sourjya261 Goto Github PK

brain_tumor_segmentation

cair_proposal

Proposed 1D Residual CNN architecture.

cell-nuclei-segmentation-using-simple-u-net

chipathy

Diabetic Retinopathy using Patch Networks.

cv_and_other_details

deep-learning-liver-segmentation

Application of Deep Learning to the segmentation of medical images

dpbh

flood_modelling_sih

hemt_simulation-optimization

This is a repository for my b tech final year project

idrid_ex_segmentation

Semantic segmentation of Hard-Exudates from fundus images using Residual UNet with squeeze and excite blocks and residual skip connections across encoder and decoder blocks.

lesion_segmentation

skin lesion segmentation using Res-unet. First segmentation project. have not implemented data augmentation.

livertumor_segmentation_w_net

neuromorphic_computing

nir-spectroscopy

explains the concept of using NIR Spectroscopy to non-invasively estimate the concentration of chemicals in biological specimens

summer_internship_brail_iitg

This repository holds all code snippets and notebooks that I used during my summer internship at BRAIL IITG under the supervision of Dr, Shyamanta. M. Hazarika. Problem Statement: Adaptation of EEG Motor Imagery Deterioration because of Mental Fatigue

u-net-for-multi-class-semantic-segmentation

verilog_projects

Assignments completed for Coursera course: FPGA Design for Embedded Systems Specialization, University of Boulder, Colorado

w-net-final

BS:16, init lr = 0.001, decay=0.3, patience=4, early_stopping = 9, asymmetric architecture

w-net-symmetric

W-Net Symmetric

w_net_idrid_ex

y-net

youtube_adview_prediction

Testing various ML models based on mse mae and rms error to find which model gives the best prediction. Decision tree regressor was found to be the best model.