fastText |CircleCI|
===================
`fastText <https://fasttext.cc/>`__ is a library for efficient learning
of word representations and sentence classification.
In this document we present how to use fastText in python.
Table of contents
-----------------
- `Requirements <#requirements>`__
- `Installation <#installation>`__
- `Usage overview <#usage-overview>`__
- `Word representation model <#word-representation-model>`__
- `Text classification model <#text-classification-model>`__
- `IMPORTANT: Preprocessing data / encoding
conventions <#important-preprocessing-data-encoding-conventions>`__
- `More examples <#more-examples>`__
- `API <#api>`__
- `train_unsupervised parameters <#train_unsupervised-parameters>`__
- `train_supervised parameters <#train_supervised-parameters>`__
- `model object <#model-object>`__
Requirements
============
`fastText <https://fasttext.cc/>`__ builds on modern Mac OS and Linux
distributions. Since it uses C++11 features, it requires a compiler with
good C++11 support. You will need `Python <https://www.python.org/>`__
(version 2.7 or ≥ 3.4), `NumPy <http://www.numpy.org/>`__ &
`SciPy <https://www.scipy.org/>`__ and
`pybind11 <https://github.com/pybind/pybind11>`__.
Installation
============
To install the latest release, you can do :
.. code:: bash
$ pip install fasttext
or, to get the latest development version of fasttext, you can install
from our github repository :
.. code:: bash
$ git clone https://github.com/facebookresearch/fastText.git
$ cd fastText
$ sudo pip install .
$ # or :
$ sudo python setup.py install
Usage overview
==============
Word representation model
-------------------------
In order to learn word vectors, as `described
here <https://fasttext.cc/docs/en/references.html#enriching-word-vectors-with-subword-information>`__,
we can use ``fasttext.train_unsupervised`` function like this:
.. code:: py
import fasttext
# Skipgram model :
model = fasttext.train_unsupervised('data.txt', model='skipgram')
# or, cbow model :
model = fasttext.train_unsupervised('data.txt', model='cbow')
where ``data.txt`` is a training file containing utf-8 encoded text.
The returned ``model`` object represents your learned model, and you can
use it to retrieve information.
.. code:: py
print(model.words) # list of words in dictionary
print(model['king']) # get the vector of the word 'king'
Saving and loading a model object
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
You can save your trained model object by calling the function
``save_model``.
.. code:: py
model.save_model("model_filename.bin")
and retrieve it later thanks to the function ``load_model`` :
.. code:: py
model = fasttext.load_model("model_filename.bin")
For more information about word representation usage of fasttext, you
can refer to our `word representations
tutorial <https://fasttext.cc/docs/en/unsupervised-tutorial.html>`__.
Text classification model
-------------------------
In order to train a text classifier using the method `described
here <https://fasttext.cc/docs/en/references.html#bag-of-tricks-for-efficient-text-classification>`__,
we can use ``fasttext.train_supervised`` function like this:
.. code:: py
import fasttext
model = fasttext.train_supervised('data.train.txt')
where ``data.train.txt`` is a text file containing a training sentence
per line along with the labels. By default, we assume that labels are
words that are prefixed by the string ``__label__``
Once the model is trained, we can retrieve the list of words and labels:
.. code:: py
print(model.words)
print(model.labels)
To evaluate our model by computing the precision at 1 (P@1) and the
recall on a test set, we use the ``test`` function:
.. code:: py
def print_results(N, p, r):
print("N\t" + str(N))
print("P@{}\t{:.3f}".format(1, p))
print("R@{}\t{:.3f}".format(1, r))
print_results(*model.test('test.txt'))
We can also predict labels for a specific text :
.. code:: py
model.predict("Which baking dish is best to bake a banana bread ?")
By default, ``predict`` returns only one label : the one with the
highest probability. You can also predict more than one label by
specifying the parameter ``k``:
.. code:: py
model.predict("Which baking dish is best to bake a banana bread ?", k=3)
If you want to predict more than one sentence you can pass an array of
strings :
.. code:: py
model.predict(["Which baking dish is best to bake a banana bread ?", "Why not put knives in the dishwasher?"], k=3)
Of course, you can also save and load a model to/from a file as `in the
word representation usage <#saving-and-loading-a-model-object>`__.
For more information about text classification usage of fasttext, you
can refer to our `text classification
tutorial <https://fasttext.cc/docs/en/supervised-tutorial.html>`__.
Compress model files with quantization
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When you want to save a supervised model file, fastText can compress it
in order to have a much smaller model file by sacrificing only a little
bit performance.
.. code:: py
# with the previously trained `model` object, call :
model.quantize(input='data.train.txt', retrain=True)
# then display results and save the new model :
print_results(*model.test(valid_data))
model.save_model("model_filename.ftz")
``model_filename.ftz`` will have a much smaller size than
``model_filename.bin``.
For further reading on quantization, you can refer to `this paragraph
from our blog
post <https://fasttext.cc/blog/2017/10/02/blog-post.html#model-compression>`__.
IMPORTANT: Preprocessing data / encoding conventions
----------------------------------------------------
In general it is important to properly preprocess your data. In
particular our example scripts in the `root
folder <https://github.com/facebookresearch/fastText>`__ do this.
fastText assumes UTF-8 encoded text. All text must be `unicode for
Python2 <https://docs.python.org/2/library/functions.html#unicode>`__
and `str for
Python3 <https://docs.python.org/3.5/library/stdtypes.html#textseq>`__.
The passed text will be `encoded as UTF-8 by
pybind11 <https://pybind11.readthedocs.io/en/master/advanced/cast/strings.html?highlight=utf-8#strings-bytes-and-unicode-conversions>`__
before passed to the fastText C++ library. This means it is important to
use UTF-8 encoded text when building a model. On Unix-like systems you
can convert text using `iconv <https://en.wikipedia.org/wiki/Iconv>`__.
fastText will tokenize (split text into pieces) based on the following
ASCII characters (bytes). In particular, it is not aware of UTF-8
whitespace. We advice the user to convert UTF-8 whitespace / word
boundaries into one of the following symbols as appropiate.
- space
- tab
- vertical tab
- carriage return
- formfeed
- the null character
The newline character is used to delimit lines of text. In particular,
the EOS token is appended to a line of text if a newline character is
encountered. The only exception is if the number of tokens exceeds the
MAX\_LINE\_SIZE constant as defined in the `Dictionary
header <https://github.com/facebookresearch/fastText/blob/master/src/dictionary.h>`__.
This means if you have text that is not separate by newlines, such as
the `fil9 dataset <http://mattmahoney.net/dc/textdata>`__, it will be
broken into chunks with MAX\_LINE\_SIZE of tokens and the EOS token is
not appended.
The length of a token is the number of UTF-8 characters by considering
the `leading two bits of a
byte <https://en.wikipedia.org/wiki/UTF-8#Description>`__ to identify
`subsequent bytes of a multi-byte
sequence <https://github.com/facebookresearch/fastText/blob/master/src/dictionary.cc>`__.
Knowing this is especially important when choosing the minimum and
maximum length of subwords. Further, the EOS token (as specified in the
`Dictionary
header <https://github.com/facebookresearch/fastText/blob/master/src/dictionary.h>`__)
is considered a character and will not be broken into subwords.
More examples
-------------
In order to have a better knowledge of fastText models, please consider
the main
`README <https://github.com/facebookresearch/fastText/blob/master/README.md>`__
and in particular `the tutorials on our
website <https://fasttext.cc/docs/en/supervised-tutorial.html>`__.
You can find further python examples in `the doc
folder <https://github.com/facebookresearch/fastText/tree/master/python/doc/examples>`__.
As with any package you can get help on any Python function using the
help function.
For example
::
+>>> import fasttext
+>>> help(fasttext.FastText)
Help on module fasttext.FastText in fasttext:
NAME
fasttext.FastText
DESCRIPTION
# Copyright (c) 2017-present, Facebook, Inc.
# All rights reserved.
#
# This source code is licensed under the MIT license found in the
# LICENSE file in the root directory of this source tree.
FUNCTIONS
load_model(path)
Load a model given a filepath and return a model object.
tokenize(text)
Given a string of text, tokenize it and return a list of tokens
[...]
API
===
``train_unsupervised`` parameters
---------------------------------
.. code:: python
input # training file path (required)
model # unsupervised fasttext model {cbow, skipgram} [skipgram]
lr # learning rate [0.05]
dim # size of word vectors [100]
ws # size of the context window [5]
epoch # number of epochs [5]
minCount # minimal number of word occurences [5]
minn # min length of char ngram [3]
maxn # max length of char ngram [6]
neg # number of negatives sampled [5]
wordNgrams # max length of word ngram [1]
loss # loss function {ns, hs, softmax, ova} [ns]
bucket # number of buckets [2000000]
thread # number of threads [number of cpus]
lrUpdateRate # change the rate of updates for the learning rate [100]
t # sampling threshold [0.0001]
verbose # verbose [2]
``train_supervised`` parameters
-------------------------------
.. code:: python
input # training file path (required)
lr # learning rate [0.1]
dim # size of word vectors [100]
ws # size of the context window [5]
epoch # number of epochs [5]
minCount # minimal number of word occurences [1]
minCountLabel # minimal number of label occurences [1]
minn # min length of char ngram [0]
maxn # max length of char ngram [0]
neg # number of negatives sampled [5]
wordNgrams # max length of word ngram [1]
loss # loss function {ns, hs, softmax, ova} [softmax]
bucket # number of buckets [2000000]
thread # number of threads [number of cpus]
lrUpdateRate # change the rate of updates for the learning rate [100]
t # sampling threshold [0.0001]
label # label prefix ['__label__']
verbose # verbose [2]
pretrainedVectors # pretrained word vectors (.vec file) for supervised learning []
``model`` object
----------------
``train_supervised``, ``train_unsupervised`` and ``load_model``
functions return an instance of ``_FastText`` class, that we generaly
name ``model`` object.
This object exposes those training arguments as properties : ``lr``,
``dim``, ``ws``, ``epoch``, ``minCount``, ``minCountLabel``, ``minn``,
``maxn``, ``neg``, ``wordNgrams``, ``loss``, ``bucket``, ``thread``,
``lrUpdateRate``, ``t``, ``label``, ``verbose``, ``pretrainedVectors``.
So ``model.wordNgrams`` will give you the max length of word ngram used
for training this model.
In addition, the object exposes several functions :
.. code:: python
get_dimension # Get the dimension (size) of a lookup vector (hidden layer).
# This is equivalent to `dim` property.
get_input_vector # Given an index, get the corresponding vector of the Input Matrix.
get_input_matrix # Get a copy of the full input matrix of a Model.
get_labels # Get the entire list of labels of the dictionary
# This is equivalent to `labels` property.
get_line # Split a line of text into words and labels.
get_output_matrix # Get a copy of the full output matrix of a Model.
get_sentence_vector # Given a string, get a single vector represenation. This function
# assumes to be given a single line of text. We split words on
# whitespace (space, newline, tab, vertical tab) and the control
# characters carriage return, formfeed and the null character.
get_subword_id # Given a subword, return the index (within input matrix) it hashes to.
get_subwords # Given a word, get the subwords and their indicies.
get_word_id # Given a word, get the word id within the dictionary.
get_word_vector # Get the vector representation of word.
get_words # Get the entire list of words of the dictionary
# This is equivalent to `words` property.
is_quantized # whether the model has been quantized
predict # Given a string, get a list of labels and a list of corresponding probabilities.
quantize # Quantize the model reducing the size of the model and it's memory footprint.
save_model # Save the model to the given path
test # Evaluate supervised model using file given by path
test_label # Return the precision and recall score for each label.
The properties ``words``, ``labels`` return the words and labels from
the dictionary :
.. code:: py
model.words # equivalent to model.get_words()
model.labels # equivalent to model.get_labels()
The object overrides ``__getitem__`` and ``__contains__`` functions in
order to return the representation of a word and to check if a word is
in the vocabulary.
.. code:: py
model['king'] # equivalent to model.get_word_vector('king')
'king' in model # equivalent to `'king' in model.get_words()`
Join the fastText community
---------------------------
- `Facebook page <https://www.facebook.com/groups/1174547215919768>`__
- `Stack
overflow <https://stackoverflow.com/questions/tagged/fasttext>`__
- `Google
group <https://groups.google.com/forum/#!forum/fasttext-library>`__
- `GitHub <https://github.com/facebookresearch/fastText>`__
.. |CircleCI| image:: https://circleci.com/gh/facebookresearch/fastText/tree/master.svg?style=svg
:target: https://circleci.com/gh/facebookresearch/fastText/tree/master
Raw data
{
"_id": null,
"home_page": "https://github.com/facebookresearch/fastText",
"name": "fasttext-numpy2-wheel",
"maintainer": null,
"docs_url": null,
"requires_python": null,
"maintainer_email": null,
"keywords": null,
"author": "Onur Celebi",
"author_email": "celebio@fb.com",
"download_url": "https://files.pythonhosted.org/packages/c0/8a/8c72c92cd91146ce9948421f3967bc2775deac5215eb8fa3b0487020881f/fasttext_numpy2_wheel-0.9.2.tar.gz",
"platform": null,
"description": "fastText |CircleCI|\n===================\n\n`fastText <https://fasttext.cc/>`__ is a library for efficient learning\nof word representations and sentence classification.\n\nIn this document we present how to use fastText in python.\n\nTable of contents\n-----------------\n\n- `Requirements <#requirements>`__\n- `Installation <#installation>`__\n- `Usage overview <#usage-overview>`__\n- `Word representation model <#word-representation-model>`__\n- `Text classification model <#text-classification-model>`__\n- `IMPORTANT: Preprocessing data / encoding\n conventions <#important-preprocessing-data-encoding-conventions>`__\n- `More examples <#more-examples>`__\n- `API <#api>`__\n- `train_unsupervised parameters <#train_unsupervised-parameters>`__\n- `train_supervised parameters <#train_supervised-parameters>`__\n- `model object <#model-object>`__\n\nRequirements\n============\n\n`fastText <https://fasttext.cc/>`__ builds on modern Mac OS and Linux\ndistributions. Since it uses C++11 features, it requires a compiler with\ngood C++11 support. You will need `Python <https://www.python.org/>`__\n(version 2.7 or \u2265 3.4), `NumPy <http://www.numpy.org/>`__ &\n`SciPy <https://www.scipy.org/>`__ and\n`pybind11 <https://github.com/pybind/pybind11>`__.\n\nInstallation\n============\n\nTo install the latest release, you can do :\n\n.. code:: bash\n\n $ pip install fasttext\n\nor, to get the latest development version of fasttext, you can install\nfrom our github repository :\n\n.. code:: bash\n\n $ git clone https://github.com/facebookresearch/fastText.git\n $ cd fastText\n $ sudo pip install .\n $ # or :\n $ sudo python setup.py install\n\nUsage overview\n==============\n\nWord representation model\n-------------------------\n\nIn order to learn word vectors, as `described\nhere <https://fasttext.cc/docs/en/references.html#enriching-word-vectors-with-subword-information>`__,\nwe can use ``fasttext.train_unsupervised`` function like this:\n\n.. code:: py\n\n import fasttext\n\n # Skipgram model :\n model = fasttext.train_unsupervised('data.txt', model='skipgram')\n\n # or, cbow model :\n model = fasttext.train_unsupervised('data.txt', model='cbow')\n\nwhere ``data.txt`` is a training file containing utf-8 encoded text.\n\nThe returned ``model`` object represents your learned model, and you can\nuse it to retrieve information.\n\n.. code:: py\n\n print(model.words) # list of words in dictionary\n print(model['king']) # get the vector of the word 'king'\n\nSaving and loading a model object\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\nYou can save your trained model object by calling the function\n``save_model``.\n\n.. code:: py\n\n model.save_model(\"model_filename.bin\")\n\nand retrieve it later thanks to the function ``load_model`` :\n\n.. code:: py\n\n model = fasttext.load_model(\"model_filename.bin\")\n\nFor more information about word representation usage of fasttext, you\ncan refer to our `word representations\ntutorial <https://fasttext.cc/docs/en/unsupervised-tutorial.html>`__.\n\nText classification model\n-------------------------\n\nIn order to train a text classifier using the method `described\nhere <https://fasttext.cc/docs/en/references.html#bag-of-tricks-for-efficient-text-classification>`__,\nwe can use ``fasttext.train_supervised`` function like this:\n\n.. code:: py\n\n import fasttext\n\n model = fasttext.train_supervised('data.train.txt')\n\nwhere ``data.train.txt`` is a text file containing a training sentence\nper line along with the labels. By default, we assume that labels are\nwords that are prefixed by the string ``__label__``\n\nOnce the model is trained, we can retrieve the list of words and labels:\n\n.. code:: py\n\n print(model.words)\n print(model.labels)\n\nTo evaluate our model by computing the precision at 1 (P@1) and the\nrecall on a test set, we use the ``test`` function:\n\n.. code:: py\n\n def print_results(N, p, r):\n print(\"N\\t\" + str(N))\n print(\"P@{}\\t{:.3f}\".format(1, p))\n print(\"R@{}\\t{:.3f}\".format(1, r))\n\n print_results(*model.test('test.txt'))\n\nWe can also predict labels for a specific text :\n\n.. code:: py\n\n model.predict(\"Which baking dish is best to bake a banana bread ?\")\n\nBy default, ``predict`` returns only one label : the one with the\nhighest probability. You can also predict more than one label by\nspecifying the parameter ``k``:\n\n.. code:: py\n\n model.predict(\"Which baking dish is best to bake a banana bread ?\", k=3)\n\nIf you want to predict more than one sentence you can pass an array of\nstrings :\n\n.. code:: py\n\n model.predict([\"Which baking dish is best to bake a banana bread ?\", \"Why not put knives in the dishwasher?\"], k=3)\n\nOf course, you can also save and load a model to/from a file as `in the\nword representation usage <#saving-and-loading-a-model-object>`__.\n\nFor more information about text classification usage of fasttext, you\ncan refer to our `text classification\ntutorial <https://fasttext.cc/docs/en/supervised-tutorial.html>`__.\n\nCompress model files with quantization\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\nWhen you want to save a supervised model file, fastText can compress it\nin order to have a much smaller model file by sacrificing only a little\nbit performance.\n\n.. code:: py\n\n # with the previously trained `model` object, call :\n model.quantize(input='data.train.txt', retrain=True)\n\n # then display results and save the new model :\n print_results(*model.test(valid_data))\n model.save_model(\"model_filename.ftz\")\n\n``model_filename.ftz`` will have a much smaller size than\n``model_filename.bin``.\n\nFor further reading on quantization, you can refer to `this paragraph\nfrom our blog\npost <https://fasttext.cc/blog/2017/10/02/blog-post.html#model-compression>`__.\n\nIMPORTANT: Preprocessing data / encoding conventions\n----------------------------------------------------\n\nIn general it is important to properly preprocess your data. In\nparticular our example scripts in the `root\nfolder <https://github.com/facebookresearch/fastText>`__ do this.\n\nfastText assumes UTF-8 encoded text. All text must be `unicode for\nPython2 <https://docs.python.org/2/library/functions.html#unicode>`__\nand `str for\nPython3 <https://docs.python.org/3.5/library/stdtypes.html#textseq>`__.\nThe passed text will be `encoded as UTF-8 by\npybind11 <https://pybind11.readthedocs.io/en/master/advanced/cast/strings.html?highlight=utf-8#strings-bytes-and-unicode-conversions>`__\nbefore passed to the fastText C++ library. This means it is important to\nuse UTF-8 encoded text when building a model. On Unix-like systems you\ncan convert text using `iconv <https://en.wikipedia.org/wiki/Iconv>`__.\n\nfastText will tokenize (split text into pieces) based on the following\nASCII characters (bytes). In particular, it is not aware of UTF-8\nwhitespace. We advice the user to convert UTF-8 whitespace / word\nboundaries into one of the following symbols as appropiate.\n\n- space\n- tab\n- vertical tab\n- carriage return\n- formfeed\n- the null character\n\nThe newline character is used to delimit lines of text. In particular,\nthe EOS token is appended to a line of text if a newline character is\nencountered. The only exception is if the number of tokens exceeds the\nMAX\\_LINE\\_SIZE constant as defined in the `Dictionary\nheader <https://github.com/facebookresearch/fastText/blob/master/src/dictionary.h>`__.\nThis means if you have text that is not separate by newlines, such as\nthe `fil9 dataset <http://mattmahoney.net/dc/textdata>`__, it will be\nbroken into chunks with MAX\\_LINE\\_SIZE of tokens and the EOS token is\nnot appended.\n\nThe length of a token is the number of UTF-8 characters by considering\nthe `leading two bits of a\nbyte <https://en.wikipedia.org/wiki/UTF-8#Description>`__ to identify\n`subsequent bytes of a multi-byte\nsequence <https://github.com/facebookresearch/fastText/blob/master/src/dictionary.cc>`__.\nKnowing this is especially important when choosing the minimum and\nmaximum length of subwords. Further, the EOS token (as specified in the\n`Dictionary\nheader <https://github.com/facebookresearch/fastText/blob/master/src/dictionary.h>`__)\nis considered a character and will not be broken into subwords.\n\nMore examples\n-------------\n\nIn order to have a better knowledge of fastText models, please consider\nthe main\n`README <https://github.com/facebookresearch/fastText/blob/master/README.md>`__\nand in particular `the tutorials on our\nwebsite <https://fasttext.cc/docs/en/supervised-tutorial.html>`__.\n\nYou can find further python examples in `the doc\nfolder <https://github.com/facebookresearch/fastText/tree/master/python/doc/examples>`__.\n\nAs with any package you can get help on any Python function using the\nhelp function.\n\nFor example\n\n::\n\n +>>> import fasttext\n +>>> help(fasttext.FastText)\n\n Help on module fasttext.FastText in fasttext:\n\n NAME\n fasttext.FastText\n\n DESCRIPTION\n # Copyright (c) 2017-present, Facebook, Inc.\n # All rights reserved.\n #\n # This source code is licensed under the MIT license found in the\n # LICENSE file in the root directory of this source tree.\n\n FUNCTIONS\n load_model(path)\n Load a model given a filepath and return a model object.\n\n tokenize(text)\n Given a string of text, tokenize it and return a list of tokens\n [...]\n\nAPI\n===\n\n``train_unsupervised`` parameters\n---------------------------------\n\n.. code:: python\n\n input # training file path (required)\n model # unsupervised fasttext model {cbow, skipgram} [skipgram]\n lr # learning rate [0.05]\n dim # size of word vectors [100]\n ws # size of the context window [5]\n epoch # number of epochs [5]\n minCount # minimal number of word occurences [5]\n minn # min length of char ngram [3]\n maxn # max length of char ngram [6]\n neg # number of negatives sampled [5]\n wordNgrams # max length of word ngram [1]\n loss # loss function {ns, hs, softmax, ova} [ns]\n bucket # number of buckets [2000000]\n thread # number of threads [number of cpus]\n lrUpdateRate # change the rate of updates for the learning rate [100]\n t # sampling threshold [0.0001]\n verbose # verbose [2]\n\n``train_supervised`` parameters\n-------------------------------\n\n.. code:: python\n\n input # training file path (required)\n lr # learning rate [0.1]\n dim # size of word vectors [100]\n ws # size of the context window [5]\n epoch # number of epochs [5]\n minCount # minimal number of word occurences [1]\n minCountLabel # minimal number of label occurences [1]\n minn # min length of char ngram [0]\n maxn # max length of char ngram [0]\n neg # number of negatives sampled [5]\n wordNgrams # max length of word ngram [1]\n loss # loss function {ns, hs, softmax, ova} [softmax]\n bucket # number of buckets [2000000]\n thread # number of threads [number of cpus]\n lrUpdateRate # change the rate of updates for the learning rate [100]\n t # sampling threshold [0.0001]\n label # label prefix ['__label__']\n verbose # verbose [2]\n pretrainedVectors # pretrained word vectors (.vec file) for supervised learning []\n\n``model`` object\n----------------\n\n``train_supervised``, ``train_unsupervised`` and ``load_model``\nfunctions return an instance of ``_FastText`` class, that we generaly\nname ``model`` object.\n\nThis object exposes those training arguments as properties : ``lr``,\n``dim``, ``ws``, ``epoch``, ``minCount``, ``minCountLabel``, ``minn``,\n``maxn``, ``neg``, ``wordNgrams``, ``loss``, ``bucket``, ``thread``,\n``lrUpdateRate``, ``t``, ``label``, ``verbose``, ``pretrainedVectors``.\nSo ``model.wordNgrams`` will give you the max length of word ngram used\nfor training this model.\n\nIn addition, the object exposes several functions :\n\n.. code:: python\n\n get_dimension # Get the dimension (size) of a lookup vector (hidden layer).\n # This is equivalent to `dim` property.\n get_input_vector # Given an index, get the corresponding vector of the Input Matrix.\n get_input_matrix # Get a copy of the full input matrix of a Model.\n get_labels # Get the entire list of labels of the dictionary\n # This is equivalent to `labels` property.\n get_line # Split a line of text into words and labels.\n get_output_matrix # Get a copy of the full output matrix of a Model.\n get_sentence_vector # Given a string, get a single vector represenation. This function\n # assumes to be given a single line of text. We split words on\n # whitespace (space, newline, tab, vertical tab) and the control\n # characters carriage return, formfeed and the null character.\n get_subword_id # Given a subword, return the index (within input matrix) it hashes to.\n get_subwords # Given a word, get the subwords and their indicies.\n get_word_id # Given a word, get the word id within the dictionary.\n get_word_vector # Get the vector representation of word.\n get_words # Get the entire list of words of the dictionary\n # This is equivalent to `words` property.\n is_quantized # whether the model has been quantized\n predict # Given a string, get a list of labels and a list of corresponding probabilities.\n quantize # Quantize the model reducing the size of the model and it's memory footprint.\n save_model # Save the model to the given path\n test # Evaluate supervised model using file given by path\n test_label # Return the precision and recall score for each label.\n\nThe properties ``words``, ``labels`` return the words and labels from\nthe dictionary :\n\n.. code:: py\n\n model.words # equivalent to model.get_words()\n model.labels # equivalent to model.get_labels()\n\nThe object overrides ``__getitem__`` and ``__contains__`` functions in\norder to return the representation of a word and to check if a word is\nin the vocabulary.\n\n.. code:: py\n\n model['king'] # equivalent to model.get_word_vector('king')\n 'king' in model # equivalent to `'king' in model.get_words()`\n\nJoin the fastText community\n---------------------------\n\n- `Facebook page <https://www.facebook.com/groups/1174547215919768>`__\n- `Stack\n overflow <https://stackoverflow.com/questions/tagged/fasttext>`__\n- `Google\n group <https://groups.google.com/forum/#!forum/fasttext-library>`__\n- `GitHub <https://github.com/facebookresearch/fastText>`__\n\n.. |CircleCI| image:: https://circleci.com/gh/facebookresearch/fastText/tree/master.svg?style=svg\n :target: https://circleci.com/gh/facebookresearch/fastText/tree/master\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "fasttext Python bindings",
"version": "0.9.2",
"project_urls": {
"Homepage": "https://github.com/facebookresearch/fastText"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "38e4e9b6407538100ced67ddde5941a071ef01482badd023ab8ebcc0d5e6797b",
"md5": "b4ee9c28dc03cd0d91f37698873f49ed",
"sha256": "5432ebdbac8cb90d91305f9d985bfd64fbd0d30a8f27de9a508c926e47d33051"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl",
"has_sig": false,
"md5_digest": "b4ee9c28dc03cd0d91f37698873f49ed",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": null,
"size": 4557717,
"upload_time": "2024-11-05T13:54:10",
"upload_time_iso_8601": "2024-11-05T13:54:10.179665Z",
"url": "https://files.pythonhosted.org/packages/38/e4/e9b6407538100ced67ddde5941a071ef01482badd023ab8ebcc0d5e6797b/fasttext_numpy2_wheel-0.9.2-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "5c45bd33da5fbcbff711dcbd1d20f28329ece7a95b707ac32709df22a8a02349",
"md5": "51d8e412dcb42aa00faf3976c3b3ba88",
"sha256": "d47fc7b0f7a4410eb45437397803f9be217240baba660c1cdc8b493365dd1421"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "51d8e412dcb42aa00faf3976c3b3ba88",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": null,
"size": 4646860,
"upload_time": "2024-11-05T13:54:07",
"upload_time_iso_8601": "2024-11-05T13:54:07.444153Z",
"url": "https://files.pythonhosted.org/packages/5c/45/bd33da5fbcbff711dcbd1d20f28329ece7a95b707ac32709df22a8a02349/fasttext_numpy2_wheel-0.9.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "92a44dce97deb92057c2a71fc5f0d26a28507ecf8c6c64cd93f9edbd410a4caa",
"md5": "6c57fdf76b91e2cfa78459bc4027685d",
"sha256": "ef2f5caf30f7186eb08d3fc3ef598895aa534fc3a1cc31150904aaacee8a3c80"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl",
"has_sig": false,
"md5_digest": "6c57fdf76b91e2cfa78459bc4027685d",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": null,
"size": 4578625,
"upload_time": "2024-11-05T13:54:15",
"upload_time_iso_8601": "2024-11-05T13:54:15.645928Z",
"url": "https://files.pythonhosted.org/packages/92/a4/4dce97deb92057c2a71fc5f0d26a28507ecf8c6c64cd93f9edbd410a4caa/fasttext_numpy2_wheel-0.9.2-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "34e7c664619e42a38f0064f8d9100b76b3bb57bf2679360dece5cf245a5b980e",
"md5": "747d8f0d157d0f182219729b9872dbfe",
"sha256": "21ca228aad6c75349e1ccb231fa454b4ec5bdc2f8a510718d2b74b5742c65f20"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "747d8f0d157d0f182219729b9872dbfe",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": null,
"size": 4669758,
"upload_time": "2024-11-05T13:54:09",
"upload_time_iso_8601": "2024-11-05T13:54:09.381068Z",
"url": "https://files.pythonhosted.org/packages/34/e7/c664619e42a38f0064f8d9100b76b3bb57bf2679360dece5cf245a5b980e/fasttext_numpy2_wheel-0.9.2-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "df13f8a21e5aa6c0a3e382c2346322b6d70fe1435c85cd57c8913b56fc72441d",
"md5": "f59ccb782bd7d0cb4c6093906cf2553f",
"sha256": "e1d50a174457a11d9317c5eaef007d0971a1baed4b15b8f5c404d69ba7ed2346"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl",
"has_sig": false,
"md5_digest": "f59ccb782bd7d0cb4c6093906cf2553f",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": null,
"size": 4583847,
"upload_time": "2024-11-05T13:54:19",
"upload_time_iso_8601": "2024-11-05T13:54:19.266121Z",
"url": "https://files.pythonhosted.org/packages/df/13/f8a21e5aa6c0a3e382c2346322b6d70fe1435c85cd57c8913b56fc72441d/fasttext_numpy2_wheel-0.9.2-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "d49da78dfd3e996f5d3c91bcd14a4e249a90da038b99287fda43211281724fdb",
"md5": "8812b53c138906c38aa192507c4c21df",
"sha256": "3e70c79c9ca670f36fcf639eea87bf44adff4082a966c62053747a7dee0f2898"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "8812b53c138906c38aa192507c4c21df",
"packagetype": "bdist_wheel",
"python_version": "cp312",
"requires_python": null,
"size": 4686609,
"upload_time": "2024-11-05T13:54:11",
"upload_time_iso_8601": "2024-11-05T13:54:11.194367Z",
"url": "https://files.pythonhosted.org/packages/d4/9d/a78dfd3e996f5d3c91bcd14a4e249a90da038b99287fda43211281724fdb/fasttext_numpy2_wheel-0.9.2-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "7b0656b18b5f7e5d75497d2a5db654078384b8ca25caffbb22ab827f7904e540",
"md5": "8dceb8ab54ad09f421fafad722033a4a",
"sha256": "74e6f4d6c5453a7cd71ac61766c70797165e42de99458c7ce5b3310ce661531a"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl",
"has_sig": false,
"md5_digest": "8dceb8ab54ad09f421fafad722033a4a",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": null,
"size": 4558804,
"upload_time": "2024-11-05T13:54:23",
"upload_time_iso_8601": "2024-11-05T13:54:23.524985Z",
"url": "https://files.pythonhosted.org/packages/7b/06/56b18b5f7e5d75497d2a5db654078384b8ca25caffbb22ab827f7904e540/fasttext_numpy2_wheel-0.9.2-cp38-cp38-manylinux_2_17_i686.manylinux2014_i686.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "12eab9d18381f9c27fa2a3e052b8737dde1d53960dfbb650e489be0f77d19993",
"md5": "86db755bfca18aa023520665f47232a6",
"sha256": "5c7baab6f9419c99c4a69b01c06886314d147ea547c7be87e1ba68d37316bb1c"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "86db755bfca18aa023520665f47232a6",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": null,
"size": 4650883,
"upload_time": "2024-11-05T13:54:15",
"upload_time_iso_8601": "2024-11-05T13:54:15.584987Z",
"url": "https://files.pythonhosted.org/packages/12/ea/b9d18381f9c27fa2a3e052b8737dde1d53960dfbb650e489be0f77d19993/fasttext_numpy2_wheel-0.9.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "826d10b91f802316ae538a4cc781e6f4896139728a9975ca7c8bbd02c335ba71",
"md5": "e73bef591aadbe79ba84928c20af2ff0",
"sha256": "7f612905192f23be4ff883ece3c1ee44fca52f0c308c2bc4f2e4b6c0e86d99a2"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl",
"has_sig": false,
"md5_digest": "e73bef591aadbe79ba84928c20af2ff0",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": null,
"size": 4552775,
"upload_time": "2024-11-05T13:54:26",
"upload_time_iso_8601": "2024-11-05T13:54:26.259077Z",
"url": "https://files.pythonhosted.org/packages/82/6d/10b91f802316ae538a4cc781e6f4896139728a9975ca7c8bbd02c335ba71/fasttext_numpy2_wheel-0.9.2-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "75fa917d1437a70f6a4ef69c44025d6ab1f13c40903956ed203993fe1709a48e",
"md5": "4a06420c8969acf413c9ff7f96b966fe",
"sha256": "e50ab4f2ddd050a42fe7d634fb6237e855739e6d186aa9e6f879f0606bff8256"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "4a06420c8969acf413c9ff7f96b966fe",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": null,
"size": 4645625,
"upload_time": "2024-11-05T13:54:16",
"upload_time_iso_8601": "2024-11-05T13:54:16.943668Z",
"url": "https://files.pythonhosted.org/packages/75/fa/917d1437a70f6a4ef69c44025d6ab1f13c40903956ed203993fe1709a48e/fasttext_numpy2_wheel-0.9.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "c08a8c72c92cd91146ce9948421f3967bc2775deac5215eb8fa3b0487020881f",
"md5": "2257b327976fe58e958ba087648f1762",
"sha256": "484bb7efb0d07c5b6235c8ab44d5e7ddcde5727a25dbc04d18175f777c1e799c"
},
"downloads": -1,
"filename": "fasttext_numpy2_wheel-0.9.2.tar.gz",
"has_sig": false,
"md5_digest": "2257b327976fe58e958ba087648f1762",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 73494,
"upload_time": "2024-11-05T13:54:19",
"upload_time_iso_8601": "2024-11-05T13:54:19.159373Z",
"url": "https://files.pythonhosted.org/packages/c0/8a/8c72c92cd91146ce9948421f3967bc2775deac5215eb8fa3b0487020881f/fasttext_numpy2_wheel-0.9.2.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-11-05 13:54:19",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "facebookresearch",
"github_project": "fastText",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "fasttext-numpy2-wheel"
}
JSON
