TakeLab Podium

What is Podium?

Podium is a framework agnostic Python natural language processing library which standardizes data loading and preprocessing. Our goal is to accelerate users' development of NLP models whichever aspect of the library they decide to use.

Contents

• Installation
• Usage examples
• Contributing
• Versioning
• Authors
• License

Installation

Follow these instructions to get a copy of the project up and running on your local machine for development and testing purposes.

We also recommend usage of a virtual environment:

• conda
• virtualenv

Installing from source

Commands to install podium from source

git clone git@github.com:mttk/podium.git && cd podium
pip install .

Installing from wheel

The following release wheels are available for Podium:

• Version 1.1.0:

pip install http://takelab.fer.hr/podium/releases/podium-1.1.0-py3-none-any.whl

• Version 1.0.1:

pip install http://takelab.fer.hr/podium/releases/podium-1.0.1-py3-none-any.whl

• Version 1.0.0:

pip install http://takelab.fer.hr/podium/releases/podium-1.0.0-py3-none-any.whl

Installing from pip

[Coming soon] You can also install podium using pip

pip install podium-nlp

For more detailed installation instructions, check the installation page in the documentation.

Usage examples

For usage examples see the documentation pages walkthrough and examples

Loading datasets

Use some of our pre-defined datasets:

>>> from podium.datasets import SST
>>> sst_train, sst_dev, sst_test = SST.get_dataset_splits()
>>> print(sst_train)
SST({
    size: 6920,
    fields: [
        Field({
            name: text,
            keep_raw: False,
            is_target: False,
            vocab: Vocab({specials: ('<UNK>', '<PAD>'), eager: False, finalized: True, size: 16284})
        }),
        LabelField({
            name: label,
            keep_raw: False,
            is_target: True,
            vocab: Vocab({specials: (), eager: False, finalized: True, size: 2})
        })
    ]
})
>>> print(sst_train[222]) # A short example
Example({'text': (None, ['A', 'slick', ',', 'engrossing', 'melodrama', '.']), 'label': (None, 'positive')})

Load datasets from 🤗/datasets:

  >>> from podium.datasets.hf import HFDatasetConverter
  >>> import datasets
  >>> # Load the huggingface dataset
  >>> imdb = datasets.load_dataset('imdb')
  >>> print(imdb.keys())
  dict_keys(['train', 'test', 'unsupervised'])
  >>> # Wrap it so it can be used in Podium (without being loaded in memory!)
  >>> imdb_train, imdb_test, imdb_unsupervised = HFDatasetConverter.from_dataset_dict(imdb).values()
  >>> # We need to trigger Vocab construction
  >>> imdb_train.finalize_fields()
  >>> print(imdb_train)
  HFDatasetConverter({
    size: 25000,
    fields: [
        Field({
            name: text,
            keep_raw: False,
            is_target: False,
            vocab: Vocab({specials: ('<UNK>', '<PAD>'), eager: False, finalized: True, size: 280619})
        }),
        LabelField({
            name: label,
            keep_raw: False,
            is_target: True})
    ]
  })

Load your own dataset from a standardized tabular format (e.g. csv, tsv, jsonl):

>>> from podium.datasets import TabularDataset
>>> from podium import Vocab, Field, LabelField
>>> fields = {'premise':   Field('premise', numericalizer=Vocab()),
...           'hypothesis':Field('hypothesis', numericalizer=Vocab()),
...           'label':     LabelField('label')}
>>> dataset = TabularDataset('my_dataset.csv', format='csv', fields=fields)
>>> print(dataset)
TabularDataset({
    size: 1,
    fields: [
        Field({
            name: premise,
            is_target: False, 
            vocab: Vocab({specials: ('<UNK>', '<PAD>'), eager: False, finalized: True, size: 19})
        }),
        Field({
            name: hypothesis,
            is_target: False, 
            vocab: Vocab({specials: ('<UNK>', '<PAD>'), eager: False, finalized: True, size: 19})
        }),
        LabelField({
            name: label,
            is_target: True, 
            vocab: Vocab({specials: (), eager: False, finalized: True, size: 1})
        }),
    ]
})

Or define your own Dataset subclass (tutorial coming soon)

Define your preprocessing

We wrap dataset pre-processing in customizable Field classes. Each Field has an optional Vocab instance which automatically handles token-to-index conversion.

>>> from podium import Vocab, Field, LabelField
>>> vocab = Vocab(max_size=5000, min_freq=2)
>>> text = Field(name='text', numericalizer=vocab)
>>> label = LabelField(name='label')
>>> fields = {'text': text, 'label': label}
>>> sst_train, sst_dev, sst_test = SST.get_dataset_splits(fields=fields)
>>> print(vocab)
Vocab({specials: ('<UNK>', '<PAD>'), eager: True, finalized: True, size: 5000})

Each Field allows the user full flexibility modify the data in multiple stages:

• Prior to tokenization (by using pre-tokenization hooks)
• During tokenization (by using your own tokenizer)
• Post tokenization (by using post-tokenization hooks)

You can also completely disregard our preprocessing and define your own by setting your own numericalizer.

You could decide to lowercase all the characters and filter out all non-alphanumeric tokens:

>>> def lowercase(raw):
...     return raw.lower()
>>> def filter_alnum(raw, tokenized):
...     filtered_tokens = [token for token in tokenized if
...                        any([char.isalnum() for char in token])]
...     return raw, filtered_tokens
>>> text.add_pretokenize_hook(lowercase)
>>> text.add_posttokenize_hook(filter_alnum)
>>> fields = {'text': text, 'label': label}
>>> sst_train, sst_dev, sst_test = SST.get_dataset_splits(fields=fields)
>>> print(sst_train[222])
Example({'text': (None, ['a', 'slick', 'engrossing', 'melodrama']), 'label': (None, 'positive')})

Pre-tokenization hooks do not see the tokenized data and are applied (and modify) only raw data. Post-tokenization hooks have access to tokenized data, and can be applied to either raw or tokenized data.

Use preprocessing from other libraries

A common use-case is to incorporate existing components of pretrained language models, such as BERT. This is extremely simple to incorporate as part of our Fields. This snippet requires installation of the transformers (pip install transformers) library.

>>> from transformers import BertTokenizer
>>> # Load the tokenizer and fetch pad index
>>> tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
>>> pad_index = tokenizer.convert_tokens_to_ids(tokenizer.pad_token)
>>> # Define a BERT subword Field
>>> subword_field = Field(name="subword",
...                       padding_token=pad_index,
...                       tokenizer=tokenizer.tokenize,
...                       numericalizer=tokenizer.convert_tokens_to_ids)
>>> fields = {'text': subword_field, 'label': label}
>>> sst_train, sst_dev, sst_test = SST.get_dataset_splits(fields=fields)
>>> print(sst_train[222])
Example({'subword': (None, ['a', 'slick', ',', 'eng', '##ross', '##ing', 'mel', '##od', '##rama', '.']), 'label': (None, 'positive')})

Contributing

To learn more about making a contribution to Podium, please see our Contribution page.

Versioning

We use SemVer for versioning. For the versions available, see the tags on this repository.

Authors

• Podium is currently maintained by Ivan Smoković, Silvije Skudar, Mario Šaško, Filip Boltužić and Martin Tutek. A non-exhaustive but growing list of collaborators needs to mention: Domagoj Pluščec, Marin Kačan, Dunja Vesinger, Mate Mijolović.
• Project made as part of TakeLab at Faculty of Electrical Engineering and Computing, University of Zagreb

See also the list of contributors who participated in this project.

License

This project is licensed under the BSD 3-Clause - see the LICENSE file for details.