following medical question. 

    ### Question:
    {}

    ### Response:
    <think>
    {}
    </think>
    {}"""

    inputs = examples["Question"]
    cots = examples["Complex_CoT"]
    outputs = examples["Response"]
    texts = []
    forinput, cot, output inzip(inputs, cots, outputs):
        text = train_prompt_style.format(input, cot, output) + EOS_TOKEN
        texts.append(text)
    return {"text": texts}


# 設(shè)置LoRA配置
defsetup_lora(model):
    """
    設(shè)置LoRA（低秩適配器）配置，并將其應(yīng)用到模型。

    Args:
        model (AutoModelForCausalLM): 加載的模型

    Returns:
        model (AutoModelForCausalLM): 應(yīng)用LoRA后的模型
    """
    print("設(shè)置LoRA: Setting up LoRA configuration...")
    lora_config = LoraConfig(
        r=8,  # adapter的秩
        lora_alpha=32,  # 縮放因子
        lora_dropout=0.1,  # LoRA層的dropout
        bias="none",  # LoRA的偏置項
    )
    return get_peft_model(model, lora_config)


# 加載數(shù)據(jù)集
defload_dataset_func(dataset_path, train_size=100):
    """
    從指定路徑加載數(shù)據(jù)集，訓(xùn)練集大小為 train_size，評估集為訓(xùn)練集的10%，但至少為1。
    """
    print(f"從 {dataset_path} 加載數(shù)據(jù)集...")
    # 加載數(shù)據(jù)集
    dataset = load_dataset(dataset_path, "en", split="train", trust_remote_code=True)

    # 計算評估集大小
    eval_size = max(1, int(train_size * 0.1))  # 評估集大小是訓(xùn)練集的10%，但至少為1

    # 切分數(shù)據(jù)集
    train_dataset = dataset.select(range(train_size))  # 使用前 train_size 條作為訓(xùn)練集
    eval_dataset = dataset.select(range(train_size, train_size + eval_size))  # 剩余部分作為評估集

    print(f"訓(xùn)練集大小: {len(train_dataset)}, 評估集大小: {len(eval_dataset)}")
    return train_dataset, eval_dataset


# 配置訓(xùn)練參數(shù)
defsetup_training_args(final_model_dir, enable_evaluation=True):
    """
    設(shè)置訓(xùn)練參數(shù)，包括輸出目錄、學(xué)習(xí)率、批處理大小等，并根據(jù)參數(shù)控制是否啟用評估。

    Args:
        final_model_dir (str): 微調(diào)后模型保存的路徑
        enable_evaluation (bool): 是否啟用評估。默認為True，啟用評估；為False時禁用評估。

    Returns:
        training_args (TrainingArguments): 訓(xùn)練參數(shù)
    """
    # 根據(jù)是否啟用評估設(shè)置 evaluation_strategy
    evaluation_strategy = "epoch"if enable_evaluation else"no"

    training_args = TrainingArguments(
        output_dir=final_model_dir,
        evaluation_strategy=evaluation_strategy,  # 控制評估策略
        learning_rate=5e-5,
        per_device_train_batch_size=2,  # 適當減少批處理大小（根據(jù)M3 Pro的內(nèi)存限制）
        gradient_accumulation_steps=4,  # 使用梯度累積，模擬更大的批量
        num_train_epochs=3,  # 訓(xùn)練3個周期
        report_to="wandb",  # 使用wandb進行訓(xùn)練日志記錄
        weight_decay=0.01,
        logging_dir=os.path.join(setting.root_dir, 'logs'),
        logging_steps=50,  # 減少日志記錄頻率
        save_steps=500,  # 增加模型保存的步數(shù)頻率，減少頻繁保存
        save_total_limit=2,  # 保存最多2個模型
        dataloader_num_workers=4,  # 設(shè)置數(shù)據(jù)加載器的并行數(shù)（根據(jù)需要調(diào)整）
    )
    return training_args



# 訓(xùn)練模型
deftrain_model(model, training_args, dataset, eval_dataset, tokenizer, enable_evaluation=True):
    """
    使用SFTTrainer進行模型訓(xùn)練。

    Args:
        model (AutoModelForCausalLM): 需要訓(xùn)練的模型
        training_args (TrainingArguments): 訓(xùn)練參數(shù)
        dataset (Dataset): 用于訓(xùn)練的數(shù)據(jù)集
        eval_dataset (Dataset): 用于評估的數(shù)據(jù)集
        tokenizer (AutoTokenizer): 分詞器
        enable_evaluation (bool): 是否進行評估

    Returns:
        trainer (SFTTrainer): 訓(xùn)練器實例
    """
    # 如果啟用了評估，傳遞評估集
    trainer = SFTTrainer(
        model=model,
        args=training_args,
        train_dataset=dataset,
        eval_dataset=eval_dataset if enable_evaluation elseNone,  # 根據(jù)參數(shù)決定是否傳遞評估集
        tokenizer=tokenizer,
        data_collator=None,  # 可以選擇合適的data collator
    )
    trainer.train()
    return trainer


# 保存模型
defsave_model(trainer, final_model_dir):
    """
    保存訓(xùn)練后的模型到指定目錄。

    Args:
        trainer (SFTTrainer): 訓(xùn)練器實例
        final_model_dir (str): 模型保存路徑
    """
    print("Saving model...")
    trainer.save_model(final_model_dir)



defmerge_models(models, weights, device="cpu"):
    """
    合并多個模型的權(quán)重（加權(quán)平均）。

    Args:
        models (list): 模型列表
        weights (list): 權(quán)重列表，權(quán)重數(shù)量與模型數(shù)量一致
        device (str): 設(shè)備，可以是 "cuda" 或 "cpu"

    Returns:
        merged_model (nn.Module): 合并后的模型
    """
    # 確保模型數(shù)量與權(quán)重數(shù)量一致
    assertlen(models) == len(weights), "模型數(shù)量與權(quán)重數(shù)量不一致"

    # 將所有模型加載到相同的設(shè)備
    for i inrange(len(models)):
        models[i] = models[i].to(device)

    # 獲取第一個模型的狀態(tài)字典
    merged_state_dict = models[0].state_dict()

    # 對每一層的權(quán)重進行加權(quán)平均
    for key in merged_state_dict.keys():
        merged_state_dict[key] = torch.zeros_like(merged_state_dict[key])
        for model, weight inzip(models, weights):
            merged_state_dict[key] += model.state_dict()[key] * weight

    # 創(chuàng)建一個新的模型并加載合并后的權(quán)重
    merged_model = models[0].__class__.from_pretrained(models[0].config)
    merged_model.load_state_dict(merged_state_dict)
    return merged_model


# 主函數(shù)
defmain():
    """
    主函數(shù)，執(zhí)行整個訓(xùn)練流程：設(shè)置路徑、加載模型、訓(xùn)練并保存模型。

    參數(shù)設(shè)置：
            enable_evaluation = False  # 設(shè)置為False以禁用評估 如果性能慢可以設(shè)置 False

    加載數(shù)據(jù)集：
        train_size=10 設(shè)置數(shù)據(jù)集大小，評估集是數(shù)據(jù)集百分之10（如果小于1 則等于1 ）
        train_dataset, eval_dataset = load_dataset_func(dataset_path, train_size=10)


    """
    setup_wandb()  # 登錄wandb
    model_dir, dataset_path, final_model_dir = set_paths()  # 設(shè)置路徑

    model, tokenizer, EOS_TOKEN = load_model_and_tokenizer(model_dir)  # 加載模型和分詞器

    train_dataset, eval_dataset = load_dataset_func(dataset_path, train_size=5)  # 加載數(shù)據(jù)集
    train_dataset = train_dataset.map(lambda examples: formatting_prompts_func(examples, EOS_TOKEN), batched=True)  # 格式化數(shù)據(jù)集
    eval_dataset = eval_dataset.map(lambda examples: formatting_prompts_func(examples, EOS_TOKEN), batched=True)  # 格式化評估集
    print(train_dataset["text"][0])  # 打印格式化后的數(shù)據(jù)

    model = setup_lora(model)  # 配置LoRA
    # 設(shè)置是否開啟評估
    enable_evaluation = True# 設(shè)置為False以禁用評估
    training_args = setup_training_args(final_model_dir,enable_evaluation)  # 配置訓(xùn)練參數(shù)
    trainer = train_model(model, training_args, train_dataset, eval_dataset, tokenizer, enable_evaluation)  # 開始訓(xùn)練

    save_model(trainer, final_model_dir)  # 保存模型
    wandb.finish()  # 完成wandb記錄




# 執(zhí)行主函數(shù)
if __name__ == "__main__":
    main()

"""
保存在本地 models/final_model 路徑下

"""

defsave_model(trainer, final_model_dir):
    """
    保存訓(xùn)練后的模型到指定目錄。

    Args:
        trainer (SFTTrainer): 訓(xùn)練器實例
        final_model_dir (str): 模型保存路徑
    """
    print("Saving model...")
    trainer.save_model(final_model_dir)

#01 執(zhí)行即可
new_model_local = "DeepSeek-R1-Medical-COT-Tiny"
model.save_pretrained(new_model_local) 
tokenizer.save_pretrained(new_model_local)
model.save_pretrained_merged(new_model_local, tokenizer, save_method = "merged_16bit",)

#02 詳細日志
wandb: ?? View project at https://wandb.ai/z15119911990-beijing/huggingface
wandb: ?? View run at https://wandb.ai/z15119911990-beijing/huggingface/runs/mgrko2jv
  0%|          | 0/3 [00:00<?, ?it/s]
{'eval_runtime': 14.8693, 'eval_samples_per_second': 0.067, 'eval_steps_per_second': 0.067, 'epoch': 0}
                                     
  0%|          | 0/3 [00:30<?, ?it/s]
100%|██████████| 1/1 [00:00<00:00, 1461.94it/s]
                                               
                                     
{'eval_runtime': 21.2073, 'eval_samples_per_second': 0.047, 'eval_steps_per_second': 0.047, 'epoch': 0}
  0%|          | 0/3 [02:11<?, ?it/s]
100%|██████████| 1/1 [00:00<00:00, 33.69it/s]
                                             
                                     
  0%|          | 0/3 [04:02<?, ?it/s]
100%|██████████| 1/1 [00:00<00:00, 334.66it/s]
                                              {'eval_runtime': 18.3908, 'eval_samples_per_second': 0.054, 'eval_steps_per_second': 0.054, 'epoch': 0}
{'train_runtime': 251.7582, 'train_samples_per_second': 0.06, 'train_steps_per_second': 0.012, 'train_loss': 14435.3666305542, 'epoch': 0}
  0%|          | 0/3 [04:10<?, ?it/s]
wandb:                                                                                
wandb: 
wandb: Run history:
wandb:            eval/runtime ▁█▅
wandb: eval/samples_per_second █▁▃
wandb:   eval/steps_per_second █▁▃
wandb:             train/epoch ▁▁▁▁
wandb:       train/global_step ▁▁▁▁
wandb: 
wandb: Run summary:
wandb:             eval/runtime 18.3908
wandb:  eval/samples_per_second 0.054
wandb:    eval/steps_per_second 0.054
wandb:               total_flos 43804457687040.0
wandb:              train/epoch 0
wandb:        train/global_step 0
wandb:               train_loss 14435.36663
wandb:            train_runtime 251.7582
wandb: train_samples_per_second 0.06
wandb:   train_steps_per_second 0.012
wandb: 
wandb: ?? View run /Users/ningcaichen/Documents/02-python相關(guān)文檔/01-AI系列/LoRA-DeepSeek-R1/models/final_model at: https://wandb.ai/z15119911990-beijing/huggingface/runs/mgrko2jv
wandb: ?? View project at: https://wandb.ai/z15119911990-beijing/huggingface
wandb: Synced 5 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)
wandb: Find logs at: ./wandb/run-20250212_133457-mgrko2jv/logs