Mihir-Mavalankar

Stanford_CS_224N_NLPwithDeepLearning

DeepCubeA

Code for DeepCubeA, a Deep Reinforcement Learning algorithm that can learn to solve the Rubik's cube.

iMaterialist-Fashion-2020-at-FGVC7

For Kaggle Competition

Global-Wheat-Detction-Kaggle-2020

For the Kaggle Competition on object detection with same name. 1) models used are DETR, EfficientDet, YOLOv5, RetinaNet, FasterRCNN. 2) Ensemble inference using Weighted Box Fusion 3) Pseudo Learning to deal with smaller labeled datasets. 4) Mixup as an augmentation technique

pytorch-retinanet

Pytorch implementation of RetinaNet object detection.

EfficientDet

The pytorch re-implement of the official efficientdet with SOTA performance in real time and pretrained weights.

detr

End-to-End Object Detection with Transformers

roi_head_class

Stanford_CS234_RL

DeepRL_SymmetricAntGait