AdvancedLiterateMachinery icon indicating copy to clipboard operation
AdvancedLiterateMachinery copied to clipboard

A collection of original, innovative ideas and algorithms towards Advanced Literate Machinery. This project is maintained by the OCR Team in the Language Technology Lab, Tongyi Lab, Alibaba Group.

Results 143 AdvancedLiterateMachinery issues
Sort by recently updated
recently updated
newest added
trafficstars

训练命令: python main.py --task ctdet \ --dataset_name table \ --exp_id training_wtw \ --dataset_name WTW \ --image_dir /train/images \ --wiz_4ps \ --wiz_stacking \ --wiz_pairloss \ --tsfm_layers 3 \ --stacking_layers 3 \...

1. I’ve already set up the pretrained_folder. 2. When I try to use the weight(swin_base_patch4_window7_224_22k.pth) from that folder to resume, I get the following error (see error details below). 3....

The header in [`DocumentUnderstanding/VGT/object_detection/ditod/VGTbackbone.py`](https://github.com/AlibabaResearch/AdvancedLiterateMachinery/blob/main/DocumentUnderstanding/VGT/object_detection/ditod/VGTbackbone.py) states that the code is under GPL/Commercial dual licensing from ETRI. The actual implementation is very different from the [original code](https://github.com/youngwanLEE/MPViT/blob/main/detectron2/mpvit/backbone.py) so I am not sure...

一个小的代码修改意见: pt_seq_length=1024 所以在transformer.py decoder逐步输出inference结果时的for循环里 for i in range(self.args.pt_seq_length): i从0取到1023 embedding对于Position的编码应该也是0到1023 但是pt_hs = self.decode(pt_seq, memory, mask, pos_embed, 'pt')这一步输入到decode函数里的pt_seq,在i=0的时候,已经长度为7(应该是有prompt),所以会存在数组越界的情况。 数组越界就报错: RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasLtMatmul( ltHandle, computeDesc.descriptor(), &alpha_val, mat1_ptr, Adesc.descriptor(),...

能否提供LORE-TSR论文中的icdar2013和2019的标签,我自己将数据集标签修改为coco格式,测试识别的acc只有0.34,能否提供一下你们处理完成的数据集

In the DocXLayout README.md I cannot find the model_path/DocX_dla34_230829.pth, could you show me it?

Platypus 是否有计划开源训练代码?

I am able to train it with one 2080Ti with 12 GB of VRAM but I would have to resort to having a batch size of 1. Training with a...

Hi all, I'm attempting to export the VGT model to ONNX format, but I'm encountering an issue where I get the following error: ```arduino File ...\VGT\object_detection\ditod\VGTbeit.py:795 vis_x = checkpoint.checkpoint(blk, vis_x,...