Bert Model for Text Classification
I use the Bert Pretrained model and fine tunned it for Toxic Comment Classification organized by Jigsaw/Conversation AI. I had also got my first Bronze model on this notebook with more then 1100+ views.
Twitter Sentiment Analysis
This event was held by an Analytics vidya in which more than 15000+ candidates registered for the event and I was successfully able to secured the 24 positions in the event. This event was to predict the sentiments based on Tweets by the peoples.
Holiday Season
It was an image classification problem held at Hackerrank in which more than 4200+ candidates had registered and I used the SOTA InceptionRestnetV2 Pretrained model and successfully secured 21 positions in the event.
Recommendation System
Working on the recommendation System motivation - Tensor Girl Notebook and Abhishek Thakur
Snake Breed Detection
P.S - The government has been facing a long-standing issue of wild animals entering residential areas due to various reasons. It’s of critical importance that if any such dangerous animal is encountered, the concerned authority should be notified immediately. Reptiles, especially snakes, are among the most dangerous animals and they often enter residential areas.
Dog vs Cat Classifier
In this competition, you’ll write an algorithm to classify whether images contain either a dog or a cat.
RandomForest and Optuna
I used RandomForest Classifier to classify MNIST digits and used Optuna for hyperparameter optimization.
Sentiment Analysis Using Pyspark
I used the pyspark for spam detection and automated the Email detection.
XGBRegressorGPU
I used XGBRegressor and configured it with GPU for the regression task in Tabular Playground Series - Feb 2021.