marko1616

cmake编译报错

3

-- Building for: NMake Makefiles CMake Deprecation Warning at CMakeLists.txt:1 (cmake_minimum_required): Compatibility with CMake < 2.8.12 will be removed from a future version of CMake. Update the VERSION argument value...

在transformers>4.36.2的情况下合并Lora模型崩溃

2

### Reminder - [x] I have read the README and searched the existing issues. ### Reproduction 运行命令行： ``` python src/export_model.py --model_name_or_path “J:\AI\models\chinese-alpaca-2-13b” --adapter_name_or_path “J:\AI\models\chinese-alpaca-2-13b-lora\checkpoint-120” --finetuning_type lora --export_dir “J:\AI\models\markobot_merged” --template llama2_zh...

# What does this PR do? Fixes #2943 ## Before submitting - [x] Did you read the [contributor guideline](https://github.com/hiyouga/LLaMA-Factory/blob/main/.github/CONTRIBUTING.md)? 已知默认生成配置会导致保存问题的模型: - Llama2 - chinese-alpaca-2

Add template&support(tested)

3

# What does this PR do? Add support for #3184 ## Before submitting - [x] Did you read the [contributor guideline](https://github.com/hiyouga/LLaMA-Factory/blob/main/.github/CONTRIBUTING.md)?

比如从某一步在开始以一定频率保存，基于指标的保存etc.

[Bug] `ValueError: negative shift count` at calculating hand.(Test code included)

1

**Description:** I encountered a potential issue with the tile reduction logic used in the hand calculation process. The current implementation seems to decrement the tile count incorrectly, which might lead...

Feature/support qwenvl glm4-v phi3-v(conflict resolving)

15

# What does this PR do? Fixes #4375 ## Before submitting - [x] Did you read the [contributor guideline](https://github.com/hiyouga/LLaMA-Factory/blob/main/.github/CONTRIBUTING.md)? - [x] Did you write any new necessary tests?

## [Feature request]支持Qwen-VL ### 相关issue - 虽然这在 #2112 与 #2093 被提过，但目前的版本依旧不支持Qwen-VL。 ### Qwen-VL支持的必要性 - 在大部分图像任务中Qwen-VL的能力超越了LLAVA与YI-VL（Qwen数据集一直都更好） - 一个更通用的多模态支持?（GLM4的支持类似这个） - 可以嵌入多张图像 ### 实现相关 - 我已经写了个小 [测试](https://github.com/marko1616/LLaMA-Factory/tree/feature/Support-Qwenvl) (用于webui的chat) #### 图像嵌入方法 - 给template添加单独的Role/修改message格式(**glm4v**就是这么做的) -...

Fix Issue #4946 Basic folder upload for sftp electron.

3

# 🚀 Basic Folder Upload Feature for SFTP in Electron > **Status:** _Under Review_ ## 📝 Description I have implemented a **folder upload feature** for the SFTP functionality within Electron....

# 🚀 What does this PR do? Support Llama-3.2-11B-Vision. ## ✅ Before submitting - [x] Did you read the [contributor guideline](https://github.com/hiyouga/LLaMA-Factory/blob/main/.github/CONTRIBUTING.md)? - [x] Did you write any new necessary tests?...