Deployment of sentiment analysis model using Amazon sagemaker

Computer Vision Nanodegree second project, generating image captions. We take a prebuilt Resnet CNN model to encode the image and captions, use the RNN with LSTM to train and decode images and predict captions

Udacity - Computer Vision Nanodegree - Project 2: Image Captioning

predicting country of origin from names

RAF: A RNN-based Chatbot Development Tool

基于pytorch的RNN、LSTM模型构建，RNN进行MNIST数据集分类，LSTM进行古诗生成

Neural Networks for Neuroscience Tasks

This is the 3rd project of the Deep Learning Nanodegree.

My solution to the Udacity project of training all Seinfeld Scripts in an RNN to predict a new script based on which character talks first

2018-2019 Semester2 at Soton, lab practice for Deep Learning

Predict whether the number of confirmed people will increase or not with the use of RNN.

Retrieved financial institutions data using Twitter’s API, created a database, and performed pre-processing of this data • Created an AI module which classified each tweet into three different sentiment groups: Positive, Negative and Neutral Developed a POC which predicted the stock price of these financial institutions based on the different se…

A fully playable snake game application that features an intelligent agent trained using Deep Recurrent Q-Learning

RNN-based generator of Polish-like words

The main task of the character-level language model is to predict the next character given all previous characters in a sequence of data, i.e. generates text character by character.

Classifying names of people into the language of origin.

identify the language of a given word

RNN-based security patch identification with oversampling samples. This is an extension code in the MILCOM'21 paper "PatchRNN: A Deep Learning-Based System for Security Patch Identification".