CameraTraps/roadmaps.md

59 строки
2.1 KiB
Markdown

### Development roadmap
Here you can find details of the milestone roadmap for **Pytorch-Wildlife**:
<!-- <img src="images/Finished_tasks_announcement.png" width="700"> -->
## October 0.1
✔ Packaging
- [x] Add the YOLOv5 detection model.<br>
- [x] Add the Megadetector pretrained weights.<br>
- [x] Add a Notebook to create, load and run inference on the YOLOv5 detection model using Megadetector pretrained weights.<br>
- [x] Add an animal species classification model and pretrained weights.
<!-- <img src="images/10_PyTorchWildlifeRoadmap_October.png" width="700"> -->
## November 0.2
✔ Packaging
- [x] Add a documented animal video detection pipeline.<br>
- [x] Add a user-friendly interface for animal detection.<br>
- [x] Add a module for custom datasets.<br>
- [x] Add a Jupyter notebook for the video detection tutorial.
<!-- <img src="images/11_PyTorchWildlifeRoadmap_November.png" width="700"> -->
## December 0.3
✔ Packaging
- [x] Add the video detection pipeline to the user interface.<br>
- [x] Add animal classification to the user interface.<br>
- [x] Complete documentation on how to use the user-interface.<br>
- [x] Add a module for fine-tuning and training from scratch.<br>
<!-- - [x] Add a dataset module to download LILA subsets.<br>
- [x] Add visualization functions for LILA subsets. -->
<!-- <img src="images/12_PyTorchWildlifeRoadmap_December.png" width="700"> -->
## January 0.4
✔ Packaging
- [x] Add a utilities module for data pre-processing and post-processing.<br>
- [x] Add Snapshot Serengeti classification weights.<br>
<!-- - [x] 2.1.3 Add a new animal classification dataset.<br> -->
<!-- ✔ Improvements
- [x] 2.1.2 Add a new animal detection dataset.<br>
- [x] 2.2.3 Implement a New detection model.<br>
- [x] 2.2.4 Implement a new classification model. -->
<!-- <img src="images/01_PyTorchWildlifeRoadmap_January.png" width="700"> -->
<!-- ## February 0.5
✔ Packaging -->
<!-- - [x] 2.1.3 Add a new classification dataset.<br> -->
<!-- - [x] Implement a new segmentation model.<br> -->
<!-- - [x] Add full documentation.<br> -->
<!-- <img src="images/02_PyTorchWildlifeRoadmap_February.png" width="700"> -->