Suchismita Sahu's repositories
Lipreading-Using-Mutimodal-Speech-Recognition
Multimodal Speech Recognition for phoneme level prediction using Audio-Visual data from TCDTIMIT dataset implementing RNNs with LSTMs for the audio subnetwork and CNN-LSTMs for the video subnetwork.
Autoencoder-on-MNIST-in-Pytorch
The Autoencoder contains an encoder and decoder where encoder stores the images input in a compressed form and decoder retrieves back the Images.
Human-Activity-Recognition-using-Time-Series-Classification
Classified the activities of humans based on time series obtained by a Wireless Sensor Network.
Interpretable-FeedForward-CNNs
Interpretable Feedforward Convolutional Neural Networks: Implemented using Saab transform in cascade with Least Square Regressor for constructing the layers. Test on MNIST Handwritten digit dataset.
SIFT-Feature-Extraction-Texture-Analysis-and-Image-Matching
Implement texture classification and segmentation based on the 5x5 Laws Filters. Implement Scale Invariant Feature Transform (SIFT) which is an image feature extractor useful for representing the image information in a low dimensional form based on paper Lowe, David G. "Object recognition from local scale-invariant features." iccv. Ieee, 1999.. Used SIFT Features for Image Matching
AthenaHacks2019_starter
The starter project for AthenaHacks 2019 Unity Bootcamp.
athenahacks_demo
stuff for athenahacks
Color-Histogram-Manipulation-and-Denoising
Image Processing Algorithms for Conversion of of grayscale to color image, Histogram Manipulation methods for image contrast enhancement and Denoising for uniform, salt and pepper, and mixed noise.
nemo_asr_app
Spanish_ASR
Studying-the-Robustness-of-GANs-on-Noisy-Data-using-MNIST-and-EMNIST-Datasets
Pytorch implementation of GANs on MNIST dataset and testing the robustness of the GANs by incorporating Gaussian Noise and EMNIST dataset.
VQA-System
Visual Question Answering: given an image and a question related to the image in natural language, the systems answer the question in natural language from the image scene.
awesome-ai-cancer
Awesome artificial intelligence in cancer diagnostics and oncology
awesome-ai-cardiology
Awesome resources for artificial intelligence in cardiology
Banknote-Authentication-Dataset-Analysis-KNN
Banknote Authentication Dataset is used for Exploratory Analysis and KNN algorithm is used to classify, experimenting with different parameters for optimization and performance analysis.
CtCI-6th-Edition-Python
Cracking the Coding Interview 6th Ed. Python Solutions
ECG_GAN_MBD
This repository is for the paper "Synthesis of Realistic ECG using Generative Adversarial Networks".
Edge-Detection-and-Digital-Halftoning
Edge Detection Techniques and Digital half-toning methods and compare their performance.
Generative-Adversarial-Networks-using-MNIST
Implementation of Generative Adversarial Networks using the MNIST handwritten digits recognition dataset.
GenerativeAIExamples
Generative AI reference workflows optimized for accelerated infrastructure and microservice architecture.
MedicalNet
Many studies have shown that the performance on deep learning is significantly affected by volume of training data. The MedicalNet project provides a series of 3D-ResNet pre-trained models and relative code.
Megatron-LM
Ongoing research training transformer language models at scale, including: BERT & GPT-2
Multi-class-and-Multi-Label-Classification-Using-Support-Vector-Machines
Classification of Anuran species using Anuran Calls Dataset by implementing Support Vector Machines
On-the-Prediction-of-Apply-Rate
The problem of interest is the prediction of apply rate. Imagine a user visiting Glassdoor, and performing a job search.
Tensorflow-MultiGPU-VAE-GAN
A single jupyter notebook multi gpu VAE-GAN example with latent space algebra and receptive field visualizations.