Chunjiang Ge (葛春江)
Chunjiang Ge (葛春江)
### Describe the issue Issue: I find download ocr vqa images are slow and I implement a multi-thread download file. Command: ``` import concurrent.futures def download_image(k): ext = os.path.splitext(data[k]['imageURL'])[1] outputFile...
Does vlmeval support multi card inference and batch size > 1?
In the paper, you mention learning rate decay is important. I would like to know how could I set layer decay in the code. Do you have plans to release...