nagakiran1 Goto Github PK

4-simple-steps-in-builiding-ocr

Optical character recognition (OCR) is process of classification of opti- cal patterns contained in a digital image. The character recognition is achieved through segmentation, feature extraction and classification. Keras Deep learning Network is used at here in recognising the Text characters and OpenCV is used in segmenting the text and Noise normalization.

carnd-advanced-lane-lines

carnd-behavioral-cloning-p3

Starting files for the Udacity CarND Behavioral Cloning Project

carnd-extended-kalman-filter-project

Self-Driving Car Nanodegree Program Starter Code for the Extended Kalman Filter Project

carnd-traffic-sign-classifier-project

Classify Traffic Signs.

classification_with_deep_nn

Deep Learning Neural Network is used in doing Classification problem.

crowd-counting-cnn-density-based

Project is to count the number group of people in an image and to count number of persons associated with each group. Here two enhanced Convolutional models have used in building effective crowd counting architecture.

data

just to keep active data

demo-repo

Some demo reposint

elsa-aad

essentials-of-data-structures

extending-google-bert-as-question-and-answering-model-and-chatbot

BERT Question and Answer system meant and works well for only limited number of words summary like 1 to 2 paragraphs only. It can’t be able to answer well from understanding more than 10 pages of data. We can extend the BERT question and answer model to work as chatbot on large text. To accomplish the understanding of more than 10 pages of data, here we have used a specific approach of picking the data.

face-recognition

Deep face recognition with Keras, Dlib and OpenCV

fausi-email-prioritization

Text preprocessing by removing stop words, URLs and Numbers which will not give much meaning to sentence. Creating a Feature Vector for each word in the Review by taking the word meaning from dictionary and using the vector to Sequential LSTM model Network to define the Class of Review Category which it belongs.

gpt-2

Code for the paper "Language Models are Unsupervised Multitask Learners"

keystem

to extract keywords from the documents and group the relevant clusters with single entity.

mask_rcnn

Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow

meshgraphnets

Rewrite deepmind/meshgraphnets into pytorch

meshgraphnets-1

PHYS 449 course project. In this repository, we attempt replication of results of the Learning Mesh Based Simulations with Graph Networks. Specifically the airfoil and flag experiments.

models

Models and examples built with TensorFlow

nagakiran1

nuhman-bititude

paddleocr

Awesome multilingual OCR toolkits based on PaddlePaddle (practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices)

receipt-image-classifier

Classifying the image either as Receipt or not, based on the considered Image patterns and the Extracted text associated with the considered Image.

semantic-feature-generation-for-words

Building a Natural language Responsive system by Preprocessing(removing stop words , numbers, urls and stemmming ) raw text, generating features for words, Extracting Entities(Named Entity Recognition) based on specific application, Feeding the preprocessed and required text to Deep Learning Neural Network which can generate a responsive sentence for the given sentence.

server

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

simplehtr_retrain

Handwritten Text Recognition (HTR) system implemented with TensorFlow.

transformers

🤗 Transformers: State-of-the-art Natural Language Processing for TensorFlow 2.0 and PyTorch.