Sparsh Nagpal's repositories
Music-genre-classifier
Implementing different machine learning models on input music data and finding the most efficient method.
Ads-with-AI
A mega project that includes multiple features to enhance Ads and their Click-Through Rate. The llama-2 model will be fine-tuned for generating query captions. Stable Diffusion to generate Ad Creatives. Datasets like the Meta Ads dataset will help the product determine its target audience.
Indigo-Twitter-Analysis
Extracting live tweets through Tweepy library, analysing IndiGo airlines tagged tweets, Entity Recognition
Live-stocks-scraping
The above code pulls the live Nifty50 stocks statistics using web scraping
PDF-Query-LLM-with-VectorDB
This is a Python notebook that would guide you to build a Langchain LLM based model that intakes a pdf file, converts text into text chunks, which get converted to Text Embeddings and get saved to the VectorDB. VectorDB wold the using similarity search on vectors for quick response of the question answering.
Sketch-to-Face-using-Multi-GANs
A sketch to face image translation model which uses Contextual GANs, GFPGANs and DeOldify to generate results.
Sparks-Tasks-DataScience
Sparks Foundation internship tasks
Train-Rush-Predictor
The model takes data of train as input, cleans the data and get's output if the rush will be high, low or medium
Zero-shot-subject-classification
The above model uses transformers library with zero-shot-classification pipeline.
gpt3-whatsapp-bot
Using powerful services like OpenAI, Twilio, and Azure, I was able to create a system that scours the internet for topic-specific articles, summarizes them, and sends them straight to my WhatsApp.
Website-Q-A-using-RAG
Leveraging the power of the RAG model, the pipeline processes the extracted web data to generate informative responses, summaries, or answers to user queries.
auto-sklearn
Automated Machine Learning with scikit-learn
Basic-Langchain-QnA
Using LLM, OpenAI Keys and StreamLit, you can have a simple model for your Question Answering ChatBot
cats-vs-dogs-coursera-assignment
A coursera assignment to classify between images of dogs and cats using CNN in tensorflow with keras library.
Cover-Letter-Generator-using-Llama-2
Using Llama-2-7B quantized version (llama-2-7b ggml), I built a Streamlit code which can input commands about the job description and then revert with a suitable cover letter
Horses-vs-Humans-Transfer-Learning-CNN-Coursera-assignment
A CNN based coursera assignement using tensorflow where Horse and Human classification takes places using the means of transfer learning. The accuracy is measured upto 99.99% in the model. The user has to upload the image to view result.
igel
a delightful machine learning tool that allows you to train, test, and use models without writing code
KYSee-Online-KYC
KYC using ML and CV
Maths-write-and-play
A maths addition game where you'll get random numbers as input. You have to write their sum on the box and try to get it right. The code uses Tensorflow model of mnist dataset for handwriting recognition and flask to interact with html js for the values changing realtime.
sagar-viradiya
About me
streambook
A website to create a 'watchlist' of your favourite shows and movies
T098_Scrapshut
The given code is for providing user an access to our NGO website where they can see our demands and donate to our needs.
UW-IMT-589D-Building-and-Applying-Large-Language-Models
Repository for assignment solutions for the University of Washington's course IMT 589D - Building and Applying Large Language Models by Professor Chirag Shah