Help writing the Documentation

[Zrahay]

Write a clear documentation for the whole project for bringing clarity to other users who want to use our product.

[Abhislit]

Hello @Zrahay,

I would like to work on this issue and contribute to improving the project documentation. I can create clear, documentation that explains the project overview, setup, usage, and contribution guidelines.

Please assign this issue to me if it is available.
Thank you.

[Zrahay]

Yeah, this issue is still available. Assigning it to you! @Abhislit

[sm-28601]

Hi @Zrahay ,

Adding a few more concrete points I noticed while following the documentation step-by-step as a new user. These may be helpful to clarify in the docs:

1. Toolchain verification step
   The docs mention:

cd tools && chmod +x test_tools.sh && ./test_tools.sh

However, test_tools.sh is not present in the tools/ directory.
This makes the verification step confusing for new users.
It might help to either remove this step or replace it with explicit manual checks (e.g. clang --version, riscv64-linux-gnu-gcc --version, qemu-riscv64 --version).

2. Project structure clarity
   The repository contains multiple similarly named directories (data/, training_data/, preprocessing/, preprocessing_output/, inference_folder/), but there’s no explanation of which stage of the pipeline each corresponds to. A short section or diagram explaining the data flow would reduce confusion.

3. Model selection for inference
   There are multiple .pth files (execution_time, binary_size, best, final), but the docs don’t explain which model is intended for inference or how their outputs differ. This can lead to shape mismatch errors.

4. Git LFS requirement
   Large JSON/model files are tracked via Git LFS, but it’s not clearly stated that users must run git lfs pull after cloning. Without this, files appear corrupted.

I’d be happy to help document or fix any of these points if useful.
Thanks!

[Incharajayaram]

@sm-28601 yes these points are right
we were expecting the same from this issue
please document whatever you feel is right
and we will check and remove the things which aren't relevant

[Incharajayaram]

@sm-28601 also can you please start a discussion in the discussions tab? we can further clear your doubts there

[sm-28601]

Thanks for the clarification @Incharajayaram 🙌

Sure, I’ll start a discussion in the Discussions tab to consolidate the setup and inference-related points, and then document the relevant ones accordingly.

I’ll keep the changes focused on common setup issues faced by new users.
Thanks for the guidance!

[sm-28601]

Hi @Zrahay @Incharajayaram @rishabhlakhotia ,

I’ve created a PR that updates CONTRIBUTING.md to clarify the local setup and testing steps I faced as a new contributor.
It covers points like Git LFS usage, manual toolchain verification, current project structure, and the actual testing workflow.

Please have a look and let me know if any part should be adjusted or removed. Happy to refine it further.
Thanks!