Aditya Oke
Aditya Oke
## 🚀 Feature I guess this project has matured a lot now. It has lot of default and configurations. Some of them are sadly hard-coded with the repository name being...
## 🚀 Feature Unsure if this might help in taking down CI test times ### Motivation Cache helps to reduce time required to fetch packages. This can reduce CI time....
## 🚀 Feature Slightly important as engines are concerned here :sweat_smile: Metrics parameter in training. ### Motivation What currently we do in `train_step`, `val_step` or `fit` API returns a dictionary...
## 🚀 Feature Implement YOLO v5 from `torch.hub`. This library removes such `dataset` abstraction and aims to provides a clean modular interface to models. Some key points to note: -...
## 🚀 Feature There is a lot of scope to improve the CNN Trainer. - [ ] LR Finder Implementation. - [ ] Metrics for Lightning (use Lightning metrics only)...
## 🚀 Feature A layers API for set of re-usable layers. ### Motivation Most models tend to re-use Layers. Note that this is different from components that we currently have....
- Adding testing for losses and layers Plan for layers - Check that we can use these modules. - Check values of known cases. Since layers are bit generic, Plan...
## 🐛 Bug **Describe the bug** I 'm training model with 2 class. Error appears at line 61 in `metrics/accuracy.py`. By default, `maxk` always equal 5 because `topk` is fixed...
## 🚀 Feature Let's not abuse the public FB Servers, check for pretrained weights in a cronjob, maybe once per day? ### Motivation Our CI is slow due to these...