me

About Me

Enthusiastic about the field of data science, particularly in constructing robust models and performing exploratory data analysis (EDA) on datasets. Enjoy leveraging analytical skills to uncover meaningful insights and drive data-informed decision making. I am a quick and enthusiastic learner.

Projects

Machine Learning Diabetic Model

Developed a diabetic model with LOGISTIC REGRESSION, achieving 85% accuracy in determining diabetes presence. Implemented FlASK and HTML for the Frontend, and successfully deployed the model on HEROKU.

Deep Neural Network Diabetic Model

Devised high-accuracy DEEP NEURAL NETWORK MODELS models for Diabetes datasets by applying Hyperparameter tuning with TENSORFLOW. Achieved accuracy rates exceeding 90%.

Machine Learning HandWritten Scripts

Utilized Perceptron models on the MNIST dataset and implemented classification tasks with 95% accuracy.

Convolutional Neural Networks LENET - 5

Built the LeNet-5 architecture using TensorFlow. Through meticulous design and implementation, achieved an exceptional accuracy rate of 99.80%. Demonstrating my expertise in neural network architecture design and my ability to harness the power of TensorFlow to deliver high-performing models

Machine Learning Titanic Model

Built Titanic survival model using DECISION TREE & LOGISTIC. Utilized Exploratory Data Analysis for key insights. Precise data processing enhanced accuracy. High accuracy rates achieved through informed feature engineering.

Deep Neural Network HandWritten Sripts

Improved accuracy of MNIST dataset with DNNs to exceed 90% accuracy, and achieved remarkable results of over 99% accuracy using CNNs. Demonstrates proficiency in advancing models and achieving exceptional performance.

Technical Skills

Python

NumPy , Pandas, Scikit-learn, Matplotlib, Seaborn.

MySQL

Joins, Normalization, SubQueries, DML DCL & DDL commands.

Machine Learning

Scikit-learn, RandomForest, Decision Tree, Linear, Logistic, KNN, SVM.

Deep Learning

TensorFlow, Keras, keras Tuning, Tensor Board, Hyperparameter Tuning.

Framework & TOOLS

Flask, Streamlit, Langchain(CHATGPT API), Jupyter, Tableau.

Convolutional Neural Network

Conv2d, Pooling, Transfer Learning, Fine Tuning, LeNet-5

Latest Work

Anime Recommendation

Created Anime Recommendation Model utilizing Langchain framework & Chat GPT API. Empowers users to receive personalized anime suggestions based on provided genres. Developed interactive frontend using Python's Streamlit framework.

End-to-End MLOps

Created a dog and cat classification model using TensorFlow, implementing MLOps techniques for streamlined deployment on AWS EC2. Achieved robustness and scalability through data curation, CNN architecture, and CI/CD automation..

HR Data Analysis Dashboard

Constructed HR analytics strategies resulting in a 16% attrition rate reduction. Identified 56% of attrition in the R&D department and peak attrition within the 25-34 age group. Utilized Python, Excel, SQL, and Tableau for analysis and visualization.

GET IN TOUCH

    You can send me message and I will get back to you as soon as possible.
    You can also mail me on akash.vishwakarma.5477@gmail.com

Your message has been sent. Thank you!

© Built by Akash Vishwakarma. All rights reserved.