I have experience in implementing ML and DL models using tenserflow, keras and scikit-learn. I have completed many projects. Some of these are as:
1. Human Activity Recognition (HAR) :- I have tried three models Support Vector Machines (SVM), K-Nearest Neighbors (k-NN) and Convolutional Neural Network (CNN) to classify six activities(Walking ,Walking Upstairs, Walking Downstairs, Sitting, Standing, Laying). The most accurate model tried is SVM with accuracy of 98.47%.
2. Brain Computer Interface(BCI) with the P300 Speller which aims for helping patients incapable to activate muscles to spell words by methods for their brain signal activities. The first classification is to identify the presence of a P300 in the electroencephalogram(EEG). The second classification is the combination of various P300 response for deciding the correct character to spell.
3. Handwritten Digit Recognition using CNN :- Implemented a CNN model using tenserflow to identify handwritten digits from the MNIST dataset.
I will first discuss problem statement and your requirement properly and then I will propose an approach and estimated outcomes with you and if you will agree with that then I will code that and complete it. Your requirement and statisfaction will be my first priority.
Looking forward to hearing from you.
Thank You.