This repository is in constant evolution to show and explain the technical side of AI and its applications to solve problems

Important

All models and cases on these repositories would be adapted to be used with the less possible requirement and computational power, to establish a minimum base to use them

Machine Learning Models

Note

It was used ScikitLearn to do Supervised and unsupervised Learning

• Classification.ipynb
• Clustering Hierarchical & agglomerative.ipynb
• Clustering K-Means.ipynb
• Multi-Class Classification KNN & SVM.ipynb
• Multiclass Classification Decision-Tree.ipynb
• Polynomial_Regression.ipynb
• Simple Linear_Regression.ipynb

Deep Learning techniques

Note

It is recomended to use CUDA or a graphic card, for ensuring deployment of pipelines and models It was used PyTorch, TensorFlow, Keras and Cyton to understand techniques and bases to made deep learning

Builded samples and Structure

• 1 Recap machine learning fundamentals Ensemble learning I: Bagging Random Forest

• 2 Ensemble learning II: Boosting XGBoost

• 3 Neural network I: Basics

• 4 Neural network II: CNN Convolutional Neural Network

• 5 Neural network III: RNN Recurrent Neural Network

• 6 Natural language processing I Tokenization, Bag-of-words, Word embeddings

• 7 Natural language processing II Naïve Bayes Classifier, Hidden Markov Models

• 8 Reinforcement learning Q-Learning

  Convolutional neural networks (CNNs):

  These are commonly used for image classification and object detection tasks. They are designed to recognize patterns in images by using filters that can detect edges, corners, and other features

  Recurrent neural networks (RNNs):

  These are used for tasks that involve sequential data, such as speech recognition and natural language processing. They are designed to process data that has a temporal component, such as audio signals or text

  Generative adversarial networks (GANs):

  These are used for generating new data that is similar to a given dataset. They consist of two neural networks that compete with each other to generate realistic data

NLP and some foundational models

Note

Under construction, It was used PyTorch, TensorFlow, Keras and Cyton to build and work models and its transformers

• Topic modeling with LSA, LDA and Word2Vec
• BERT (Bidirectional Encoder Representation from Transformers)

For future building samples

• GPT (Generative pre-trained transformer)
• ResNet (A deep neural network architecture for image classification)
• YOLOv5 (A real-time object detection system)
• CLIP (model developed by OpenAI that can understand the relationship between images and text)