Ishparsh Uprety's repositories
Nepali-Transliteration
While working with many Nepali documents, we encountered lots of data of Nepali names which includes names, surname, address and number Extracting the data was not a easy task but working with its romanize transliteration was hard. Many different packages are created for transliteration but they were not quite accurate. This package contains large amount of Nepali litral and words which are mapped to its respective romanized literal and word. But that was not the challenging part, still it was not giving the accurate result for instance "नेपाल" was showing Nepala as the "ल" is mapped as "la". So we have worked with these type of issues also.
River-Network-From-Satellite-Image-Using-Unet-Tensorflow
We presented our initial model for detecting rivers and watersheds from satellite pictures in this study, which was based on image processing methodology. The methodology was tested on a set of images obtained from the Sentinel-2 Satellite. For better recognition of rivers and watersheds, a new level of segmentation was used. We obtained a good accuracy using our proposed model, which is significantly higher than other U-net and Tensorflow implemented models available to date.
contact_form
simple student form web application
covid_prediction
This project is about covid prediction through X-Rays of chest
image_predict_cnn
Image Prediction by CNN model
PDF_to_image
This project is about how to convert the PDF file into image file.
predict_cnn
Predict Churn Model of Bank by ANN model
spring5webapp
Simple Web Application
Image-Alignment
When scanning a document, a slight skew gets into the scanned image. If you are using the scanned image to extract information from it, detecting and correcting skew is crucial.
Nepali-Transliteration-API
FastAPI is an excellent tool for putting your machine learning models into production. In this article, I briefly explain how you can easily put your FastAPI in production to an AWS EC2 instance using Nginx.
Object-Detection
This project is used to visualize the vehicles, person, traffic object using Yolo
Diabetes-Prediction-Using-Kmeans
In the beginning, the algorithm chooses k centroids in the dataset randomly after shuffling the data. Then it calculates the distance of each point to each centroid using the euclidean distance calculation method.
Face-Recognition-using-Picamera
This project is about how to deploy the project in Picamera for Face Recognition
getting-started
Getting started with Docker