Soumik Rakshit

Modifications added in this PR: - Download the model automatically from the release based on an alias - Download the image for inference from a URL - `tools/infer.py` and `yolov6/core/inferer.py`...

# Pull requests instruction 1. Put the bibtex in `awesome-NeRF/citations/.txt`. 2. Modify the README and follow the format TITLE, AUTHOR, CONFERENCE YEAR | OPTIONAL LINK | BIBTEX. For example: -...

Train, Validate and Detect using Weights & Biases Artifacts

1

## Features Added - Pass a model/checkpoint artifact (such as `wandb/hand-detection-yolov5/run_2mza8br0_model:v0`) to the `weights` parameter of `train.py` to load the weights from the artifact to the model for training -...

# Utilities Added - A function `rle_to_mask2d` is added that converts a Run-length encoded segmentation map to a 2-dimensional segmentation map. This function was adapted from [this StackOverflow answer](https://stackoverflow.com/a/58842473/8209968). -...

Integration for Weights and Biases

4

This PR includes the following features: - Streamlining [main.py](https://github.com/ZPdesu/Barbershop/blob/main/main.py) to include the logic from [align_face.py](https://github.com/ZPdesu/Barbershop/blob/main/align_face.py). - Updated the command line interface for [main.py](https://github.com/ZPdesu/Barbershop/blob/main/main.py). - Formatting the codebase using [black](https://github.com/psf/black). -...

Module.tabulate to return the Table

2

Updated `treex.Module.tabulate` to accept an additional parameter `return_table` which when set to `True` would return the `rich.table.Table` object.

Hi, I was trying to follow the excellent article by @straaljager. While trying to run this codebase using Cuda 11, I am getting the following error: ```main.cu(191): error: initial value...

Weights & Biases initialization should be decoupled from the Logger

3

Currently, a Weights & Biases run is initialized inside the constructor of [`WandBSGLogger`](https://github.com/Deci-AI/super-gradients/blob/master/src/super_gradients/common/sg_loggers/wandb_sg_logger.py#L100). However, if the run initialization is not decoupled from the logger, it would not be possible for...

Log inference results to Weights & Biases

1

Optionally log inference results in [Weights & Biases](https://wandb.ai/site) dashboard using [bounding-box image overlays](https://docs.wandb.ai/guides/track/log/media#image-overlays). Add an optional parameter `visualize_on_wandb` to `ImagesDetectionPrediction.show()` which when set to `True` would log the predictions using...

Update Weights & Biases Integration + Add Prediction Visualization for Object Detection

2

# Changes made in this PR ## Log inference results to Weights & Biases Fixes https://github.com/Deci-AI/super-gradients/issues/953 Optionally log inference results for object detection in [Weights & Biases](https://wandb.ai/site) dashboard using [bounding-box...