NN_ML_AI_bases_for_devs

Neural Networks, Machine Learning and AI bases for devs

The most basic and fundamental concepts to understand Neural Networks, Machine Learning and AI

# Install Python using Miniconda (recommended)
# https://docs.anaconda.com/miniconda/
# Quick command line install for Windows
# https://docs.anaconda.com/miniconda/#quick-command-line-install
curl https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe -o miniconda.exe
start /wait "" miniconda.exe /S
del miniconda.exe

# Create a new conda environment and install the required packages
conda create -n nn_ml_ai python=3.10
conda activate nn_ml_ai
pip install -r requirements.txt

Execution

• To execute examples for each task, run the xx-main.py file related in the scr folder.
• Download training files from ../data/ (i.e.: Binary_Train.npz, Binary_Dev.npz, MNIST.npz)
• For tasks 27 and 28, download trained models from: https://s3.amazonaws.com/intranet-projects-files/holbertonschool-ml/27-saved.pkl and https://s3.amazonaws.com/intranet-projects-files/holbertonschool-ml/28-saved.pkl

datasets

Binary_Train.npz https://s3.eu-west-3.amazonaws.com/hbtn.intranet.project.files/holbertonschool-ml/

Binary_Dev.npz https://s3.eu-west-3.amazonaws.com/hbtn.intranet.project.files/holbertonschool-ml/Binary_Dev.npz

MNIST.npz https://s3.eu-west-3.amazonaws.com/hbtn.intranet.project.files/holbertonschool-ml/MNIST.npz

Tasks

Based on the Holberton School Machine Learning curriculum, in this repository you will find the following tasks:

1. Perceptron - One neuron Neural Network (NN)
2. One layer NN - binary classification
3. Multi layer NN (deep neural network) - binary classificator

24-One-Hot Encode 25-One-Hot Decode 27. Update DeepNeuralNetwork 28. All the Activations (multiclass classification)

29. Activation functions: https://www.linkedin.com/pulse/activation-functions-neural-networks-leonardo-calderon-j-

Built With

• Python 3.10.14
• NumPy 2.1.0
• Matplotlib 3.9.2

LeoCJJ

Leonardo Calderon J.

2024