Aman Priyanshu's Projects
AdaptKeyBERT: keyword/keyphrase extraction with zero-shot and few-shot semi-supervised domain adaptation
My Profile
Anomaly detection without Machine Learning, cause that's not the only way
Sem 6: Android Lab - MIT Manipal B.Tech Information Technology (2019-23)
Attention in PyTorch
Running case hold for amrbart and text
Building a chatbot from scratch.
code to perform detailed data exploration for binary_classification
Base Classifiers might not be as viable an option, but often times the architectures relevant for even simple neural networks might not be present, therefore this repository aims to combine simple classifiers using binary combinations to produce better performances.
Experimenting with BiPedal Walker
Creating a BART backbone for - the [EMNLP 2020] Keep CALM and Explore: Language Models for Action Generation in Text-based Games
Creating a chatroom using python. Where multiple people can chat
An introductory exploration of PyGame
Creating a tensorflow Callback to allow messages of update over Colab.
My code for breaking or testing different ciphers they will include both the encryption function as well as the decryption function
Some data on Covid19
Decentralized Cross Device Learning for Hierarchical Aggregation of Medical Symptoms
Distributed denial of service (DDoS) attacks are a subclass of denial of service (DoS) attacks. A DDoS attack involves multiple connected online devices, collectively known as a botnet, which are used to overwhelm a target website with fake traffic. DDoS detection is the process of distinguishing distributed denial of service (DDoS) attacks from normal network traffic in order to perform effective attack mitigation.
Decision Tree through ID3 algorithm
DeCrise: Social Media platforms hold the potential to avert and support public utility services for crisis management. We use continual and federated learning to amplify our NLP model.
The aim of this repository is to create RBMs, EBMs and DBNs in generalized manner, so as to allow modification and variation in model types.
The Denoising Autoencoder is an extension of the autoencoder. Just as a standard autoencoder, itβs composed of an encoder, that compresses the data into the latent code, extracting the most relevant features, and a decoder, which decompress it and reconstructs the original input. There is only a slight modification: the Denoising Autoencoder takes a noisy image as input and the target for the output layer is the original input without noise.
Official code of our work, PolicyQA: A Reading Comprehension Dataset for Privacy Policies [Findings of EMNLP 2020].
Creating a Discord Bot which can allow use of Keras remotely. A python enabler will also be created.
This repo will mostly be simple articles wherein I explore topics of ML/DL. It won't be in detail but hopefully will help me someday, feel free to contribute if you have any topics you want to discuss.
DP-HyperparamTuning offers an array of tools for fast and easy hypertuning of various hyperparameters for the DP-SGD algorithm.
A collection of methodologies for synthetic table generation with differential privacy.