Close

About Me

Nuno Figueiredo Pires

Data Scientist | Business Transformation & Analytics

Data-driven professional transitioning into data science with 20+ years of experience in digital transformation, operations, and strategic decision-making. Proficient in Python, SQL, data science and machine learning, with a strong ability to translate business challenges into data-driven solutions. Passionate about leveraging analytics, automation, and Al to optimize business processes and drive innovation.

Categories

.

running output

📂 Organize Media AI

Smart media organizer using deep learning, metadata, and hashing for automated classification and cleanup.

This tool helps you structure a photo/video archive by recursively scanning folders, extracting date and event info, classifying images using ResNet-50, and identifying exact and perceptual duplicates (images and videos).

After consolidating all the media files, I ended with 780 GiB of data in more than 90_000 individual files. It will take ages to organize this data system by hand

🚀 Key Features

🔧 Technologies Used

Python, PyTorch, Transforms, TIMM, OpenCV, PIL, CNNs, Residual Networks, ImageHash, EXIFRead, PyHEIF, JSON

📂 Folder Output

/destination/
├── 2024-
│   └── 03-10/
│       └── Birthday/
│           ├── pict00001_dog.jpg
│           └── vid00001_party.mp4
├── duplicated/
├── near_duplicated/
└── noevaluate/

📄 Sample Output

Files renamed as:
pict00001_dog.jpg
vid00005_sunset.mp4