koaning / embetter

just a bunch of useful embeddings
https://koaning.github.io/embetter/
MIT License
469 stars 15 forks source link

embetter

"Just a bunch of useful embeddings to get started quickly."


Embetter implements scikit-learn compatible embeddings for computer vision and text. It should make it very easy to quickly build proof of concepts using scikit-learn pipelines and, in particular, should help with bulk labelling. It's also meant to play nice with bulk and scikit-partial but it can also be used together with your favorite ANN solution like lancedb.

Install

You can install via pip.

python -m pip install embetter

Many of the embeddings are optional depending on your use-case, so if you want to nit-pick to download only the tools that you need:

python -m pip install "embetter[text]"
python -m pip install "embetter[spacy]"
python -m pip install "embetter[sense2vec]"
python -m pip install "embetter[gensim]"
python -m pip install "embetter[bpemb]"
python -m pip install "embetter[vision]"
python -m pip install "embetter[all]"

API Design

This is what's being implemented now.

# Helpers to grab text or image from pandas column.
from embetter.grab import ColumnGrabber

# Representations/Helpers for computer vision
from embetter.vision import ImageLoader, TimmEncoder, ColorHistogramEncoder

# Representations for text
from embetter.text import SentenceEncoder, MatryoshkaEncoder, Sense2VecEncoder, BytePairEncoder, spaCyEncoder, GensimEncoder

# Representations from multi-modal models
from embetter.multi import ClipEncoder

# Finetuning components 
from embetter.finetune import FeedForwardTuner, ContrastiveTuner, ContrastiveLearner, SbertLearner

# External embedding providers, typically needs an API key
from embetter.external import CohereEncoder, OpenAIEncoder

All of these components are scikit-learn compatible, which means that you can apply them as you would normally in a scikit-learn pipeline. Just be aware that these components are stateless. They won't require training as these are all pretrained tools.

Text Example

import pandas as pd
from sklearn.pipeline import make_pipeline 
from sklearn.linear_model import LogisticRegression

from embetter.grab import ColumnGrabber
from embetter.text import SentenceEncoder

# This pipeline grabs the `text` column from a dataframe
# which then get fed into Sentence-Transformers' all-MiniLM-L6-v2.
text_emb_pipeline = make_pipeline(
  ColumnGrabber("text"),
  SentenceEncoder('all-MiniLM-L6-v2')
)

# This pipeline can also be trained to make predictions, using
# the embedded features. 
text_clf_pipeline = make_pipeline(
  text_emb_pipeline,
  LogisticRegression()
)

dataf = pd.DataFrame({
  "text": ["positive sentiment", "super negative"],
  "label_col": ["pos", "neg"]
})
X = text_emb_pipeline.fit_transform(dataf, dataf['label_col'])
text_clf_pipeline.fit(dataf, dataf['label_col']).predict(dataf)

Image Example

The goal of the API is to allow pipelines like this:

import pandas as pd
from sklearn.pipeline import make_pipeline 
from sklearn.linear_model import LogisticRegression

from embetter.grab import ColumnGrabber
from embetter.vision import ImageLoader
from embetter.multi import ClipEncoder

# This pipeline grabs the `img_path` column from a dataframe
# then it grabs the image paths and turns them into `PIL.Image` objects
# which then get fed into CLIP which can also handle images.
image_emb_pipeline = make_pipeline(
  ColumnGrabber("img_path"),
  ImageLoader(convert="RGB"),
  ClipEncoder()
)

dataf = pd.DataFrame({
  "img_path": ["tests/data/thiscatdoesnotexist.jpeg"]
})
image_emb_pipeline.fit_transform(dataf)

Batched Learning

All of the encoding tools you've seen here are also compatible with the partial_fit mechanic in scikit-learn. That means you can leverage scikit-partial to build pipelines that can handle out-of-core datasets.