Datasetpublic

everestt/osrs-inferno-images

Description: This is a dataset of Old School RuneScape Inferno monster images taken from various gameplay recordings with varying game settings and client overlay plugin configurations. Dataset Info Total Examples: 3288 Raw Bit Depths: uint8 Processed Shapes: 128x128x3 (BGR) Average Raw File Size: 66.18 KB Raw File Size Distribution (Bins): 10KB: 494 10-50KB: 1058 50-100KB: 1018 100-500KB: 718 500KB: 0 Class Distribution: bat: 349 blob: 367 jad: 202 jad_healer: 224 mager:… See the full description on the dataset page: https://huggingface.co/datasets/everestt/osrs-inferno-images.

sourceHugging Facecc-by-sa-4.0updated 5mo agoView on Hugging Face

0likes6downloads

Dataset Card

Description: This is a dataset of Old School RuneScape Inferno monster images taken from various gameplay recordings with varying game settings and client overlay plugin configurations.

Dataset Info ============

—Total Examples: 3288
—Raw Bit Depths: uint8
—Processed Shapes: 128x128x3 (BGR)
—Average Raw File Size: 66.18 KB

Raw File Size Distribution (Bins):

—10KB: 494
—10-50KB: 1058
—50-100KB: 1018
—100-500KB: 718
—500KB: 0

Class Distribution:

—bat: 349
—blob: 367
—jad: 202
—jad_healer: 224
—mager: 452
—meleer: 451
—nibbler: 270
—ranger: 424
—zuk: 454
—zuk_healer: 95

CSV Column Information: ----------------------

—'class': The label/category of the image.
—'p0' to 'p49151': Flattened pixel data. Pixels are stored in BGR format (Blue, Green, Red). For a 128x128 image, there are 1281283 = 49,152 pixel values. Mapping: p(i) where i covers [row width channels + col * channels + channel].

How to use this data: --------------------

1.Load with pandas: df = pd.readcsv('pixeldata.csv')
2.Separate features and labels: X = df.drop('class', axis=1).values y = df['class'].values
3.Reshape pixels back to images (if needed): images = X.reshape(-1, 128, 128, 3)
4.Normalize: X_normalized = X / 255.0