Instructions to use GabrielCheng/Deepseek-r1-finetuned-drone-safty with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use GabrielCheng/Deepseek-r1-finetuned-drone-safty with Transformers:

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

pipe = pipeline("text-generation", model="GabrielCheng/Deepseek-r1-finetuned-drone-safty")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("GabrielCheng/Deepseek-r1-finetuned-drone-safty")
model = AutoModelForCausalLM.from_pretrained("GabrielCheng/Deepseek-r1-finetuned-drone-safty")
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]:]))

PEFT
How to use GabrielCheng/Deepseek-r1-finetuned-drone-safty with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use GabrielCheng/Deepseek-r1-finetuned-drone-safty with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "GabrielCheng/Deepseek-r1-finetuned-drone-safty"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "GabrielCheng/Deepseek-r1-finetuned-drone-safty",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/GabrielCheng/Deepseek-r1-finetuned-drone-safty

SGLang

How to use GabrielCheng/Deepseek-r1-finetuned-drone-safty 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 "GabrielCheng/Deepseek-r1-finetuned-drone-safty" \
    --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": "GabrielCheng/Deepseek-r1-finetuned-drone-safty",
		"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 "GabrielCheng/Deepseek-r1-finetuned-drone-safty" \
        --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": "GabrielCheng/Deepseek-r1-finetuned-drone-safty",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use GabrielCheng/Deepseek-r1-finetuned-drone-safty with Docker Model Runner:
```
docker model run hf.co/GabrielCheng/Deepseek-r1-finetuned-drone-safty
```

Deepseek-r1-finetuned-drone-safty

基于 DeepSeek-R1-Distill-Qwen-1.5B 微调的无人机飞行安全问答专家测试模型

这是一个基于 DeepSeek-R1-Distill-Qwen-1.5B，使用 PEFT LoRA 进行微调的语言模型。它的主要特点是能够针对无人机飞行安全监控和风险相关问题，生成包含思考过程（以 <think> 标签标识）和最终答案的形式化的回复。

微调数据集来源： (GabrielCheng/Drone-flight-monitoring-reasoning-SFT)

注：微调数据集中只有问答文本数据，没有实时的飞行轨迹、环境信息等数据。所以模型不具备在真实动态场景中的实用性。仅用于展示形式化的微调训练的效果。

模型调用示例

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_id = "GabrielCheng/Deepseek-r1-finetuned-drone-safty"
model = AutoModelForCausalLM.from_pretrained(model_id).to("cuda")
tokenizer = AutoTokenizer.from_pretrained(model_id)

pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

question = "在低能见度环境下，如何利用飞行轨迹数据综合评估无人机的安全风险？"

prompt = f"""以下指令描述了一项任务，并附带了相关背景信息。
请用中文编写一个回复，以恰当地完成此任务请求。
在回答之前，请仔细思考问题，并创建一个逻辑连贯的思考过程，以确保回答准确无误。

### 指令:
你是一位无人机飞行安全监测专家。
请回答以下关于无人机飞行的安全和风险相关问题。

### 问题:
{question}

### 回答:
<think>""" 

response = pipe(prompt, max_length=1000, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)  
print(response[0]['generated_text'])

Downloads last month: 3

Safetensors

Model size

2B params

Tensor type

F32

Model tree for GabrielCheng/Deepseek-r1-finetuned-drone-safty

Base model

deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B

Adapter

(315)

this model

GabrielCheng
/

Deepseek-r1-finetuned-drone-safty

Deepseek-r1-finetuned-drone-safty

基于 DeepSeek-R1-Distill-Qwen-1.5B 微调的无人机飞行安全问答专家测试模型

模型调用示例

Model tree for GabrielCheng/Deepseek-r1-finetuned-drone-safty

Dataset used to train GabrielCheng/Deepseek-r1-finetuned-drone-safty