ML-AND-DL-

Repository Structure

• MachineLearning: Traditional machine learning algorithms.
  • K-Means: Clustering.
  • KNN: K-Nearest Neighbors.
  • NaiveBayes: Naive Bayes classifiers.
  • PrincipalComponentAnalysis: PCA for dimensionality reduction.
  • Regression: Linear, Logistic, Polynomial, and Ridge Regression.
• DeepLearning: Deep learning implementations.
  • Keras: Keras-based solutions.
  • Neural_Network_Implementation_using_NumPy: Custom neural network with NumPy.
  • Pytorch: PyTorch implementations (CNNs, GANs).
  • Tensorflow: TensorFlow implementations (MLPs, CNNs, RNNs, GANs).

Each subdirectory contains code, notebooks, and occasionally datasets or model weights.

Visualisation

Problem	Description	Implementation	Dataset
Picking best computer game to try	K-Means clustering for visualizing top positions	Python (raw)	Kaggle - Video Game Sales

AI Programming Showcase

This section highlights my work with AI algorithms and frameworks, focusing on understanding their mechanics rather than solving complex problems. Implementations use classical or generated datasets.

Raw Python

Machine Learning

Algorithm	Description	Implementation	Dataset
Linear Regression	Basic regression	Python (raw)	Generated Numbers
Ridge Regression	Compared with Linear Regression	Python (raw)	Generated Numbers
Polynomial Regression	Degree 2 approximation	Python (raw)	Generated Numbers
Polynomial Regression	Degree 3 approximation	Python (raw)	Generated Numbers
KNN	Manhattan, Euclidean similarity	Python (raw)	iris
PCA	Dimensionality reduction	Python (raw)	Generated Numbers
Naive Bayes	Gaussian distribution	Python (raw)	Pima Indian Diabetes
Logistic Regression	Binary classification	Python (raw)	Titanic Disaster Data

Deep Learning

Net Type	Problem	Description	Implementation	Dataset
MLP	Digit Classification	2-layers, mini-batch	Python (raw)	MNIST

scikit-learn

Algorithm	Description	Implementation	Dataset
Linear Regression	Basic regression	Python (sklearn)	Generated Numbers
Polynomial Regression	Degree 2 approximation	Python (sklearn)	Generated Numbers
Polynomial Regression	Degree 3 approximation	Python (sklearn)	Generated Numbers
KNN	Euclidean similarity	Python (sklearn)	iris

TensorFlow

Machine Learning

Algorithm	Description	Implementation	Dataset
Linear Regression	Basic regression	Python (Tensorflow)	Generated Numbers

Deep Learning

Net Type	Problem	Description	Implementation	Dataset
MLP	Digit Classification	2-layers, mini-batch, dropout	Python (Tensorflow)	MNIST
MLP	Encrypting data with Autoencoder	1-layer Encoder/Decoder, mini-batch	Python (Tensorflow)	MNIST
MLP	Digit Classification	Batch normalization, dropout	Python (Tensorflow)	MNIST
CNN	10 Classes Color Images	Dropout regularization	Python (Tensorflow)	CIFAR-10
CNN	10 Classes Color Images	Dropout regularization	Python (Tensorflow)	CIFAR-10
CNN	10 Classes Color Images	Batch normalization, dropout	Python (Tensorflow)	CIFAR-10
RNN	Simple Language Translator	Basic RNN	Python (Tensorflow)	French-English corpus
RNN	"The Simpsons" Script Generation	Character-level RNN	Python (Tensorflow)	"The Simpsons" script
DCGAN	Generating Human Face Miniatures	Deep Convolutional GAN

Keras

Net Type	Problem	Description	Implementation	Dataset
MLP	Digit Classification	2-layers, mini-batch, BN	Python (Keras)	MNIST
MLP	Clothes Images Classification	2-layers, mini-batch, BN	Python (Keras)	Fashion MNIST
MLP	Letters Classification	2-layers, mini-batch, BN	Python (Keras)	EMNIST
MLP	Review Sentiment Classification	Bag of Words	Python (Keras)	IMDB Reviews
MLP	Boston House Prices Regression	1-layer, mini-batch	Python (Keras)	Boston House Prices
CNN	Ten Color Image Classes	VGG15	Python (Keras)	CIFAR10
CNN	Letter Classification	32x32x64x64, 512, BN	Python (Keras)	EMNIST
CNN	Clothes Images Classification	16x16x32x32, 256x128, BN	Python (Keras)	Fashion MNIST
CNN	Digit Classification	16x32x64, 128, BN