[🇧🇷 Português] [🇺🇸 English]

1- 🧠 Machine Learning / Main Repository

Main repo for PUC-SP 5th semester Machine Learning course (2026): weekly classes, PyTorch / TensorFlow notebooks, CNN / RNN / GAN projects, seminars, and extensionist social initiatives by Prof. Roney Coelho.

Machine Learning Integrated Project - PUC-SP 5th Semester (2026)

Institution: Pontifical Catholic University of São Paulo (PUC-SP Humanistic AI & Data Science • 5º Semestre • 2026
School: Faculty of Interdisciplinary Studies
Course Repo: INTEGRATED PROJECT: MACHINE LEARNING - 72 Hours
Professor: ✨ Rooney Ribeiro Albuquerque Coelho
Extensionist Activities: Social projects with open-source software for community support.

Note

⚠️ Heads Up

• Projects and deliverables may be made publicly available whenever possible.

• The course emphasizes practical, hands-on experience with real datasets to simulate professional consulting scenarios in the fields of Machine Learning and Neural Networks for partner organizations and institutions affiliated with the university.

• All activities comply with the academic and ethical guidelines of PUC-SP.

• Any content not authorized for public disclosure will remain confidential and securely stored in private repositories.

What is Machine Learning?

Imagine teaching a robot puppy to fetch a ball. You show it many examples (data), it tries, makes mistakes, and gets better each time without you telling it every single step. That's Machine Learning (ML)! Computers learn patterns from data, like how kids learn by playing and trying.

Key topics here:

• Neural Networks: Like a brain made of tiny math "neurons" connected together. They guess answers and fix mistakes automatically.
• MLP (Multilayer Perceptron): A simple "brain" with layers – input (sees data), hidden (thinks), output (decides). Like stacking Lego blocks to solve puzzles.
• CNN (Convolutional Neural Networks): Super for pictures! Spots edges, shapes, then faces – like your eyes scanning a photo. Great for cat vs. dog pics.
• Frameworks: PyTorch (fast on Apple M-series chips, feels like Python magic) and TensorFlow (Google's tool, runs anywhere). They build these "brains" without supercomputers.

Inside.a.CNN.when.an.image.flows.through.the.network.mp4

Table of Contents

• Course Overview
• Core Learning Pillars
• Architecture Applications
• Weekly Classes - Sub-Repositories
• Folder Structure
• Related Project Repositories
• How to Use This Repo
• Grading & Assessment
• Learning Resources
• Tooling Stack
• Contributing Guidelines
• Bibliographic References
• Contact Me

Course Overview

Present basic and advanced concepts of artificial neural networks (ANNs), their biological inspiration, and real-world applications. Focus on building practical systems with supervised/unsupervised models like MLPs, CNNs, RNNs, GANs, and Reinforcement Learning. Includes extensionist projects for social good (open-source repos for communities).

Architecture Applications - Core Learning Pillars

Acronym	Full Name	Primary Application	Real-World Use
CNN	Convolutional Neural Network	Computer Vision	Image classification, object detection, facial recognition
MLP	Multilayer Perceptron	Classic Neural Networks	Tabular data prediction, regression, binary classification
RNN	Recurrent Neural Network	Sequential Data	Text generation, time series forecasting, speech recognition
GAN	Generative Adversarial Network	Data Generation	Image synthesis, deepfakes, data augmentation, art generation

Weekly Classes

All materials (slides, code, notebooks) go in folders like /week-1/, /week-2/, etc. Update as classes happen.

Week	Date	Topic Summary	Notes/Files Placeholder
1	19 Feb	Introduction to Machine Learning	/week-1/intro-ml.ipynb
2	26 Feb	What are neural networks? Intro to Perceptron. Basic shallow NN model.	/week-2/perceptron/
3	05 Mar	Supervised training: Loss functions, hyperparameter tuning.	/week-3/training/
4	12 Mar	Building shallow NNs with TensorFlow & PyTorch (dense layers).	/week-4/tf-pytorch/
5	19 Mar	Evaluating shallow NNs: Metrics. Loading data (CSV, HDF5, images).	/week-5/evaluation/
6	26 Mar	Data preprocessing (normalization, one-hot). Advanced shallow NNs.	/week-6/preprocessing/
7	02 Apr	Academic recess (Easter). No class.	-
8	09 Apr	Continued: Preprocessing & advanced NNs in PyTorch/TensorFlow.	/week-8/advanced-nns/
9	16 Apr	Seminar 1	/seminar-1/
10	23 Apr	CNNs: Intro, convolutions, pooling, fully connected layers.	/week-10/cnn-intro/
11	30 Apr	Training CNNs: Loss, optimizers, regularization (dropout, L1/L2), transfer learning.	/week-11/cnn-training/
12	07 May	CNN apps: Image classification, detection, segmentation. Data aug, visualization.	/week-12/cnn-apps/
13	14 May	RNNs intro: LSTM/GRU layers, info flow, training.	/week-13/rnns/
14	21 May	Encoder-Decoder: Machine translation, text generation. Training challenges.	/week-14/encoder-decoder/
15	28 May	GANs intro: Generator vs. Discriminator, training.	/week-15/gans/
16	04 Jun	National Holiday (Corpus Christi). No class.	-
17	11 Jun	Reinforcement Learning: Agents, Markov processes, Q-Learning, SARSA.	/week-17/rl/
18	18 Jun	Seminar 2	/seminar-2/

Folder Structure

/computer-vision/ ← CNN - Computer Vision
/classic-nn/ ← MLP - Classic Neural Networks
/sequential-data/ ← RNN - Sequential Data
/data-generation/ ← GAN - Data Generation
/projects/ ← Multi-architecture projects
/notebooks/ ← Demo implementations

Related Project Repositories

Project	Architecture	Status
Image Classifier	CNN	Coming Soon
Time Series	RNN	Coming Soon
Image Generator	GAN	Coming Soon

How to Use This Repo

1. Clone: git clone https://github.com/yourusername/PUC-SP-ML-Integrated-Project-2026.git.
2. Add weekly folders with README.md, .ipynb, .py files.
3. For PyTorch (local/Apple M): pip install torch. Fast on M-chips!
4. TensorFlow: pip install tensorflow.
5. Run notebooks in Colab or Jupyter. Share publicly for extensionist credit.

Grading & Assessment

Component	Weight	Type
Weekly Labs	20%	Individual
Projects	40%	Team
Presentations	20%	Team
Final Exam	20%	Individual

- Seminar 1 (16 Apr 2026): Individual, weight 0.5.
- Seminar 2 (18 Jun 2026): Individual, weight 0.5.

Methods: Dialogued lectures, TF / PyTorch projects, active methodologies, continuous evals.

Learning Resources

- PyTorch Tutorials
- Fast.ai Practical Deep Learning
- Papers With Code

Tooling Stack

pip install torch torchvision tensorflow pandas numpy matplotlib wandb

Contributing Guidelines

1. Fork → Clone → Branch (feat/cnn-week3)
2. Add notebooks to architecture folders
3. Update weekly schedule table
4. Submit PR with results

Bibliographic References

* Core Literature:

- Goodfellow et al. Deep Learning (2016) - All architectures
- LeCun et al. LeNet CNN (1998) - Computer Vision
- Hochreiter LSTM (1997) - Sequential Data
- Goodfellow GAN (2014) - Data Generation

* Basic

- NETTO, A.; MACIEL, F. Python for data science and machine learning made simple. Alta Books, 2021. [ppl-ai-file-upload.s3.amazonaws]
- SILVA, F. M. da et al. Artificial Intelligence: Applications in various human activities. Sagah, 2019. [ppl-ai-file-upload.s3.amazonaws]
- WITTEN, I. H. et al. Artificial Intelligence: A machine learning approach. LTC, 2021. [ppl-ai-file-upload.s3.amazonaws]

* Complementary

- BIFET, A. et al. Machine learning for data streams. MIT Press, 2018.
- CANO, A. Social media and machine learning. IntechOpen, 2020.
- HUTTER, F.; KOTTHOFF, L.; VANSCHOREN, J. Automated machine learning: methods, systems, challenges. Springer Nature, 2019.
- SUD, K. et al. Introduction to data science and machine learning. IntechOpen, 2020.
- THOMAS, C. Data mining. IntechOpen, 2018.

💌 Let the data flow... Ping Me !

🛸๋ My Contacts Hub

────────────── ⊹🔭๋ ──────────────

➣➢➤ Back to Top

Copyright 2026 Mindful-AI-Assistants. Code released under the MIT license.