aiswarya-nandakumar Goto Github PK

aiswarya-nandakumar

Config files for my GitHub profile.

binary-splitting-algorithm-from-scratch

Binary splitting algorithm is used in Kmeans clustering. Binary splitting algorithm from scratch is implemented in Python.

classification-of-eeg-signals

Deep learning for classification of EEG signals. EEG signals are classified using convolutional neural networks. 1D CNN is implemented using Keras and Tensorflow. this is a binary classification project. EEG signals are classified into normal and abnormal signals based on their values. This classification can be used for prediction of various kinds of brain disorders.

classification-of-eeg-signals-using-support-vector-machines

Classification of EG signals into normal and abnormal using Support Vector Machines (SVM).

classification-of-rice-grains-using-kmeans-clustering

There are different rice grains being cultivated. these rice grains have different sizes and shapes. In this project, different rice grains are clustered using Kmeans clustering based on their sizes.

comparing-the-performance-of-different-ml-algorithms-on-uci-har-data

Human Activity Recognition database built from the recordings of 30 subjects performing activities of daily living (ADL) while carrying a waist-mounted smartphone with embedded inertial sensors. Different Machine Learning algorithms, Support Vector Machines, K Nearest Neighbors, Decision Trees and Random Forest are applied to the dataset for classification based on the HAR recorded signals. The performance of different models are evaluated by using standard metrics and confusion metrics and compared.

converting-an-rgb-image-into-binary-image

converting an RGB image into a binary image

deep-learning-for-classification-of-eeg-signals

Classification of EEG signals using Deep learning. This is a binary classification problem. The EEG signals are classified into 2 target classes, normal and abnormal signals. Keras and tensorflow are used. this classification can be used for the prediction of various brain disorders.

deep-learning-for-detction-of-alzheimer-s-from-mri-images

deep-learning-for-detection-of-alzheimer-s-from-longitudinal-mri-data

Deep learning approach for detection of Alzheimer's from longitudinal MRI data (OASIS) dataset. Concept of transfer learning has been applied here. Different Networks are used and results are compared.

deep-learning-for-digit-recognition-using-mnist-dataset

Deep learning for digit recognition using MNIST dataset. Single digits are classified and predicted here. This can be extended to alphabets and multiple digits also.

eeg-signal-classification-using-artificial-neural-networks

Classification of EEG signals into normal and abnormal target classes using Artificial Neural Networks(ANN). This is a binary classification problem. Based on this classification different brain disorders can be predicted. Keras and tensorflow are used. Python code is attached.

explainable-ai-for-medical-image-classification

Using LIME for understanding how the features are learned by the ML model for classifying the medical images.

feed-forward-computation-from-scratch

Implementation of feed forward computation from scratch

github-slideshow

A robot powered training repository :robot:

heat-equation-using-python

In this a steady state heat conduction equation is considered which governs the steady state temperature distribution along a plate of dimension 10X10. The heat distribution is such that the top boundary of the plate is kept at 900K, while the bottom is kept at 300K. The left and right ends of the plate are kept at 200K and 600K respectively. In Python the temperature distribution is implemented using Central Difference Scheme.

heat-equation-with-temperature-source

Here a steady state @ dimensional heat generation equation is considered which governs the steady state temperature distribution along a plate of dimension 10X10. The temperature distribution is such that the top boundary of the plate is kept at 900K, while the bottom is kept at 300K. The left and right ends of the plate are kept at 200 and 600K respectively. In Python the temperature distribution is implemented. The heat source is maintained at 1000K temperature an it is kept at the middle of the plate such that the (x, y) co-ordinates are (5,5).

kmeans-clustering-for-classifying-customers-of-a-mall-based-on-annual-income-and-spending-score

Kmeans is a useful algorithm in machine learning. Its used for clustering the data. In this project, I have clustered the customers of a mall based on their annual income and spending score using Kmeans clustering.

laplace-equation

2D Laplace equation/ Diffusion equation in Python. 2 Functions are defined one that plots data in a 3D projection and the second one that iterates to solve for p until the change in the L1norm of p is less than a specified value.

plotting-images-as-subplots-in-python

Beginners often find difficulty in plotting images as subplots in Python. Here is a simple code for plotting various images as sub images in Python.

plotting-velocity-of-a-freely-falling-ball

Plotting the velocity of a freely falling ball by using computer vision and Python.

prediction-of-breast-cancer-from-histopathological-images

Different networks, 2D CNN, 1X1 filter, VGG16 and Inception network is used for predicting breast cancer from histopathological images. Performance of different models are compared and found out that InceptionNet performed the best.

projects

These are the codes of my various projects. They are created as separate repositories, where you can read about their descriptions and get the codes

random-forest-for-binary-classification-of-eeg-signals

Random Forest for binary classification of EEG signals. The data set consists of 14000 EEG values. They are classified into 2 classes using Random Forest algorithm based on normal and abnormal EEG values.

real-time-security-alarm

Real time security alarm using ESP NODE MCU and sensors in Arduino

rock-paper-scissor-game

Implementation of the popular game Rock Paper Scissors in Python.

seaborn-plots

Data visualization is very much useful and fun! For every projects, we will have to visualize the data. Practicing the different visualization techniques regularly can make the reports and presentations creative and easy. These are some data visualizations done using Seaborn in Python

support-vector-machine-for-breast-cancer-detection

SVM/ Support Vector Machines for Breast caner detection

temporal-graph-networks-for-user-item-interaction-prediction

TGN for user-item interaction prediction in an ecommerce platform.

test