Projects | Amit Pant

Paper Replication from Scratch using PyTorch

Tue, 20 Aug 2024 00:00:00 +0000

Paper replications for various classification and object detection tasks, all implemented from scratch using PyTorch. Additionally, I have pruned certain networks to enhance their real-time performance, allowing for a tunable trade-off between accuracy and efficiency.

FPGA-Optimized Neural Architecture Search for Enhanced Hardware Efficiency (FONAS)

Thu, 04 Apr 2024 00:00:00 +0000

Searched for the set of efficient deep neural architectures (FPGANets) for image classification with two constraints: arithmetic intensity and latency for FPGA. This approach outperformed many existing networks in terms of both latency and accuracy on ImageNet-1k.

EfficientNetV2 Optimization

Tue, 02 Jan 2024 00:00:00 +0000

EfficientNetV2 Optimization

Pruned 88% of channels of EfficientNetV2, resulting in a model that is 14x smaller, 2.5x faster, with 14x fewer parameters and just a 2% loss in accuracy.

Nepali Celeb Localized Face Dataset

Sat, 20 May 2023 00:00:00 +0000

The dataset contains images of 49 famous individuals from Nepal, including actors, actresses, singers, and social figures. The individuals vary in age, gender, and profession.

Potential uses for this dataset include face recognition, face detection, emotion recognition, and training machine learning models related to facial analysis.

For more details and to access the dataset, visit .

Procedure

Images of the actors were collected from the internet using a Python script.
The bing_image_downloader package was used to download images.
Images were manually verified for quality and unsuitable ones were removed.
OpenCV library was used to detect faces and crop images to extract only the face region.
The cropped face images were resized to a standard size.
Processed images were saved in a separate directory with appropriate naming convention.

If you’re interested in the technical details, you can check out the dataset on Kaggle and review the Python scripts used for data collection and preprocessing.

Lung Vision: X-ray-based Classification of Pulmonary Diseases

Tue, 04 Apr 2023 00:00:00 +0000

Prepared a dataset containing Normal, Tuberculosis (TB), and Pneumonia lung X-rays, and implemented a Deep Neural Network (DNN) to accurately classify these three categories. The challenge was to tackle the class imbalance inherent in the dataset.

Finance Portfolio Manager

Wed, 26 Oct 2022 00:00:00 +0000

Finance Portfolio Manager

Developed a Flask web app for stock trading, enabling user registration, querying for stocks, buying, selling, viewing the portfolio, and their transaction history with security features.