Datasets:

Krishna5T
/

trial1

configs:
  - config_name: passage
    data_files:
      - split: train
        path: Passage/*.csv
  - config_name: mcq_4way
    data_files:
      - split: train
        path: mcq(4-way)/*.csv
  - config_name: constructed_response
    data_files:
      - split: train
        path: Constructed-response/*.csv
  - config_name: general
    data_files:
      - split: train
        path: '*.csv'
task_categories:
  - question-answering
  - image-to-text

Trial1 Dataset

This dataset contains four configurations for different types of educational data, including images.

Configurations

Passage: Data related to long-form reading passages and figures.
MCQ (4-way): Multiple choice questions with 4 options.
Constructed Response: Open-ended questions requiring written answers.
General: Root level data files.

How to use

from datasets import load_dataset

# Load a specific config
dataset = load_dataset("Krishna5T/trial1", "mcq_4way")