-
1. I used `llamafactory-cli train examples/train_lora/qwen2vl_lora_sft.yaml` to train a model
2. I used `llamafactory-cli export examples/merge_lora/qwen2vl_lora_sft.yaml` to get merged model
3. `g…
-
If I run the following with the image below, Python uses around 56GB memory and the generation speed dramatically decreases as it generates more tokens.
Is this normal? It seems pretty high usage for…
-
SIZE_FACTOR=16 MAX_PIXELS=301056 CUDA_VISIBLE_DEVICES=0,1,2,3 NPROC_PER_NODE=4 swift sft \
--model_type qwen2-vl-7b-instruct \
……
--freeze_vit true \
--sft_type full \
--learning_rate …
-
hi, I have images like this
,after fine-tuning , if I give image later like this .
![Screenshot_85]…
-
### System Info / 系統信息
cuda: 12.2（安装的pytorch是cuda-12.1的）
transformers：4.44.0
python：3.10
OS：kylinV10
显卡是：NVIDIA A100-SXM4-40GB（隔壁有问是不是这个问题，所以一并列出：https://github.com/QwenLM/Qwen2-VL/issues/44 ）
…
-
### Proposal to improve performance
Only perform image preprocessing once
### Report of performance regression
_No response_
### Misc discussion on performance
test code: https://…
-
非常感谢您们的工作，可以麻烦回复下精度对不齐的问题吗？https://github.com/QwenLM/Qwen2-VL/issues/27
-
### Reminder
- [X] I have read the README and searched the existing issues.
### System Info
- `llamafactory` version: 0.9.1.dev0
- Platform: Linux-5.15.0-119-generic-x86_64-with-glibc2.31
- Pytho…
J0eky updated
2 months ago
-
At current time 1514-0902-2024, the LLaMA-Factory repo doesn't contain qwen2vl_lora_dpo.yaml which is menti…
-
### Proposal to improve performance
_No response_
### Report of performance regression
build with latest vllm code and start Qwen2-VL-7B-Instruct
![image](https://github.com/user-attachments/ass…