datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
ROCStories2016
ROCStories 2016 Dataset
This is the HuggingFace version of ROCStories Dataset.
We take a few changes.
Column names are renamed to match in each split, for example in test we have originally InputSentenceX, that was renamed do sentenceX to match the train split.
Columns missing in different splits were add to the other split with empty values. For example train split contains storytitle column. This column as added in test and validation splits.
There was three columns in… See the full description on the dataset page: https://huggingface.co/datasets/igormorgado/ROCStories2016.ROCStories2018
ROCStories 2018 Dataset
This is the HuggingFace version of ROCStories Dataset.
We take a few changes.
Column names are renamed to match in each split, for example in test we have originally InputSentenceX, that was renamed do sentenceX to match the train split.
Columns missing in different splits were add to the other split with empty values. For example train split contains storytitle column. This column as added in test and validation splits.
There was three columns in… See the full description on the dataset page: https://huggingface.co/datasets/igormorgado/ROCStories2018.RocketReviews
RocketReviews Dataset
A structured dataset scraped from RocketReviews.com for use in AI/ML pipelines and vector databases.
Collection Status
Legend: [ ] not started · [~] in progress · [x] complete
Primary Tables
Table
Description
Script
Output
Status
reviews
Kit and product reviews with ratings and text sections
scripts/reviews/01_scrape.py
source/reviews/
[~]
flights
Member flight logs with conditions and notes
scripts/flights/01_scrape.py… See the full description on the dataset page: https://huggingface.co/datasets/rocketsmith/RocketReviews.ROCOThis dataset is made using raw data from ROCO(Radiology Objects in COntext), a multimodal image dataset, with the aim of detecting the interplay between visual elements and semantic relations present in radiology images.
For each image in the original raw dataset we used the associated caption to generate a simulated conversation about the image between a user and a chatbot.