Instructions to use SylvanL/ChatTCM-7B-SFT with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SylvanL/ChatTCM-7B-SFT with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="SylvanL/ChatTCM-7B-SFT")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("SylvanL/ChatTCM-7B-SFT")
model = AutoModelForCausalLM.from_pretrained("SylvanL/ChatTCM-7B-SFT")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use SylvanL/ChatTCM-7B-SFT with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SylvanL/ChatTCM-7B-SFT"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SylvanL/ChatTCM-7B-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SylvanL/ChatTCM-7B-SFT

SGLang

How to use SylvanL/ChatTCM-7B-SFT with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "SylvanL/ChatTCM-7B-SFT" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SylvanL/ChatTCM-7B-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "SylvanL/ChatTCM-7B-SFT" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SylvanL/ChatTCM-7B-SFT",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use SylvanL/ChatTCM-7B-SFT with Docker Model Runner:
```
docker model run hf.co/SylvanL/ChatTCM-7B-SFT
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

正在写英文论文？

请支持一下作者的最新产品 👉 www.thesisagent.ai

海外学子的AI学术工具，提供从日常写作到科研论文的全方位辅助。

邀请码：N91M8BE33

国内首个从数据集到训练方式到模型权重完全开源的中医大模型.

在2张A800-80G上,

基于SylvanL/ChatTCM-7B-Pretrain, 在llamafactory框架上,

使用SylvanL/Traditional-Chinese-Medicine-Dataset-SFT进行了2个epoch的全参数量有监督微调(full Supervised Fine-tuning).

在不出现明显指令丢失或灾难性遗忘的前提下，使模型具备以下能力：

具有将文言文/古文翻译为现代文的能力, 以加强对于中医典籍的理解与使用.
具有向主流派别执业医生靠拢的临床诊断逻辑与推方能力, 可以理解输入的患者情况并进行判断与分析.
具有良好的中医知识问答能力, 可以针对中医领域的知识点进行全面且可靠的解答.
加强模型面向中医术语的基础nlp能力, 可以更好的赋能如中医命名实体识别, 关系抽取, 关联性分析, 同义实体消岐, 拼写检查与纠错等通用功能.

P.S.: 模型并没有进行任何identify的植入

可选Instruction:

将输入的古文翻译成现代文。
请为输入的现代文找到其对应的古文原文与出处。
基于输入的患者医案记录，直接给出你的证型诊断，无需给出原因。
基于输入的患者医案记录，直接给出你的疾病诊断，无需给出原因。
基于输入的患者医案记录，直接给出你认为的方剂中药组成。
基于输入的患者医案记录，直接给出你认为的【治疗方案】{可多选}∈["中药", "成药", "方剂"]，和【诊断】{可多选}∈["证型", "治法", "西医诊断", "中医诊断"]：

epoch 1: 
    "num_input_tokens_seen": 1649269888,
    "total_flos": 3298213988794368.0,
    "train_loss": 1.0691444667014194,
    "train_runtime": 587389.2072,
    "train_samples_per_second": 3.483,
    "train_steps_per_second": 0.016

epoch 2: 
    "num_input_tokens_seen": 1649269888,
    "total_flos": 3298213988794368.0,
    "train_loss": 0.7717254485568724,
    "train_runtime": 600800.1758,
    "train_samples_per_second": 3.406,
    "train_steps_per_second": 0.015

llamafactory-cli train \
    --stage sft \
    --do_train True \
    --model_name_or_path {SylvanL/ChatTCM-7B-Pretrain} \
    --preprocessing_num_workers 16 \
    --finetuning_type full \
    --template default \
    --flash_attn auto \
    --dataset_dir {dataset_dir} \
    --dataset SFT_medicalKnowledge_source1_548404,SFT_medicalKnowledge_source2_99334,SFT_medicalKnowledge_source3_556540,SFT_nlpDiseaseDiagnosed_61486,SFT_nlpSyndromeDiagnosed_48665,SFT_structGeneral_310860,SFT_structPrescription_92896,_SFT_traditionalTrans_1959542.json,{BAAI/COIG},{m-a-p/COIG-CQIA} \
    --cutoff_len 1024 \
    --learning_rate 5e-05 \
    --num_train_epochs 2.0 \
    --max_samples 1000000 \
    --per_device_train_batch_size 28 \
    --gradient_accumulation_steps 4 \
    --lr_scheduler_type cosine \
    --max_grad_norm 1.0 \
    --logging_steps 1 \
    --save_steps 1000 \
    --warmup_steps 0 \
    --optim adamw_torch \
    --packing False \
    --report_to none \
    --output_dir {output_dir} \
    --bf16 True \
    --plot_loss True \
    --ddp_timeout 180000000 \
    --include_num_input_tokens_seen True \
    --deepspeed cache/ds_z3_offload_config.json

Downloads last month: 60

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for SylvanL/ChatTCM-7B-SFT

Base model

Qwen/Qwen2-7B

Finetuned

Qwen/Qwen2-7B-Instruct

Finetuned

SylvanL/ChatTCM-7B-Pretrain

Finetuned

(1)

this model

Quantizations

2 models

SylvanL
/

ChatTCM-7B-SFT

正在写英文论文？

请支持一下作者的最新产品 👉 www.thesisagent.ai

海外学子的AI学术工具，提供从日常写作到科研论文的全方位辅助。

邀请码：N91M8BE33

国内首个从数据集到训练方式到模型权重完全开源的中医大模型.

Model tree for SylvanL/ChatTCM-7B-SFT

Dataset used to train SylvanL/ChatTCM-7B-SFT

正在写英文论文？

请支持一下作者的最新产品 👉 www.thesisagent.ai

海外学子的AI学术工具， 提供从日常写作到科研论文的全方位辅助。

邀请码：N91M8BE33

国内首个从数据集到训练方式到模型权重完全开源的中医大模型.

Model tree for SylvanL/ChatTCM-7B-SFT

Dataset used to train SylvanL/ChatTCM-7B-SFT

海外学子的AI学术工具，提供从日常写作到科研论文的全方位辅助。