Datasetpublic

Yeongkwon/MNIST-M-flatten

MNIST-M (tabular) A flattened version of Mike0307/MNIST-M for tabular learning (XGBoost, AutoGluon, etc.). Flatten procedure Each row's image (PNG bytes) is decoded with PIL and converted to RGB. The 32x32x3 pixels are flattened in row-major (H, W, C) order into 3072 features. Columns: px_0 ... px_3071 (uint8, 0-255) plus label (int). The original train/test split is preserved: train.csv, test.csv. Restoring an image import pandas as pd, numpy as… See the full description on the dataset page: https://huggingface.co/datasets/Yeongkwon/MNIST-M-flatten.

sourceHugging Faceupdated 3mo agoView on Hugging Face

0likes25downloads

Dataset Card

MNIST-M (tabular)

A flattened version of Mike0307/MNIST-M for tabular learning (XGBoost, AutoGluon, etc.).

Flatten procedure

—Each row's image (PNG bytes) is decoded with PIL and converted to RGB.
—The 32x32x3 pixels are flattened in row-major (H, W, C) order into 3072 features.
—Columns: px_0 ... px_3071 (uint8, 0-255) plus label (int).
—The original train/test split is preserved: train.csv, test.csv.

Restoring an image

python

import pandas as pd, numpy as np
from PIL import Image

row = pd.read_csv("train.csv", nrows=1).iloc[0]
arr = row.drop("label").to_numpy(np.uint8).reshape(32, 32, 3)
Image.fromarray(arr)