swift icon indicating copy to clipboard operation
swift copied to clipboard

ms-swift: Use PEFT or Full-parameter to finetune 250+ LLMs or 25+ MLLMs

Results 206 swift issues
Sort by recently updated
recently updated
newest added

依据官方文档中的步骤对CodeQwen1.5-7B-Chat进行微调,流程全部跑通了,但是加载数据后并没有看出微调后的效果,甚至于自我认知微调都没有work ![image](https://github.com/modelscope/swift/assets/100772629/1341bdf3-e127-4833-bcd3-8854f9b00162) ```python sft_args = SftArguments( model_type=ModelType.codeqwen1half_7b_chat, model_id_or_path="qwen/CodeQwen1.5-7B-Chat", # dataset=[DatasetName.coig_cqia_chinese_traditional], custom_train_dataset_path=["F:/AAA/train.json"], train_dataset_sample=1000, logging_steps=5, max_length=2048, learning_rate=5e-5, warmup_ratio=0.4, output_dir='output', lora_target_modules=['ALL'], self_cognition_sample=500, dataloader_num_workers=0, model_name=['小黄', 'Xiao Huang'], model_author=['魔搭', 'ModelScope']) ``` train.json中只有1条数据,复制了100遍,训练完之后也没有work ```python # Experimental...

(swift) [sunkaijie@localhost swift]$ CUDA_VISIBLE_DEVICES=0 swift infer --model_type llava1d6-mistral-7b-instruct run sh: `python /home/sunkaijie/project/swift/swift/cli/infer.py --model_type llava1d6-mistral-7b-instruct` 2024-04-22 08:11:33,587 - modelscope - INFO - PyTorch version 2.2.2 Found. 2024-04-22 08:11:33,588 - modelscope -...

请问:在cli上运行时,如何使用自定义的数据集; 例如:微调MiniCPM-V时候,找不到对应的sft.sh,只找到了minicpm_2b_sft_chat和minicpm_moe_8x2b;然而利用cli: **# Experimental environment: A10, 3090, V100, ... **# 10GB GPU memory** CUDA_VISIBLE_DEVICES=0 swift sft \ --model_type minicpm-v-3b-chat \ --dataset coco-mini-en-2 \** 是否需要修改--dataset? 还是怎么做,我找不到相关方法,能否提供更加易懂和全面的信息及方法。

# PR type - [x] Bug Fix - [ ] New Feature - [ ] Document Updates - [ ] More Models or Datasets Support # PR information Write the...

**Describe the feature** Please describe the feature requested here(请在这里描述需求) **Paste any useful information** Paste any useful information, including papers, github links, etc.(请在这里描述其他有用的信息,比如相关的论文地址,github链接等) **Additional context** Add any other context or information...

**Describe the feature** Please describe the feature requested here(请在这里描述需求) **Paste any useful information** Paste any useful information, including papers, github links, etc.(请在这里描述其他有用的信息,比如相关的论文地址,github链接等) https://github.com/ymcui/Chinese-LLaMA-Alpaca-2 **Additional context** Add any other context or...

enhancement

你好 我看swift框架上都是对模型的微调,有预训练的功能吗? 请问,如何用swift对千问1.5-chat-14b模型预训练呢?数据格式有什么什么样子的呢? 或者有对应的预训练脚本吗? 谢谢!

question

**Describe the bug** 推理llama-3-8B模型显示没有这个类型的模型 ![image](https://github.com/modelscope/swift/assets/28260618/5ce84365-e23b-4883-b43d-a37f25e33b6c) **Your hardware and system info** 这个是因为还未支持llama3吗?看README说已经支持了 **Additional context**

question

**Describe the feature** Please describe the feature requested here(请在这里描述需求) **Paste any useful information** Paste any useful information, including papers, github links, etc.(请在这里描述其他有用的信息,比如相关的论文地址,github链接等) **Additional context** Add any other context or information...

**Describe the feature** 希望可以支持JetMoE系列 **Paste any useful information** https://huggingface.co/jetmoe/jetmoe-8b-chat