Hugging Face's logo

Datasets:
tanuj437
/
sanskrit-pos-tagged-corpus

Languages:
Sanskrit
Tags:
sanskrit
pos-tagging
token-classification
nlp
License:
Dataset card Files
xet
 Community
Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    ValueError
Message:      Bad split: train_rich. Available splits: ['train', 'validation', 'test']
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                         ^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/utils.py", line 61, in get_rows
                  ds = load_dataset(
                       ^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/load.py", line 1505, in load_dataset
                  return builder_instance.as_streaming_dataset(split=split)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1196, in as_streaming_dataset
                  raise ValueError(f"Bad split: {split}. Available splits: {list(splits_generators)}")
              ValueError: Bad split: train_rich. Available splits: ['train', 'validation', 'test']

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Sanskrit POS Tagging Dataset

This dataset is designed for fine-tuning the SanskritBERT model on Part-of-Speech (POS) tagging tasks. It contains sentences tokenized and annotated with POS tags.

The dataset provides two variants of annotations:

  1. Rich: Contains detailed morphological information or fine-grained POS tags, suitable for tasks requiring deep linguistic analysis.
  2. Simple: Contains simplified POS tags (likely Universal POS tags), suitable for general-purpose tagging and lighter models.

Dataset Structure

The data is provided in JSONL (JSON Lines) format. Each line represents a single sentence entry.

Data Fields

  • id: A unique identifier for the sentence.
  • tokens: A list of strings, where each string is a word token in the sentence.
  • labels: A list of strings corresponding to the POS tags for each token.

File Organization

The dataset is split into rich and simple subsets, each with train, test, and valid splits:

Rich Tags (Detailed)

  • pos_rich_train.jsonl
  • pos_rich_test.jsonl
  • pos_rich_valid.jsonl

Simple Tags (Simplified)

  • pos_simple_train.jsonl
  • pos_simple_test.jsonl
  • pos_simple_valid.jsonl

Usage

You can load this dataset using the Hugging Face datasets library:

from datasets import load_dataset

# Load the simple dataset
dataset_simple = load_dataset("json", data_files={
    "train": "pos_simple_train.jsonl",
    "test": "pos_simple_test.jsonl",
    "validation": "pos_simple_valid.jsonl"
})

# Load the rich dataset
dataset_rich = load_dataset("json", data_files={
    "train": "pos_rich_train.jsonl",
    "test": "pos_rich_test.jsonl",
    "validation": "pos_rich_valid.jsonl"
})

Citation

If you use this dataset in your research, please cite the SanskritBERT paper or the source of this data.
Downloads last month
58
Total file size:
618 MB

Models trained or fine-tuned on tanuj437/sanskrit-pos-tagged-corpus