Yunfei Chu
Yunfei Chu
Here: https://github.com/OFA-Sys/OFA/tree/feature/add_transformers/transformers/src/transformers/models/ofa/generate
Yes, this is the result after finetuning the OFA-tiny model on Refcoco.
Thanks for your support! I just released a version of the pretrain script. Please try [pretrain.sh](https://github.com/OFA-Sys/OFA-Compress/blob/main/scripts/pretrain.sh). But I do not recommend using this framework for pre-training. We have tried pre-training...
Yes, the performance refers to training from scratch, and we haven't tried pertaining with new data from loaded ckpt. If you are interested, welcome to try it. In the future,...
您好，我们会尽快更新出来针对中文OFA的转换脚本，感谢关注！
> When do we expect to support qwen2 audio We expect to support qwen2audio within one month.
> Thanks for implementing this! Please run `format.sh` to fix the lint errors. @DarkLight1337 Already fixed them.
> To ease testing, can you also update `BaseMultiModalItemTracker._placeholder_str` inside `vllm/entrypoints/chat_utils.py` to support online inference? > > Also, it would be great if you could include this model in `examples/offline_inference_audio_language.py`!...
> @faychu 推理时间长的音频 输入的input.tokens长度不对 @umie0128 Qwen2audio only supports audio lengths of up to 30 seconds.