deepseek-ai DeepSeek-Coder issues

deepseek-ai / DeepSeek-Coder

DeepSeek Coder: Let the Code Write Itself

https://coder.deepseek.com/

MIT License

5.98k stars 431 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Long Code Arena

#170 DifferentialityDevelopment opened 1 week ago
0
Where is DeepSeek-Coder-V2?

#169 RoacherM closed 1 week ago
0
RuntimeError: CUDA error: no kernel image is available for execution on the device

#168 TobiMoelti closed 2 days ago
0
为什么在进行一次训练加载后，会出现找不到显卡no slot的报错呢？

#167 ZhiyuYUE opened 2 weeks ago
0
Deepseekcoder 6 spitting out corrupt output for code generation question

#166 kodergeek opened 2 weeks ago
0
疑惑：为什么 base 模型的 tokenizer 词表中也有类似 <|Assistant|> 这样多用于 chat 模型的 special tokens？

#165 yucc-leon opened 3 weeks ago
0
训练数据切分问题

#164 sm307 opened 4 weeks ago
1
Are EOS tokens masked during pre-training? If so, how does FIM mode know how to connect to the `text_after`?

#163 zhzhangcc opened 1 month ago
0
Fix env name

#162 bigstomach opened 1 month ago
0
用vllm加速推理框架推理速度还是很慢

#161 zhuzhiwei88 opened 1 month ago
0
并发数目

#160 ChenVadder opened 1 month ago
0
使用vllm加载33b-base或33b-instruct后，使用DS-1000、Program-Aided Math Reasoning (PAL)评估集进行评估，得分很低，与论文上的数据不符

#159 aigc001 opened 2 months ago
0
使用vllm加速inference后输出容易不符合格式要求

#158 zhengrongz opened 2 months ago
0
How to use fine-tuned model?

#157 aldialimucaj opened 2 months ago
2
本地部署怎么实现vscode自动代码补全？

#156 lingyezhixing closed 2 months ago
1
微调完的模型，如何跟基础模型合并？

#155 libingbingd opened 2 months ago
1
markdown格式的数据预训练

#154 huangqingyi-code opened 2 months ago
3
请问支持function call吗？支持在RAG中实现inline citations吗？

#153 hiber-niu closed 2 months ago
0
What is the base context length of the model before extension to 16k?

#152 Calvinnncy97 closed 3 months ago
1
Why generate "GGGGG...." ,when the input string is longer than a certain length in GGUF model?

#151 hzgdeerHo opened 3 months ago
1
Does DeepSeek-Coder have wasm related knowledge?

#150 XinyuShe opened 3 months ago
1
使用react调用接口错误

#148 trookie2000 closed 3 months ago
0
clarification on the sentinel token format

#147 Zane-XY closed 2 months ago
0
Are NTP and FIM 2 separate stages of training, or are they combined?

#146 Calvinnncy97 closed 3 months ago
4
How can I do continue pretraining?

#145 hwaking opened 3 months ago
1
Fail to fine-tune V1.5 model with custom llama script

#144 lijierui closed 3 months ago
1
Align Scheduler Configuration with Finetuning Script

#143 richardodliu opened 3 months ago
0
33B inference too slowly

#142 ZJXNEFU opened 3 months ago
1
Leetcode数据集的构建脚本请问可以开源吗

#141 jzzzf opened 3 months ago
0
官方提供的微调训练脚本是否支持33B模型训练？(及训练相关问题)

#140 tongyuhome closed 2 months ago
1
如何构建微调的CoT数据

#139 wangqn1 opened 3 months ago
1
33B AWQ量化+vLLM部署问题

#138 CarolXh opened 3 months ago
0
Trying to finetune DeepSeek-Coder on custom Dataset

#137 A-Janj closed 3 months ago
13
chat completion任务时输出大量<|EOT|> token

#136 CarolXh closed 3 months ago
3
Complete missing `import`

#135 AntiQuality closed 3 months ago
0
Catastrophic forgetting problem

#134 shatealaboxiaowang opened 3 months ago
2
模型推理完成后怎么一直占用显存呢？

#133 chris-rong opened 3 months ago
2
Pretraining code

#132 Calvinnncy97 closed 3 months ago
2
Code to generate data

#131 tbressers opened 3 months ago
1
Reproduce FIM Evaluation

#130 Hambaobao opened 4 months ago
1
deepseek-coder-7b-base-v1.5 tokenizer=LlamaTokenizerFast 为什么分词会有很多乱码字符呢?

#129 zheng5yu9 opened 4 months ago
1
How is the amount of training data measured?

#128 WentaoChen0813 opened 4 months ago
1
Detailed version information of test programs in different languages.

#127 Hambaobao opened 4 months ago
0
Undefined variable in `Evaluation/MBPP/human_eval/evaluation.py`

#126 ya0guang closed 3 months ago
0
Question about training dataset

#125 TJ1999 opened 4 months ago
0
tokenizer.json issue creating gguf files

#124 RonanKMcGovern opened 4 months ago
2
Finetune of FIM

#123 shatealaboxiaowang opened 4 months ago
4
Swift and Objective C?

#122 rlaferla opened 4 months ago
2
How many tokens of code in pretraining

#121 bigeagle closed 4 months ago
2
fix in-page link for detailed eval results

#120 JacobLinCool closed 4 months ago
0