# CS109B HW3 Contest 2023 — Data

来源：https://www.kaggle.com/competitions/cs109b-hw3-contest-2023/data

采集时间：2026-10-22T10:21:53Z

For this problem, we will be working with a modified version of the [Kannada MNIST dataset](assets/f568c65734a07392ae09122b.pdf), which is a large database of handwritten digits in the indigenous language *Kannada*.

This dataset consists of 60,000 28x28 grayscale images of the ten digits, along with a test set of 10,000 images.

For this homework, we will simplify the problem by only using the digits labeled `0` and `1` owing to the similarity of the two symbols, and we will use a total of 1,200 samples for training (this includes the data you will use for validation).

To understand the dataset better, we recommend this [article](https://towardsdatascience.com/a-new-handwritten-digits-dataset-in-ml-town-kannada-mnist-69df0f2d1456) by Vinay Prabhu, the curator of the dataset.

DATA FILES:

Here's a brief description of the data files:
- `kmnist_train.csv` is our training dataset and the last column contains our response class. The 784 other columns correspond to the pixel values of the 28x28 dimension image. Class 0 means a sample is the handwritten digit `0` and class `1` means a sample is the handwritten digit `1` in the Kannada written language. kmnist_train.csv has 1,200 samples.
- `kmnist_test.csv` has a structure similar to `kmnist_train.csv`, however the class label column is NOT included in with the test set. `kmnist_test.csv` has 2,000 samples.