iix/Parquet_FIles
Parquet_Files Cross Language (CL) Datasets Four datasets of language pair translations originating from CORDIS Project News (https://elrc-share.eu/) Structured as follows: | Field | Description | | --------------- | ----------------------------------------------------------------------- | | de/es/fr/it | Non-English transcripts of sentences | | en… See the full description on the dataset page: https://huggingface.co/datasets/iix/Parquet_FIles.
Parquet_Files
Cross Language (CL) Datasets
Four datasets of language pair translations originating from CORDIS Project News (https://elrc-share.eu/)
Structured as follows:
| Field | Description |
| --------------- | ----------------------------------------------------------------------- |
| de/es/fr/it | Non-English transcripts of sentences |
| en | English translations of sentences |
120k rows (approx. per parquet file)Characters.parquet
One large dataset of synthetic characters, information regarding them and their outputs. Created by combining other synthetic character datasets which can be found at: https://shorturl.at/gjIJ5, https://shorturl.at/DFQT6 & https://shorturl.at/tHX68.
Structured as follows:
| Field | Description |
| -------------- | --------------------------------------------------------------- |
| name | Character name |
| categories | Categories or genres associated with the character (array) |
| personalities | Personality traits and characteristics of the character (array) |
| description | Description of the character |
| conversation | Conversational interactions involving the character (array) |
17.6k rows Flores7Lang.parquet
Seven languages merged from the Flores-200 Dataset (https://github.com/facebookresearch/flores/blob/main/flores200/README.md)
Structured as follows:
| Field | Description |
| --------------- | ----------------------------------------------------- |
| deu | Sentence translated into German |
| eng | Sentence translated into English |
| epo | Sentence translated into Esperanto |
| fra | Sentence translated into French |
| ita | Sentence translated into Italian |
| spa | Sentence translated into Spanish |
| tur | Sentence translated into Turkish |
2.1k rowsFraud_detection.parquet
A synthetic financial dataset intended for Fraud detection tasks (https://www.kaggle.com/datasets/ealaxi/paysim1).
Structured as follows:
| Field | Description |
| --------------- | ------------------------------------------------------------- |
| Step | Maps a unit of time in the real world (1 step = 1 hour) |
| Type | CASH-IN, CASH-OUT, DEBIT, PAYMENT, or TRANSFER |
| Amount | Amount of the transaction in local currency |
| nameOrig | Customer who initiated the transaction |
| oldbalanceOrg | Initial balance before the transaction |
| newbalanceOrig | Customer's balance after the transaction |
| nameDest | Recipient ID of the transaction |
| oldbalanceDest | Initial recipient balance before the transaction |
| newbalanceDest | Recipient's balance after the transaction |
| isFraud | Identifies a fraudulent transaction (1) or non-fraudulent (0) |
2.27m rows