README.md

metadata

frameworks:
  - Pytorch
license: other
tasks:
  - text-generation

Model Card for CodeFuse-QWen-14B

[中文] [English]

Model Description

CodeFuse-QWen-14B is a 14B Code-LLM finetuned by QLoRA of multiple code tasks on the base model StarCoder.

News and Updates

🔥🔥 2023-10-16 CodeFuse-QWen-14B has been released, achieving a pass@1 (greedy decoding) score of 48.78% on HumanEval, which is a 16% increase compared to Qwen-14b's 32.3%.

🔥🔥 2023-09-27 CodeFuse-StarCoder-15B has been released, achieving a pass@1 (greedy decoding) score of 54.9% on HumanEval, which is a 21% increase compared to StarCoder's 33.6%.

🔥🔥🔥 2023-09-26 We are pleased to announce the release of the 4-bit quantized version of CodeFuse-CodeLlama-34B. Despite the quantization process, the model still achieves a remarkable 73.8% accuracy (greedy decoding) on the HumanEval pass@1 metric.

🔥🔥🔥 2023-09-11 CodeFuse-CodeLlama34B has achived 74.4% of pass@1 (greedy decoding) on HumanEval, which is SOTA results for openspurced LLMs at present.

Code Community

Homepage: 🏡 https://github.com/codefuse-ai (Please give us your support with a Star🌟 + Fork🚀 + Watch👀)

• If you wish to fine-tune the model yourself, you can visit ✨MFTCoder✨✨

• If you wish to deploy the model yourself, you can visit ✨FasterTransformer4CodeFuse✨✨

• If you wish to see a demo of the model, you can visit ✨CodeFuse Demo✨✨

Performance

Code

Model	HumanEval(pass@1)	Date
CodeFuse-CodeLlama-34B	74.4%	2023.9
CodeFuse-CodeLlama-34B-4bits	73.8%	2023.9
WizardCoder-Python-34B-V1.0	73.2%	2023.8
GPT-4(zero-shot)	67.0%	2023.3
PanGu-Coder2 15B	61.6%	2023.8
CodeLlama-34b-Python	53.7%	2023.8
CodeLlama-34b	48.8%	2023.8
GPT-3.5(zero-shot)	48.1%	2022.11
OctoCoder	46.2%	2023.8
StarCoder-15B	33.6%	2023.5
Qwen-14b	32.3%	2023.10
CodeFuse-StarCoder-15B	54.9%	2023.9
CodeFuse-QWen-14B	48.78%	2023.10

NLP

Requirements

• python>=3.8
• pytorch>=2.0.0
• transformers==4.32.0
• Sentencepiece
• CUDA 11.4
  

Inference String Format

The inference string is a concatenated string formed by combining conversation data(system, human and bot contents) in the training data format. It is used as input during the inference process. Here is an example format of the concatenated string:

"""
<s>system
System instruction
<s>human
Human 1st round input
<s>bot
Bot 1st round output<|endoftext|>
<s>human
Human 2nd round input
<s>bot
Bot 2nd round output<|endoftext|>
...
...
...
<s>human
Human n-th round input
<s>bot
{Bot output to be genreated}<|endoftext|>
"""

When applying inference, you always make your input string end with "<s>bot" to ask the model to generate answers.

Quickstart

pip install -r requirements.txt

import torch
from transformers import (
    AutoTokenizer, 
    AutoModelForCausalLM
)
tokenizer = AutoTokenizer.from_pretrained('codefuse-ai/CodeFuse-QWen-14B', trust_remote_code=True)
tokenizer.padding_side = "left"
tokenizer.pad_token_id = tokenizer.convert_tokens_to_ids("<|endoftext|>")
tokenizer.eos_token_id = tokenizer.convert_tokens_to_ids("<|endoftext|>")
tokenizer.pad_token = "<|endoftext|>"
tokenizer.eos_token = "<|endoftext|>"
# try 4bit loading if cuda memory not enough
model = AutoModelForCausalLM.from_pretrained(model_dir,
                                             trust_remote_code=True,
                                             load_in_4bit=False,
                                             device_map="auto",
                                             torch_dtype=torch.bfloat16)
model.eval()

HUMAN_ROLE_START_TAG = "<s>human\n"
BOT_ROLE_START_TAG = "<s>bot\n"

text = f"{HUMAN_ROLE_START_TAG}write a python function of quick sort.\n{BOT_ROLE_START_TAG}" 
inputs = tokenizer(text, return_tensors='pt', padding=True, add_special_tokens=False).to("cuda")
outputs = model.generate(
        inputs=inputs["input_ids"],
        attention_mask=inputs["attention_mask"],
        max_new_tokens=512,
        top_p=0.95,
        temperature=0.1,
        do_sample=True,
        eos_token_id=tokenizer.eos_token_id,
        pad_token_id=tokenizer.pad_token_id
    )
gen_text = tokenizer.batch_decode(outputs[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)
print(gen_text)

Citation

If you find our work useful or helpful for your R&D works, please feel free to cite our paper as below.

@article{mftcoder2023,
      title={MFTCoder: Boosting Code LLMs with Multitask Fine-Tuning}, 
      author={Bingchang Liu and Chaoyu Chen and Cong Liao and Zi Gong and Huan Wang and Zhichao Lei and Ming Liang and Dajun Chen and Min Shen and Hailian Zhou and Hang Yu and Jianguo Li},
      year={2023},
      journal={arXiv preprint arXiv},
      archivePrefix={arXiv},
      eprint={2311.02303}
}

模型简介

CodeFuse-QWen-14B 是一个通过QLoRA对基座模型QWen-14B进行多代码任务微调的代码大模型。

新闻

🔥🔥 2023-10-16开源了CodeFuse-QWen-14B模型，在HumanEval pass@1(greedy decoding)上可以达到48.78%, 比Qwen-14b提高了16%的代码能力（HumanEval）

🔥🔥 2023-09-27开源了CodeFuse-StarCoder-15B模型，在HumanEval pass@1(greedy decoding)上可以达到54.9%, 比StarCoder提高了21%的代码能力（HumanEval）

🔥🔥🔥 2023-09-26 CodeFuse-CodeLlama-34B 4bits量化版本发布，量化后模型在HumanEval pass@1指标为73.8% (贪婪解码)。

🔥🔥🔥 2023-09-11 CodeFuse-CodeLlama-34B发布，HumanEval pass@1指标达到74.4% (贪婪解码), 为当前开源SOTA。

代码社区

大本营： 🏡 https://github.com/codefuse-ai （请支持我们的项目Star🌟 + Fork🚀 + Watch👀）

• 如果您想自己微调该模型，可以访问 ✨MFTCoder✨✨

• 如果您想自己部署该模型，可以访问 ✨FasterTransformer4CodeFuse✨✨

• 如果您想观看该模型示例，可以访问 ✨CodeFuse Demo✨✨

评测表现

代码

模型	HumanEval(pass@1)	日期
CodeFuse-CodeLlama-34B	74.4%	2023.9
CodeFuse-CodeLlama-34B-4bits	73.8%	2023.9
WizardCoder-Python-34B-V1.0	73.2%	2023.8
GPT-4(zero-shot)	67.0%	2023.3
PanGu-Coder2 15B	61.6%	2023.8
CodeLlama-34b-Python	53.7%	2023.8
CodeLlama-34b	48.8%	2023.8
GPT-3.5(zero-shot)	48.1%	2022.11
OctoCoder	46.2%	2023.8
StarCoder-15B	33.6%	2023.5
Qwen-14b	32.3%	2023.10
CodeFuse-StarCoder-15B	54.9%	2023.9
CodeFuse-QWen-14B	48.78%	2023.8

NLP

Requirements

• python>=3.8
• pytorch>=2.0.0
• transformers==4.32.0
• Sentencepiece
• CUDA 11.4
  

推理数据格式

推理数据为模型在训练数据格式下拼接的字符串形式，它也是推理时输入prompt拼接的方式：

"""
<s>system
这是System指令
<s>human
这是第1轮用户输入的问题
<s>bot
这是第1轮模型生成的内容<|endoftext|>
<s>human
这是第2轮用户输入的问题
<s>bot
这是第2轮模型生成的内容<|endoftext|>
...
...
...
<s>human
这是第n轮用户输入的问题
<s>bot
{模型现在要生成的内容}<|endoftext|>
"""

推理时，请确保拼接的prompt字符串以"<s>bot\n"结尾，引导模型生成回答。

快速使用

pip install -r requirements.txt

import torch
from transformers import (
    AutoTokenizer, 
    AutoModelForCausalLM
)
tokenizer = AutoTokenizer.from_pretrained('codefuse-ai/CodeFuse-QWen-14B', trust_remote_code=True)
tokenizer.padding_side = "left"
tokenizer.pad_token_id = tokenizer.convert_tokens_to_ids("<|endoftext|>")
tokenizer.eos_token_id = tokenizer.convert_tokens_to_ids("<|endoftext|>")
tokenizer.pad_token = "<|endoftext|>"
tokenizer.eos_token = "<|endoftext|>"
# try 4bit loading if cuda memory not enough
model = AutoModelForCausalLM.from_pretrained(model_dir,
                                             trust_remote_code=True,
                                             load_in_4bit=False,
                                             device_map="auto",
                                             torch_dtype=torch.bfloat16)
model.eval()

HUMAN_ROLE_START_TAG = "<s>human\n"
BOT_ROLE_START_TAG = "<s>bot\n"

text = f"{HUMAN_ROLE_START_TAG}write a python function of quick sort.\n{BOT_ROLE_START_TAG}" 
inputs = tokenizer(text, return_tensors='pt', padding=True, add_special_tokens=False).to("cuda")
outputs = model.generate(
        inputs=inputs["input_ids"],
        attention_mask=inputs["attention_mask"],
        max_new_tokens=512,
        top_p=0.95,
        temperature=0.1,
        do_sample=True,
        eos_token_id=tokenizer.eos_token_id,
        pad_token_id=tokenizer.pad_token_id
    )
    
gen_text = tokenizer.batch_decode(outputs[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)
print(gen_text)