vscode debug Transformer源码说明

JoannaJuanCV2025-10-24 18:41

首选确认conda env 环境中未使用 install 安装Transformer；
下载源码：

bash 复制代码

git clone https://github.com/huggingface/transformers.git
cd transformers

# pip
pip install '.[torch]'

安装后通过pip list 可以看到是指向你的本地目录：

python 复制代码

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Qwen/Qwen3-0.6B"

# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True, use_fast=False)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    trust_remote_code=True,
    dtype="auto",
    device_map="auto"
)

# prepare the model input
prompt = "Give me a short introduction to large language model."
messages = [
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# conduct text completion
generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):].tolist()

content = tokenizer.decode(output_ids, skip_special_tokens=True)

print("content:", content)

vscode debug：按照以往代码debug即可；

安装插件
vscode 右下角选择你的env，如下
点击上方的三个点-》运行-》启动调试