CNN convolutional neural network
CNN has become the core concept for almost every image analysis algorithm whether it be a R-CNN or a YOLO. Convolution can efficiently extract info from image source and establish complex relationships between the entities. here I am trying to get best possible accuracy out of CNN without increasing the computation time. the accuracy depend on kernel size number of feature per layer proper balance between underfitting and overfitting.
Hakkımda
I am individual developer with passion for machine learning and AI. with a very strong background in Advanced Mathematics and Physics I can develop very efficient projects without flaws I specialize in almost every aspects of ML -> Natural Language Processing :- 1.) I specialize in developing smart conversational assistants(chatbots) using rasa stack, cakechat, deep-pavlov , alexa api. 2.) I could integrate these systems into voice service platforms and messaging platforms like. telegram, slack, alexa, google assistant etc. 3.) I also specialize in using NLP for text-mining ,entity extraction, semantic [login to view URL] NLTK,spacy,pytorch etc -> Computer Vision System with Deep learning:- 1.)I regularly make use of libraries such as pytorch, tensorflow , keras, opencv. and use them to create algorithms such as YOLO, RCNN etc. -> Machine learning:- I have good amount of experience in machine learning algorithms like clustering, deep-learning, regression, classification etc. -> Full stack Web-Development:- efficiency of AI and machine learning cannot be achieved without proper deployment. I have knowledge for deploying these models on various backends. eg apache2 , nginx , gunicorn, mod-wsgi, django,flask, rest-api etc. -> data-base development:- the knowledge of the AI needs a database system to store.I specialize in conventional RDBMS such as MySQL,postgresql, sql-server,[login to view URL] well as graph databases such as neo4j.