AmirHossein DabiriAghdam's Projects
Using a simple statistical model in order to predict whether the user has clicked on the Ad or not; Based on information about their daily internet usage, age, gender, daily time spent on site and etc.
Projects of Advanced Programming Course
Advanced Programming course - Final Project
Projects of Artificial Intelligence Course
BPE tokenizer from scratch + comparison of BPE and WordPiece from Hugging Face tokenizer on wikitext and All Around the Moon book from gutenberg
The repository for my B.Sc. Thesis
A chat room implemented using C++ and socket programming
Diagnosis of COVID-19 and Pneumonia from lung X-ray scans by training a Feed Forward Neural Network implemented using Tensorflow (Keras).
Config files for my GitHub profile.
A beautiful, simple, clean, and responsive Jekyll theme for academics
A weather station based on ESP32 microcontroller and MQTT protocol with Android/Windows/Web client applications.
Flight Booking Chatbot NLU - Summer Internship Project
Exploratory dataset analysis and implementation of some regression techniques for Kaggle “House Prices - Advanced Regression Techniques” Competition.
Projects of Instrumentation Course
Using the genetic algorithm to finding combinations of gates (AND/OR/XOR) to satisfy a given truth table.
Projects of Mechatronics Engineering Course
Mono and multi-lingual transformer based models (BERT, ParsBERT, XLM-Roberta) for textual entailment (ParsiNLU dataset), and text classification using transfer learning (MIZAN dataset) tasks
Using Naive Bayes classifier implemented from scratch for classification on sms spam collection and sentimental liar datasets + preprocessing and feature extraction
Using the Naive Bayes algorithm with additive smoothing, comments of Digikala dataset classified into two classes: recommended and not recommended. several pre-processing algorithms used before applying Naive Bayes.
Projects of Natural Language Processing Course
A Feed Forward Neural Network implementation from scratch + fine-tuning its hyperparameters and training it on Fashion MNIST Dataset.
Optimal Vehicle Routing (finding the best route with the min cost in terms of distance, etc.)
POS tagging and NER using Viterbi algorithm (implemented from scratch) and LSTM/GRU (implemented using PyTorch) for Penn Treebank dataset
Using Uninformed Search and Informed Search algorithms such as BFS, IDS, and A* to find suitable solutions for snake and food game.