sevdaasyn

Action-Recognition-3DCNNs

A video analyzer for kitchen activities cutting, baking, serving etc. Action features are extracted by using Inception-v1 I3D model trained on the Kinetics dataset training split

Context-Free-Grammers-CFGs

A random sentence generator implemented. For this, a sample CFG rule set is provided in the Chomsky Normal Form(CNF). And CYK parser as a recognizer which tells whether a given sentence is grammatically correct or not

Feed-Forward-Neural-Networks

Text Generation modeled using deep learning models such as Feed-Forward Neural Networks (FFNN) by using DyNet1 deep learning library and a poem dataset called Uni-Modal Poem,

Hidden-Markov-Model-Viterbi

NER(Named-entity recognition) tagger using HMMs. Viterbi algorithm to assign the most probable NER tag sequence for a given a word sequence.

Image-Classifiation-with-Convolutional-Neural-Networks

Transfer learning by using CNNs and how to train a network for a specific problem and pre-trained VGG-16 model usage

Image-Representation-and-Retrieval-Methods

Simple image representations and use kNN classification method to obtain categories of the query images, gabor filters to detect different orientations in the images, SIFT feature vectors extraction by using a built-in function and BoW descriptor extraction

Movie-Recommendation-KNN

Naive-Bayes-Fake-News-Detection

NGram-Language-Model

Plant-Disease-Detection

Plant disease detection using some machine learning techniques. Usefulness of deep convolutional neural networks VGG16 and ResNet50 is used.

sevdaasyn/Image-Representation-and-Retrieval-Methods

Simple image representations and use kNN classification method to obtain categories of the query images, gabor filters to detect different orientations in the images, SIFT feature vectors extraction by using a built-in function and BoW descriptor extraction

sevdaasyn/Plant-Disease-Detection

Plant disease detection using some machine learning techniques. Usefulness of deep convolutional neural networks VGG16 and ResNet50 is used.

sevdaasyn/Context-Free-Grammers-CFGs

A random sentence generator implemented. For this, a sample CFG rule set is provided in the Chomsky Normal Form(CNF). And CYK parser as a recognizer which tells whether a given sentence is grammatically correct or not

sevdaasyn/Hidden-Markov-Model-Viterbi

NER(Named-entity recognition) tagger using HMMs. Viterbi algorithm to assign the most probable NER tag sequence for a given a word sequence.

sevdaasyn/Action-Recognition-3DCNNs

A video analyzer for kitchen activities cutting, baking, serving etc. Action features are extracted by using Inception-v1 I3D model trained on the Kinetics dataset training split

sevdaasyn/Feed-Forward-Neural-Networks

Text Generation modeled using deep learning models such as Feed-Forward Neural Networks (FFNN) by using DyNet1 deep learning library and a poem dataset called Uni-Modal Poem,

sevdaasyn/Image-Classifiation-with-Convolutional-Neural-Networks

Transfer learning by using CNNs and how to train a network for a specific problem and pre-trained VGG-16 model usage

sevdaasyn/Movie-Recommendation-KNN

sevdaasyn/Naive-Bayes-Fake-News-Detection

sevdaasyn/NGram-Language-Model

sevdaasyn/Smart-Contract-Taxi

A smart contract that handles a common asset and distribution of income generated from this asset in certain time intervals. The common asset in this scenario is a taxi.