Pinned Repositories
Video_to_video_transformation_using_ComfyUI
Burning-Liquid-Classification-from-flame-using-Transfer-Learning
Cats-vs.-Dogs-classifier-using-image-augmentation_and_CNN-
Created a cats vs. dogs classifier using image augmentation and CNN as a part of Kaggle Challenge exercise
Chatbot_using_Langflow
CIFAR10-Image-Classification-with-Multilayer-Perceptron
CNN_to_classify_happy_sad_emoji_faces
A convolutional neural network that trains to 99.9% accuracy on happy/sad emoji images, and cancels training upon hitting this training accuracy threshold.
Comparative-Analysis-of-Fine-Tuning-CNNs-for-Brain-Tumor-Detection
Financial-agent
Financial agent using AutoGPT
Housing_price_predictor
A house has a base cost of 50k, and every additional bedroom adds a cost of 50k. This will make a 1 bedroom house cost 100k, a 2 bedroom house cost 150k etc. A neural network is developed that learns this relationship so that it would predict a 7 bedroom house as costing close to 400k etc.
KrantiAdsul's Repositories
KrantiAdsul/Financial-agent
Financial agent using AutoGPT
KrantiAdsul/Language_Assistant_AutoGPT
Language Assistant application using AutoGPT
KrantiAdsul/Video_to_video_transformation_using_ComfyUI
KrantiAdsul/Chatbot_using_Langflow
KrantiAdsul/Movie_pitch_site_final
KrantiAdsul/League-Tracker-Node.js-Web-Application-Development-and-Enhancement
KrantiAdsul/Leveraging-Preference-Data-to-Improve-Online-Learning
Explored impact of offline demo data on online learning, developed Warm Thompson sampling algorithm with offline dataset integration, and compared its performance to normal Thompson sampling in Linear Multi-Armed Bandit
KrantiAdsul/KrantiAdsul.github.io
KrantiAdsul/Text-Extraction-from-Natural-Scenes-with-PyTorch-R-CNN
Developed machine learning model focused on fine-tuning a PyTorch R-CNN model for text extraction from natural scenes, leveraging the Text MS-COCO dataset. Gained hands-on experience in model adaptation, data preprocessing, and in-depth loss curve analysis, showcasing strong skills in computer vision and deep learning techniques
KrantiAdsul/MNIST-Image-Classification-with-Custom-MLP
KrantiAdsul/CIFAR10-Image-Classification-with-Multilayer-Perceptron
KrantiAdsul/Burning-Liquid-Classification-from-flame-using-Transfer-Learning
KrantiAdsul/ML_EE599
Homework repo for EE599: Systems for Machine Learning
KrantiAdsul/Comparative-Analysis-of-Fine-Tuning-CNNs-for-Brain-Tumor-Detection
KrantiAdsul/Multi-class_classifier_using_sign_language_dataset_kaggle
Using Sign Language dataset from https://www.kaggle.com/datamunge/sign-language-mnist, and to build a multi-class classifier to recognize sign language
KrantiAdsul/Cats-vs.-Dogs-classifier-using-image-augmentation_and_CNN-
Created a cats vs. dogs classifier using image augmentation and CNN as a part of Kaggle Challenge exercise
KrantiAdsul/Predict_next_word_in_sequence
Train a dataset of Irish songs to create traditional-sounding poetry. Here a corpus of Shakespeare sonnets is used to train a model. Then, checking if the model is able to create poetry!
KrantiAdsul/Neural_network_to_map_BBC_archive_news_words_to_categories
BBC news reports dataset (kaggle dataset) contains articles that are classified into a number of different categories. Here neural network is designed that can be trained on this dataset to accurately determine what words determine what category the news falls under
KrantiAdsul/Tokenize_BBC_archive_kaggle_dataset
Tokenized given dataset, removing common stopwords. Dataset used here is a variation of the [BBC News Classification Dataset](https://www.kaggle.com/c/learn-ai-bbc/overview), which contains 2225 examples of news articles with their respective categories (labels)
KrantiAdsul/Transfer_learning_to_classify_horses_vs_humans
Using Transfer Learning to classify images of Horses vs Humans. Validation set accuracy should be around 95% and training should automatically stop once it reaches this desired accuracy.
KrantiAdsul/CNN_to_classify_happy_sad_emoji_faces
A convolutional neural network that trains to 99.9% accuracy on happy/sad emoji images, and cancels training upon hitting this training accuracy threshold.
KrantiAdsul/Improve_MNIST_using_CNN
Improving MNIST to 99.5% accuracy or more by adding only a single convolutional layer and a single MaxPooling 2D layer to the model
KrantiAdsul/MNIST_classifier_with_callbacks_in_Tensorflow
MNIST classifier created that trains to 99% accuracy and stops once this threshold is achieved
KrantiAdsul/Housing_price_predictor
A house has a base cost of 50k, and every additional bedroom adds a cost of 50k. This will make a 1 bedroom house cost 100k, a 2 bedroom house cost 150k etc. A neural network is developed that learns this relationship so that it would predict a 7 bedroom house as costing close to 400k etc.