Yash Sandansing

@imliupu can you please tell me which dataset did you use to get such good results? Even the first iteration in my training leads to a semantic loss of more...

https://www.kaggle.com/code/daggerx/modnet?scriptVersionId=107316373 You'll need to modify some stuff a bit but mostly this is the code that I'm currently using for training.

@SamStark-AtWork I was the one who wrote this training code and even I couldn't get it to work later. The loss remains constant more or less throughout the training process....

@skyler14 sorry for the late reply. I did try experimenting with the training but it went horribly wrong with my training loop. **The images had random blobs in the center....

Refer https://github.com/AlibabaResearch/AdvancedLiterateMachinery/issues/75 for answer

The pr currently only handles creating grid from a machine readable pdf. For png most probably, you'd need to use a ocr to populate the fields in dict_for_grid

@wolfassi123 just managed get it working. this is my code for reference: ``` !pip install --quiet tf2onnx python-dotenv==1.0.0 torch==1.12.0 torchvision==0.13.0 git+https://github.com/deepdoctection/deepdoctection#egg=deepdoctection[hf] !pip install timm --quiet import deepdoctection as dd from...

try this one @wolfassi123 https://colab.research.google.com/drive/1nk7u8lQOJbO9H8xSVYTpzVdV8kuPAZla?usp=sharing have added additional libraries needed for this