Gxr777_EduNLP

1. 简介

Gxr777/EduNLP Qwen2.5 0.5B 文本生成模型。基于 HuggingFace transformers，支持一键加载推理。权重来源：ModelScope

2. 快速使用

import torch; from transformers import AutoTokenizer, AutoModelForCausalLM
device = torch.device("npu:0" if torch.npu.is_available() else "cpu")
tokenizer = AutoTokenizer.from_pretrained("Gxr777/EduNLP", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("Gxr777/EduNLP", torch_dtype=torch.float16, trust_remote_code=True).to(device).eval()
messages = [{"role": "user", "content": "什么是机器学习？"}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(device)
outputs = model.generate(**inputs, max_new_tokens=128, do_sample=False)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

3. 精度评测

指标	值
Top-1 一致性	4/4
Max Logit Diff Ratio	8e-06
Avg KL Divergence	0.0
结论	PASS

4. 性能参考

NPU: 1722.85 ms（生成 64 个 tokens，fp16）

Gxr777_EduNLP

1. 简介

Gxr777/EduNLP Qwen2.5 0.5B 文本生成模型。基于 HuggingFace transformers，支持一键加载推理。权重来源：ModelScope

2. 快速使用

import torch; from transformers import AutoTokenizer, AutoModelForCausalLM
device = torch.device("npu:0" if torch.npu.is_available() else "cpu")
tokenizer = AutoTokenizer.from_pretrained("Gxr777/EduNLP", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("Gxr777/EduNLP", torch_dtype=torch.float16, trust_remote_code=True).to(device).eval()
messages = [{"role": "user", "content": "什么是机器学习？"}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(device)
outputs = model.generate(**inputs, max_new_tokens=128, do_sample=False)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))

3. 精度评测

指标	值
Top-1 一致性	4/4
Max Logit Diff Ratio	8e-06
Avg KL Divergence	0.0
结论	PASS

4. 性能参考

NPU: 1722.85 ms（生成 64 个 tokens，fp16）