MagicSource

Hi, I have finished training. I found the qwen4b can get a resonable performance:  But still OCR ability not very good, any suggestion to enhance OCR ability?> (Chinese open...

Seems like inputs have None. check the data or add some assersations.

For anyone who want get imediately response and help for training llava, can join this group:  if QRCode outdated, add `bojuebot` for invitation.

I think base can not be used in vlm, it doens't have chat abilities,.

@jiangjiajun I am also confused a lot..... ``` paddle2onnx --model_dir .\human_pp_humansegv2_lite_192x192_inference_model\ --model_filename model.pdmodel --params_filename model.pdiparams --save_file ppsegv2_lite_192x192.onnx --opset_version 11 --input_shape_dict "{'image':[1, 3, 192, 192]}" 2023-04-08 16:05:56 [WARNING] [Deprecated] The flag...

@evanjs Its not work, **lsd** is not working anymore unless problem solved on windows10

@alixander Oh, am using the default rendering engine, how does it support orientation per group? Also, when will all these 2 features be support? I wanna change the padding between...

Which version start this fixed? Am using 3.47.2 still get this error.

It's really hard to suggest users to downgrade cuda version to support pytorch3d, this is not the right way nor the proper way. CUDA is much more basic in whole...

@bottler Hello, no offense here. I don't think the issue you linked here the user one claimed solved its the issue poster, nor he does using CUDA11.8 on windows. I...