Text Vectorization Pipeline

This example illustrates how Dask-ML can be used to classify large textual datasets in parallel. It is adapted from this scikit-learn example.

The primary differences are that

  • We fit the entire model, including text vectorization, as a pipeline.
  • We use dask collections like Dask Bag, Dask Dataframe, and Dask Array rather than generators to work with larger than memory datasets.
In [ ]:
from dask.distributed import Client, progress

client = Client(n_workers=2, threads_per_worker=2, memory_limit='2GB')
client

Fetch the data

Scikit-Learn provides a utility to fetch the newsgroups dataset.

In [ ]:
import sklearn.datasets

bunch = sklearn.datasets.fetch_20newsgroups()

The data from scikit-learn isn't too large, so the data is just returned in memory. Each document is a string. The target we're predicting is an integer, which codes the topic of the post.

We'll load the documents and targets directly into a dask DataFrame. In practice, on a larger than memory dataset, you would likely load the documents from disk or cloud storage using dask.bag or dask.delayed.

In [ ]:
import dask.dataframe as dd
import pandas as pd

df = dd.from_pandas(pd.DataFrame({"text": bunch.data, "target": bunch.target}),
                    npartitions=25)

df

Each row in the text column has a bit of metadata and the full text of a post.

In [ ]:
print(df.head().loc[0, 'text'][:500])

Feature Hashing

Dask's HashingVectorizer provides a similar API to scikit-learn's implementation. In fact, Dask-ML's implementation uses scikit-learn's, applying it to each partition of the input dask.dataframe.Series or dask.bag.Bag.

Transformation, once we actually compute the result, happens in parallel and returns a dask Array.

In [ ]:
import dask_ml.feature_extraction.text

vect = dask_ml.feature_extraction.text.HashingVectorizer()
X = vect.fit_transform(df['text'])
X

The output array X has unknown chunk sizes becase the input dask Series or Bags don't know their own length.

Each block in X is a scipy.sparse matrix.

In [ ]:
X.blocks[0].compute()

This is a document-term matrix. Each row is the hashed representation of the original post.

Classification Pipeline

We can combine the HashingVectorizer with Incremental and a classifier like scikit-learn's SGDClassifier to create a classification pipeline.

We'll predict whether the topic was in the comp category.

In [ ]:
bunch.target_names
In [ ]:
import numpy as np

positive = np.arange(len(bunch.target_names))[['comp' in x for x in bunch.target_names]]
y = df['target'].isin(positive).astype(int)
y
In [ ]:
import numpy as np
import sklearn.linear_model
import sklearn.pipeline

import dask_ml.wrappers

Because the input comes from a dask Series, with unknown chunk sizes, we need to specify assume_equal_chunks=True. This tells Dask-ML that we know that each partition in X matches a partition in y.

In [ ]:
sgd = sklearn.linear_model.SGDClassifier(
    tol=1e-3
)
clf = dask_ml.wrappers.Incremental(
    sgd, scoring='accuracy', assume_equal_chunks=True
)
pipe = sklearn.pipeline.make_pipeline(vect, clf)

SGDClassifier.partial_fit needs to know the full set of classes up front. Becuase our sgd is wrapped inside an Incremental, we need to pass it through as the incremental__classes keyword argument in fit.

In [ ]:
pipe.fit(df['text'], y,
         incremental__classes=[0, 1]);

As usual, Incremental.predict lazily returns the predictions as a dask Array.

In [ ]:
predictions = pipe.predict(df['text'])
predictions

We can compute the predictions and score in parallel with dask_ml.metrics.accuracy_score.

In [ ]:
dask_ml.metrics.accuracy_score(y, predictions)

This simple combination of a HashingVectorizer and SGDClassifier is pretty effective at this prediction task.