abhilash1910/Accelerated_Deep_Learning

Stars
6
Rank 2,539,965 (Top 51 %)
Language
Jupyter Notebook
Created over 2 years ago
Updated almost 2 years ago

abhilash1910/Accelerated_Deep_Learning

abhilash1910

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

SpectralEmbeddings

spectralembeddings is a python library which is used to generate node embeddings from Knowledge graphs using GCN kernels and Graph Autoencoders. Variations include VanillaGCN,ChebyshevGCN and Spline GCN along with SDNe based Graph Autoencoder.

ClusterTransformer

Topic clustering library built on Transformer embeddings and cosine similarity metrics.Compatible with all BERT base transformers from huggingface.

BERTSimilarity

Sentential Semantic Similarity measurement library using BERT Embeddings for spatial distance evaluation.

Deep_Reinforcement_Learning_Trading

Deep Reinforcement Learning for Trading

Jupyter Notebook

nd013-c1-vision-starter-main

Course Submission Material for Resnet based object detection for Udacity Self Driving NanoDegree.

Jupyter Notebook

Deep-Graph-Learning

A notebook containing implementations of different graph deep node embeddings along with benchmark graph neural network models in tensorflow. This has been taken from https://www.kaggle.com/abhilash1910/nlp-workshop-ml-india-deep-graph-learning to apply GNNs/node embeddings on NLP task.

Jupyter Notebook

nd0130-c2-fusion-starter-midterm

Course submission material for Lidar point cloud based 3D Detection using Yolo for Udacity Self Driving Nanodegree

MiniAttention

Hierarchical Attention Layer for Keras for document classification.

Jupyter Notebook

K-means-clustering-for-WSN

CPP code for K means clustering for wireless sensor networks which are to be applied to optimal path to sink algorithm.

Unite-India-2019

The github repository containing all the shader optimizations and codes presented in Unite-India-2019

DistilBERT--SQuAD-v1-Notebook

Notebook (Demonstration) for training Distilbert on Glue, and uploading model to Huggingface.

Jupyter Notebook

MiniClassifier

Deep Learning Library for Text Classification.

Jupyter Notebook

Transformers-Workshop

Transformers Workshop on behalf of ML India. Contains resource notebook for training/inferring large scale transformer models for different downstream task

Jupyter Notebook

Playing-With-Transformers--NLP-Workshop

Jupyter Notebook

ndd0130-c2-fusion-starter

Course submission material for Sensor Fusion and Camera based tracking using Extended Kalman Filters for Udacity Self Driving Nanodegree.

Q-Learning-in-Unity

Template for Q-Learning in Unity with C# for developing state based games

GraphAttentionNetworks

This package is a Tensorflow2/Keras implementation for Graph Attention Network embeddings and also provides a Trainable layer for Multihead Graph Attention.

LongPegasus

LongPegasus package is used for inducing longformer self attention over base pegasus abstractive summarization model to increase the token limit and performance.

Jupyter Notebook

Datahour_Diffusion

Presentation of the session on Diffusion Principles on behalf of Analytics Vidhya datahour. Youtube: https://www.youtube.com/watch?v=DIdbsfzK7IUtube:

Multi-Armed-Bandit-C-

Contains C# implementation of the Multi-Armed-Bandit algorithm for RL .This uses standard UCB and epsilon greedy algorithm as a hybrid measure to trade between exploration and exploitation

NLP-Workshop-ML-India

NLP Workshop -ML India

Jupyter Notebook

NexGCN

Sigmoid Spectral GCN library for binary classification on Networkx Graphs using statistical (centrality,distributions)/external features

Jupyter Notebook

clip_zeroshot_tf2

TextProcessingApp

App is a text processing app made in android studio using google play services

Generative-Modelling

Jupyter Notebook

Unity-Shader-Source-Codes

Source codes of some of the most popular unity shaders

Face_Detection

Face detection of an image in drawable folder

NITDGPopensource-HashedSnapshots

A secure way to encrypt images based on hashed passwords

Math_Engine

Math_engine is a part of a central game engine which deals with vectors,matrix, and quarternion calculations.

Shaders

Contains source codes and implementations of many shaders in Unity

Render-Engine-2.0

OpenGL based render engine having separate 2d and 3d features with camera properties and normal lighting

Abhilash1_optimizers_package

Stochastic Optimizers for gradient convergence algorithms with activation functions

AI-Geometric-Learning

Jupyter Notebook

Simple-Raytracer

A primitive raytracer made using ray-sphere intersections and Lambert Cosine diffused reflection law.

DeepGenerator

Sentence Sequence Transduction Library (Seq to Seq) for text generation using sequential generative Vanilla RNN using numpy

Jupyter Notebook

Model-D-Engine-OpenGl

3d Model Engine made with opengl as an extension to the game engine development

Render_Engine

The render engine is an essential part of a game engine.It is based on Opengl API and is the primary design which allows user to create any coloured figures by specifying coordinate points in X-Y plane.All the rasterization complexity has been included in headers of the engine .More features are yet to be added.

Credit_Card_Fraud_Kaggle

This repository contains the possible solutions to the Credit Card Fraud problem set in Kaggle. It contains EDA, including dimensionality reduction,visualisations, sampling techniques , synthetic oversampling. A contrastive k fold analysis of decision trees,neighbors classifier, regression and discriminant analysis along with their confusion matrices. Alternative solutions using boosting trees and forests(lightgbm classifiers) are also used. A neural network of 4 intermediate hidden odes with relu and sigmoid activations is also provided as an alternative approach to get a higher accuracy.

Jupyter Notebook

EuroPython-21-QuantumDeepLearning

Materials and Resources for the EuroPython 21 Talk on "Introduction to Quantum Deep Learning" on 28/7/2021.

Jupyter Notebook

TFModelQuantizer

A model weights quantization library which converts weights to fp32, fp16 or calibrated int8 for faster inference using Nvidia Tensor-rt backend . Compatible with vision models inside keras applications.