ABHIJIT PAL's repositories
image-mix-with-controlnet
A sample project to test out the features of streamlit. Provides a way to mix content and style of two images with help controlnet and clip-interrogator
TripplannerBot
This a streamlit app with langchain. It makes use of Bing maps API, OpenStreetMaps API and FourSquare API.
Projected-Gradient-Descent-with-CIFAR10
Implementation of PGD attack on a model trained on cifar10 dataset in TensorFlow. Also, FID between original images and generated images has been calculated.
DeepSAT-6-Satellite-Image-Classification
Using the pre-trained ImageNet models and cyclical learning rates, we tried to classify the DeepSAT-6 dataset (https://csc.lsu.edu/~saikat/deepsat/) into 6 categories (barren land, trees, grassland, roads, buildings and water bodies). Due to the absence of occlusion by the cloud, we dropped the NIR channel of the data.
Image2StyleGAN
Implementation of 'Image2StyleGAN'
abhijitpal1247
Profile
animatediff_exp
Experiments with Animate diff and LoRA
cleanlab
The standard data-centric AI package for data quality and machine learning with messy, real-world data and labels.
mindsdb
MindsDB connects AI models to databases.
MNIST-Retrain-with-DeepXplore
DeepXplore (https://arxiv.org/abs/1705.06640) is a white-box framework for testing deep neural networks. Here, I have used the examples generated by the framework to retrain LeNet-5, LeNet-4 and LeNet-1.
Neural-style-transfer
Implementation of A Neural Algorithm of Artistic Style (https://arxiv.org/pdf/1508.06576.pdf)
Practice
Various Practice sessions and notes
Real-time-object-detection-using-SURF
Real-time object tracking using SURF feature detection, implemented using MATLAB.
OGD-Data-Analysis
Performed data analysis on different OGD (Open Government Data) datasets
RAG-exp
This is an experimental repo for performing RAG experiments with LangChain
revise-tool
REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasets --- https://arxiv.org/abs/2004.07999
Sentiment-Analysis
Sentiment analysis on IMDB dataset using GloVe pre-trained embeddings and LSTM.