File size: 15,127 Bytes
eeca2ab 4d56635 93f09ae eeca2ab 4d56635 019ee51 a407009 650609c a407009 3d02821 69e8773 3d02821 76b530c 39cda0e 14d5f13 39cda0e 9032fbf 692637d 9032fbf 39cda0e d1ddbb0 650609c 76b530c 650609c bd9a4a1 650609c bd9a4a1 650609c bd9a4a1 650609c 76b530c 650609c ce17c94 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 |
---
license: bigscience-openrail-m
metrics:
- code_eval
library_name: transformers
tags:
- code
model-index:
- name: WizardCoder
results:
- task:
type: text-generation
dataset:
type: openai_humaneval
name: HumanEval
metrics:
- name: pass@1
type: pass@1
value: 0.573
verified: false
---
This is the Full-Weight of WizardCoder.
**Repository**: https://github.com/nlpxucan/WizardLM/tree/main/WizardCoder
**Twitter**: https://twitter.com/WizardLM_AI/status/1669109414559911937
**Paper**: [WizardCoder: Empowering Code Large Language Models with Evol-Instruct](https://arxiv.org/abs/2306.08568)
## WizardLM: Empowering Large Pre-Trained Language Models to Follow Complex Instructions
<p align="center">
π€ <a href="https://huggingface.co/WizardLM" target="_blank">HF Repo</a> β’ π¦ <a href="https://twitter.com/WizardLM_AI" target="_blank">Twitter</a> β’ π <a href="https://arxiv.org/abs/2304.12244" target="_blank">[WizardLM]</a> β’ π <a href="https://arxiv.org/abs/2306.08568" target="_blank">[WizardCoder]</a> <br>
</p>
<p align="center">
π Join our <a href="https://discord.gg/VZjjHtWrKs" target="_blank">Discord</a>
</p>
## News
- π₯ π₯ π₯ [08/11/2023] We release **WizardMath** Models.
- π₯ Our **WizardMath-70B-V1.0** model slightly outperforms some closed-source LLMs on the GSM8K, including **ChatGPT 3.5**, **Claude Instant 1** and **PaLM 2 540B**.
- π₯ Our **WizardMath-70B-V1.0** model achieves **81.6 pass@1** on the [GSM8k Benchmarks](https://github.com/openai/grade-school-math), which is **24.8** points higher than the SOTA open-source LLM.
- π₯ Our **WizardMath-70B-V1.0** model achieves **22.7 pass@1** on the [MATH Benchmarks](https://github.com/hendrycks/math), which is **9.2** points higher than the SOTA open-source LLM.
| Model | Checkpoint | Paper | GSM8k | MATH |Online Demo| License|
| ----- |------| ---- |------|-------| ----- | ----- |
| WizardMath-70B-V1.0 | π€ <a href="https://huggingface.co/WizardLM/WizardMath-70B-V1.0" target="_blank">HF Link</a> | π <a href="https://github.com/nlpxucan/WizardLM/blob/main/WizardMath/WizardMath_Paper.pdf" target="_blank">[WizardMath]</a>| **81.6** | **22.7** |[Demo (only support English)](http://47.103.63.15:50083/)| <a href="https://ai.meta.com/resources/models-and-libraries/llama-downloads/" target="_blank">Llama 2 </a> |
| WizardMath-13B-V1.0 | π€ <a href="https://huggingface.co/WizardLM/WizardMath-13B-V1.0" target="_blank">HF Link</a> | π <a href="https://github.com/nlpxucan/WizardLM/blob/main/WizardMath/WizardMath_Paper.pdf" target="_blank">[WizardMath]</a>| **63.9** | **14.0** |[Demo (only support English)](http://47.103.63.15:50082/)| <a href="https://ai.meta.com/resources/models-and-libraries/llama-downloads/" target="_blank">Llama 2 </a> |
| WizardMath-7B-V1.0 | π€ <a href="https://huggingface.co/WizardLM/WizardMath-7B-V1.0" target="_blank">HF Link</a> | π <a href="https://github.com/nlpxucan/WizardLM/blob/main/WizardMath/WizardMath_Paper.pdf" target="_blank">[WizardMath]</a>| **54.9** | **10.7** | [Demo (only support English)](http://47.103.63.15:50080/)| <a href="https://ai.meta.com/resources/models-and-libraries/llama-downloads/" target="_blank">Llama 2 </a>|
<font size=4>
| <sup>Model</sup> | <sup>Checkpoint</sup> | <sup>Paper</sup> |<sup>MT-Bench</sup> | <sup>AlpacaEval</sup> | <sup>WizardEval</sup> | <sup>HumanEval</sup> | <sup>License</sup>|
| ----- |------| ---- |------|-------| ----- | ----- | ----- |
| <sup>WizardLM-13B-V1.2</sup> | <sup>π€ <a href="https://huggingface.co/WizardLM/WizardLM-13B-V1.2" target="_blank">HF Link</a> </sup>| | <sup>7.06</sup> | <sup>89.17%</sup> | <sup>101.4% </sup>|<sup>36.6 pass@1</sup>|<sup> <a href="https://ai.meta.com/resources/models-and-libraries/llama-downloads/" target="_blank">Llama 2 License </a></sup> |
| <sup>WizardLM-13B-V1.1</sup> |<sup> π€ <a href="https://huggingface.co/WizardLM/WizardLM-13B-V1.1" target="_blank">HF Link</a> </sup> | | <sup>6.76</sup> |<sup>86.32%</sup> | <sup>99.3% </sup> |<sup>25.0 pass@1</sup>| <sup>Non-commercial</sup>|
| <sup>WizardLM-30B-V1.0</sup> | <sup>π€ <a href="https://huggingface.co/WizardLM/WizardLM-30B-V1.0" target="_blank">HF Link</a></sup> | | <sup>7.01</sup> | | <sup>97.8% </sup> | <sup>37.8 pass@1</sup>| <sup>Non-commercial</sup> |
| <sup>WizardLM-13B-V1.0</sup> | <sup>π€ <a href="https://huggingface.co/WizardLM/WizardLM-13B-V1.0" target="_blank">HF Link</a> </sup> | | <sup>6.35</sup> | <sup>75.31%</sup> | <sup>89.1% </sup> |<sup> 24.0 pass@1 </sup> | <sup>Non-commercial</sup>|
| <sup>WizardLM-7B-V1.0 </sup>| <sup>π€ <a href="https://huggingface.co/WizardLM/WizardLM-7B-V1.0" target="_blank">HF Link</a> </sup> |<sup> π <a href="https://arxiv.org/abs/2304.12244" target="_blank">[WizardLM]</a> </sup>| | | <sup>78.0% </sup> |<sup>19.1 pass@1 </sup>|<sup> Non-commercial</sup>|
| <sup>WizardCoder-15B-V1.0</sup> | <sup> π€ <a href="https://huggingface.co/WizardLM/WizardCoder-15B-V1.0" target="_blank">HF Link</a></sup> | <sup>π <a href="https://arxiv.org/abs/2306.08568" target="_blank">[WizardCoder]</a></sup> | || |<sup> 57.3 pass@1 </sup> | <sup> <a href="https://huggingface.co/spaces/bigcode/bigcode-model-license-agreement" target="_blank">OpenRAIL-M</a></sup> |
</font>
# WizardCoder: Empowering Code Large Language Models with Evol-Instruct
To develop our WizardCoder model, we begin by adapting the Evol-Instruct method specifically for coding tasks. This involves tailoring the prompt to the domain of code-related instructions. Subsequently, we fine-tune the Code LLM, StarCoder, utilizing the newly created instruction-following training set.
## News
- π₯ Our **WizardCoder-15B-v1.0** model achieves the **57.3 pass@1** on the [HumanEval Benchmarks](https://github.com/openai/human-eval), which is **22.3** points higher than the SOTA open-source Code LLMs.
- π₯ We released **WizardCoder-15B-v1.0** trained with **78k** evolved code instructions. Please checkout the [Model Weights](https://huggingface.co/WizardLM/WizardCoder-15B-V1.0), and [Paper]().
- 📣 Please refer to our Twitter account https://twitter.com/WizardLM_AI and HuggingFace Repo https://huggingface.co/WizardLM . We will use them to announce any new release at the 1st time.
## Comparing WizardCoder with the Closed-Source Models.
π₯ The following figure shows that our **WizardCoder attains the third position in this benchmark**, surpassing Claude-Plus (59.8 vs. 53.0) and Bard (59.8 vs. 44.5). Notably, our model exhibits a substantially smaller size compared to these models.
<p align="center" width="100%">
<a ><img src="https://raw.githubusercontent.com/nlpxucan/WizardLM/main/WizardCoder/imgs/pass1.png" alt="WizardCoder" style="width: 86%; min-width: 300px; display: block; margin: auto;"></a>
</p>
β**Note: In this study, we copy the scores for HumanEval and HumanEval+ from the [LLM-Humaneval-Benchmarks](https://github.com/my-other-github-account/llm-humaneval-benchmarks). Notably, all the mentioned models generate code solutions for each problem utilizing a **single attempt**, and the resulting pass rate percentage is reported. Our **WizardCoder** generates answers using greedy decoding and tests with the same [code](https://github.com/evalplus/evalplus).**
## Comparing WizardCoder with the Open-Source Models.
The following table clearly demonstrates that our **WizardCoder** exhibits a substantial performance advantage over all the open-source models. β**If you are confused with the different scores of our model (57.3 and 59.8), please check the Notes.**
| Model | HumanEval Pass@1 | MBPP Pass@1 |
|------------------|------------------|-------------|
| CodeGen-16B-Multi| 18.3 |20.9 |
| CodeGeeX | 22.9 |24.4 |
| LLaMA-33B | 21.7 |30.2 |
| LLaMA-65B | 23.7 |37.7 |
| PaLM-540B | 26.2 |36.8 |
| PaLM-Coder-540B | 36.0 |47.0 |
| PaLM 2-S | 37.6 |50.0 |
| CodeGen-16B-Mono | 29.3 |35.3 |
| Code-Cushman-001 | 33.5 |45.9 |
| StarCoder-15B | 33.6 |43.6* |
| InstructCodeT5+ | 35.0 |-- |
| WizardLM-30B 1.0| 37.8 |-- |
| WizardCoder-15B 1.0 | **57.3** |**51.8** |
β**Note: The reproduced result of StarCoder on MBPP.**
β**Note: The above table conducts a comprehensive comparison of our **WizardCoder** with other models on the HumanEval and MBPP benchmarks. We adhere to the approach outlined in previous studies by generating **20 samples** for each problem to estimate the pass@1 score and evaluate with the same [code](https://github.com/openai/human-eval/tree/master). The scores of GPT4 and GPT3.5 reported by [OpenAI](https://openai.com/research/gpt-4) are 67.0 and 48.1 (maybe these are the early version GPT4&3.5).**
## Call for Feedbacks
We welcome everyone to use your professional and difficult instructions to evaluate WizardCoder, and show us examples of poor performance and your suggestions in the [issue discussion](https://github.com/nlpxucan/WizardLM/issues) area. We are focusing on improving the Evol-Instruct now and hope to relieve existing weaknesses and issues in the the next version of WizardCoder. After that, we will open the code and pipeline of up-to-date Evol-Instruct algorithm and work with you together to improve it.
## Contents
1. [Online Demo](#online-demo)
2. [Fine-tuning](#fine-tuning)
3. [Inference](#inference)
4. [Evaluation](#evaluation)
5. [Citation](#citation)
6. [Disclaimer](#disclaimer)
## Online Demo
We will provide our latest models for you to try for as long as possible. If you find a link is not working, please try another one. At the same time, please try as many **real-world** and **challenging** code-related problems that you encounter in your work and life as possible. We will continue to evolve our models with your feedbacks.
## Fine-tuning
We fine-tune WizardCoder using the modified code `train.py` from [Llama-X](https://github.com/AetherCortex/Llama-X).
We fine-tune StarCoder-15B with the following hyperparameters:
| Hyperparameter | StarCoder-15B |
|----------------|---------------|
| Batch size | 512 |
| Learning rate | 2e-5 |
| Epochs | 3 |
| Max length | 2048 |
| Warmup step | 30 |
| LR scheduler | cosine |
To reproduce our fine-tuning of WizardCoder, please follow the following steps:
1. According to the instructions of [Llama-X](https://github.com/AetherCortex/Llama-X), install the environment, download the training code, and deploy. (Note: `deepspeed==0.9.2` and `transformers==4.29.2`)
2. Replace the `train.py` with the `train_wizardcoder.py` in our repo (`src/train_wizardcoder.py`)
3. Login Huggingface:
```bash
huggingface-cli login
```
4. Execute the following training command:
```bash
deepspeed train_wizardcoder.py \
--model_name_or_path "bigcode/starcoder" \
--data_path "/your/path/to/code_instruction_data.json" \
--output_dir "/your/path/to/ckpt" \
--num_train_epochs 3 \
--model_max_length 2048 \
--per_device_train_batch_size 16 \
--per_device_eval_batch_size 1 \
--gradient_accumulation_steps 4 \
--evaluation_strategy "no" \
--save_strategy "steps" \
--save_steps 50 \
--save_total_limit 2 \
--learning_rate 2e-5 \
--warmup_steps 30 \
--logging_steps 2 \
--lr_scheduler_type "cosine" \
--report_to "tensorboard" \
--gradient_checkpointing True \
--deepspeed configs/deepspeed_config.json \
--fp16 True
```
## Inference
We provide the decoding script for WizardCoder, which reads a input file and generates corresponding responses for each sample, and finally consolidates them into an output file.
You can specify `base_model`, `input_data_path` and `output_data_path` in `src\inference_wizardcoder.py` to set the decoding model, path of input file and path of output file.
```bash
pip install jsonlines
```
The decoding command is:
```
python src\inference_wizardcoder.py \
--base_model "/your/path/to/ckpt" \
--input_data_path "/your/path/to/input/data.jsonl" \
--output_data_path "/your/path/to/output/result.jsonl"
```
The format of `data.jsonl` should be:
```
{"idx": 11, "Instruction": "Write a Python code to count 1 to 10."}
{"idx": 12, "Instruction": "Write a Jave code to sum 1 to 10."}
```
The prompt for our WizardCoder in `src\inference_wizardcoder.py` is:
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{instruction}
### Response:
```
## Evaluation
We provide the evaluation script on HumanEval for WizardCoder.
1. According to the instructions of [HumanEval](https://github.com/openai/human-eval), install the environment.
2. Run the following script to generate the answer.
```bash
model="/path/to/your/model"
temp=0.2
max_len=2048
pred_num=200
num_seqs_per_iter=2
output_path=preds/T${temp}_N${pred_num}
mkdir -p ${output_path}
echo 'Output path: '$output_path
echo 'Model to eval: '$model
# 164 problems, 21 per GPU if GPU=8
index=0
gpu_num=8
for ((i = 0; i < $gpu_num; i++)); do
start_index=$((i * 21))
end_index=$(((i + 1) * 21))
gpu=$((i))
echo 'Running process #' ${i} 'from' $start_index 'to' $end_index 'on GPU' ${gpu}
((index++))
(
CUDA_VISIBLE_DEVICES=$gpu python humaneval_gen.py --model ${model} \
--start_index ${start_index} --end_index ${end_index} --temperature ${temp} \
--num_seqs_per_iter ${num_seqs_per_iter} --N ${pred_num} --max_len ${max_len} --output_path ${output_path}
) &
if (($index % $gpu_num == 0)); then wait; fi
done
```
3. Run the post processing code `src/process_humaneval.py` to collect the code completions from all answer files.
```bash
output_path=preds/T${temp}_N${pred_num}
echo 'Output path: '$output_path
python process_humaneval.py --path ${output_path} --out_path ${output_path}.jsonl --add_prompt
evaluate_functional_correctness ${output_path}.jsonl
```
## Citation
Please cite the repo if you use the data or code in this repo.
```
@misc{luo2023wizardcoder,
title={WizardCoder: Empowering Code Large Language Models with Evol-Instruct},
author={Ziyang Luo and Can Xu and Pu Zhao and Qingfeng Sun and Xiubo Geng and Wenxiang Hu and Chongyang Tao and Jing Ma and Qingwei Lin and Daxin Jiang},
year={2023},
}
```
## Disclaimer
WizardCoder model follows the same license as StarCoder. The content produced by any version of WizardCoder is influenced by uncontrollable variables such as randomness, and therefore, the accuracy of the output cannot be guaranteed by this project. This project does not accept any legal liability for the content of the model output, nor does it assume responsibility for any losses incurred due to the use of associated resources and output results. |