Datasets:
Tasks:
Image Classification
Modalities:
Image
Formats:
imagefolder
Languages:
English
Size:
< 1K
License:
| language: | |
| - en | |
| license: mit | |
| size_categories: | |
| - 10k<n<100k | |
| task_categories: | |
| - image-classification | |
| tags: | |
| - mnist | |
| - mytorch | |
| - rjit | |
| # π MyTorch Refined MNIST Dataset | |
| Curated by **Aryan Singh Chandel (Shiro)** at **Rustamji Institute of Technology (RJIT)**. | |
| This dataset contains the refined version of the MNIST handwritten digit database, specifically pre-processed for compatibility with the **MyTorch** deep learning framework. | |
| ## ποΈ Dataset Structure | |
| The data is stored in a compressed NumPy format (`mnist_raw.npz`) containing: | |
| - **X_train / y_train:** 60,000 samples for training. | |
| - **X_test / y_test:** 10,000 samples for final validation. | |
| ## β‘ Pre-processing Specification | |
| Every image in this dataset has undergone the following transformation logic used in the MyTorch 98.59% accuracy run: | |
| 1. **Flattening:** $28 \times 28$ spatial grids converted to $\mathbb{R}^{784}$ feature vectors. | |
| 2. **Global Scaling:** Pixel values $P \in [0, 255]$ normalized to $P' \in [0, 1]$ via $P' = P / 255.0$. | |
| 3. **Refinement:** Standardized tensor shapes for high-speed matrix multiplication in NumPy-based linear layers. | |
| ## π Usage | |
| To load this dataset into your own MyTorch project: | |
| ```python | |
| import numpy as np | |
| data = np.load('mnist_raw.npz') | |
| X_train, y_train = data['X_train'], data['y_train'] | |
| ``` | |
| π Citation | |
| If you use this refined dataset in your research, please attribute: | |
| Chandel, A. S. (2026). MyTorch: Deep Learning from Scratch at RJIT. | |