Amit Pant's Projects
Config files for my GitHub profile.
I pruned 88% of channels of efficientNetV2 using LAMP pruning resulting 14x smaller model, 2.5x faster in inference, 14x fewer parameters & 2.5x fewer MACops. The loss in accuracy was somewhat 2% only.
All of my projects from data mining class along with reports
Classify if a banknote is geniune or forged based on 4 attributes using decion trees
Searching for Hardware Aware Neural Networks on ImageNet for FPGA in MobilenetV3 Search Space
Neural Networks for Handwritten Digit Recognition
Implementation and evaluation of ID3 and C4.5, decision tree algorithms from scratch
A simple multiplayer game usign pygame !
Reads the Track data from xml file and creates a relation database in SQL that stores all data
Newton Vs Einstien
Performing EDA on titanicn and heart disease dataset and implementing Naive bayes classifier
Face images of 49 famous individuals from Nepal, including actors, singers
[ICLR 2020] Once for All: Train One Network and Specialize it for Efficient Deployment
This is a simple pong game with somewhat different module. The intial score is given 100 and if score falls below zero, you lose the game. The goal is to hit the ball and increase the score!
PCA on random generated number, iris flower dataset and seed dataset and comparision with the scikitlearn pca.
Magnitude-based unstructured pruning to eliminate less valuable parameters from the network. I was able to prune 81% weights
A comparative analysis of different CNN Architectures for Pulmonary Disease Classification: Inception v3 vs. ResNet-50
Dequantizing the vitis ai quantized model to evalaute it's performance on the test set in pytorch.
Acess web codes and exercises from py4e
Quantization of efficientnet model that i pruned 81% using Vitis AI and compiling it for deployment on FPAG
simple quiz app, practice flutter
A simple encrypter and Decrypter
Vitis AI is Xilinxβs development stack for AI inference on Xilinx hardware platforms, including both edge devices and Alveo cards.
Pruning yolov5 based on ultralytics yolov5 implementation
Yolov6 Small Implementation from scratch on Pytorch
Pruning YOLOv7 using torch pruning