Dog Breed Identification using CNNs

Digits drawn by hand and recognised by using a pre-trained model in android

Multi-Task Pre-Training for Plug-and-Play Task-Oriented Dialogue System (ACL 2022)

Visual Question Answering app with FastAPI and Docker

This is a web applcation built on react and fastapi. It takes an image of a dog as an input and uses deep learning to classify the emotion of dog in three classes (happy, sad, angry)

A car recognition model. It can recognize 11 types of cars

Labelling of images into one of a number of predefined classes using different CNN models.

We have trained CNN (Convolutional Neural Network) algorithm to predict sign language. There are also pretrained models which also used in this model.

Implementation of Salesforce's Moirai moel

Pre-trained Word Vector Models of 30+ Languages

Temple University, 2020 - Capstone project. Collected dog images and their projected breed(s) using the Petfinder API. Using the Stanford Dogs Dataset, created a multi-class, multi-label image classification model to predict the top 2 plausible dog breeds for each image.

2020 北京数据开放大赛

Image classification is still one of the interesting challenges for many data science practitioners today. In this project, I created several models to classify scene images where the models development are done in 2 ways, building the model from scratch and using pre-trained models. The data obtained from Kaggle, where the data was initially pu…

The official code for the "True Bilingual NMT" paper

Image classification using deep learning models with activation map visualisation and TensorRT support

pytorch中文语言模型预训练