Nininkkka/Minecraft-vision-dataset

🌍 Minecraft Vision Dataset

Manually collected Minecraft screenshots for classification tasks. Useful for training AI agents, bots, and computer vision models.

📦 What's inside

• Biomes, dimensions, time of day, and more.
• See the repository file browser for the full list of classes and current image counts.

📂 Dataset structure

The root contains one folder per class. New classes are added over time.

🔍 Current statistics

Exact numbers of classes and images are always visible on the dataset page.
Browse the "Files and versions" tab – everything is there.

🔄 Continuous updates

This dataset is regularly expanded. No manual changelog is kept – check the commit history for updates.

---

🎮 Minecraft Vision Dataset - Recommended Groups

⚠️ Important: Don't mix groups in one training run. Pick one below.

---

🧩 Recommended Groups

1. 🌍 Core biomes (128x128, 7 classes):
   caves, end, forests, nether, oceans, plains, sky
   (Most popular & balanced)

2. 🌞 Time of day (128x128, 4 classes):
   time-day, time-noon, time-night, time-midnight

3. 📸 High resolution (384x384, 5 classes):
   caves_384x384, desert_384x384, nether_384x384, end_384x384, forests_384x384

---

🚀 Quick filter example (PyTorch)

from datasets import load_dataset
ds = load_dataset("Nininkkka/Minecraft-vision-dataset", split="train", streaming=True)

# Filter Group 1
def filter_group1(x):
    path = x["image"]["path"]
    return "/v1/" in path and any(b in path for b in ["caves/","end/","forests/","nether/","oceans/","plains/","sky/"])

group1 = ds.filter(filter_group1)