datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ChartQAIf you wanna use the dataset, you need to download the zip file manually from the "Files and versions" tab.
Please note that this dataset can not be directly loaded with the load_dataset function from the datasets library.
If you want a version of the dataset that can be loaded with the load_dataset function, you can use this one: https://huggingface.co/datasets/ahmed-masry/chartqa_without_images
But it doesn't contain the chart images. Hence, you will still need to use the images stored in… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/ChartQA.ChartQAPro
ChartQAPro: A More Diverse and Challenging Benchmark for Chart Question Answering
🤗Dataset | 🖥️Code | 📄Paper
The abstract of the paper states that:
Charts are ubiquitous, as people often use them to analyze data, answer questions, and discover critical insights. However, performing complex analytical tasks with charts requires significant perceptual and cognitive effort. Chart Question Answering (CQA) systems automate this process by enabling models to interpret and reason with… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/ChartQAPro.chartqa_without_images
Dataset Card for "chartqa_without_images"
If you wanna load the dataset, you can run the following code:
from datasets import load_dataset
data = load_dataset('ahmed-masry/chartqa_without_images')
The dataset has the following structure:
DatasetDict({
train: Dataset({
features: ['imgname', 'query', 'label', 'type'],
num_rows: 28299
})
val: Dataset({
features: ['imgname', 'query', 'label', 'type'],
num_rows: 1920
})
test:… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/chartqa_without_images.ChartGemma
Dataset Card for the ChartGemma dataset.
If you wanna load the dataset, you can run the following code:
from datasets import load_dataset
dataset = load_dataset('ahmed-masry/ChartGemma')
The dataset has the following structure:
DatasetDict({
train: Dataset({
features: ['image', 'input', 'output'],
num_rows: 163240
})
})
Each row consist of the image, the input query, and the output label.
To load the image, we suggest using the following piece of code:… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/ChartGemma.unichart-qa-dataunichart-table-dataUniChart-pretrain-imagesunichart-pretrain-data
Dataset Card for "unichart-pretrain-data"
If you wanna load the dataset, you can run the following code:
from datasets import load_dataset
data = load_dataset('ahmed-masry/unichart-pretrain-data')
The dataset has the following structure:
DatasetDict({
train: Dataset({
features: ['imgname', 'query', 'label'],
num_rows: 6898333
})
})
It has 6898333 rows; each row consist of the imgename, the input query, and the output label.
Chart Images
The… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/unichart-pretrain-data.DashboardQA
DashboardQA: Benchmarking Multimodal Agents for Question Answering on Interactive Dashboards
🤗Dataset | 🖥️Code | 📄Paper
The abstract of the paper states that:
Dashboards are powerful visualization tools for data-driven decision-making, integrating multiple interactive views that allow users to explore, filter, and navigate data. Unlike static charts, dashboards support rich interactivity, which is essential for uncovering insights in real-world analytical workflows. However… See the full description on the dataset page: https://huggingface.co/datasets/ahmed-masry/DashboardQA.Om-El-Masryeen-project