The Text Clustering repository contains tools to easily embed and cluster texts as well as label clusters semantically. This repository is a work in progress and serves as a minimal codebase that can be modified and adapted to other use cases.

The pipeline consists of several distinct blocks that can be customized and the whole pipeline can run in a few minutes on a consumer laptop. Each block uses existing standard methods and works quite robustly.

Install the following libraries to get started:

pip install scikit-learn umap-learn sentence_transformers faiss-cpu plotly matplotlib datasets

Clone this repository and navigate to the folder:

git clone https://github.com/huggingface/text-clustering.git cd text-clustering

Run pipeline and visualize results:

from src.text_clustering import ClusterClassifier from datasets import load_dataset SAMPLE = 100_000 texts = load_dataset("HuggingFaceTB/cosmopedia-100k", split="train").select(range(SAMPLE))["text"] cc = ClusterClassifier(embed_device="mps") # run the pipeline: embs, labels, summaries = cc.fit(texts) # show the results cc.show() # save  cc.save("./cc_100k")

Load classifier and run inference:

from src.text_clustering import ClusterClassifier cc = ClusterClassifier(embed_device="mps") # load state cc.load("./cc_100k") # visualize cc.show() # classify new texts with k-nearest neighbour search cluster_labels, embeddings = cc.infer(some_texts, top_k=1)

If you want to reproduce the color scheme in the plot above you can add the following code before you run cc.show():

from cycler import cycler import matplotlib.pyplot as plt default_cycler = (cycler(color=[ "0F0A0A", "FF6600", "FFBE00", "496767", "87A19E", "FF9200", "0F3538", "F8E08E", "0F2021", "FAFAF0"]) ) plt.rc('axes', prop_cycle=default_cycler)

If you would like to customize the plotting further the easiest way is to customize or overwrite the _show_mpl and _show_plotly methods.

You can also run the pipeline using a script with:

# run a new pipeline python run_pipeline.py --mode run --save_load_path './cc_100k' --n_samples 100000 --build_hf_ds # load existing pipeline python run_pipeline.py --mode load --save_load_path './cc_100k' --build_hf_ds # inference mode on new texts from an input dataset python run_pipeline.py --mode infer --save_load_path './cc_100k' --n_samples <NB_INFERENCE_SAMPLES> --input_dataset <HF_DATA_FOR_INFERENCE>

The build_hf_ds flag builds and pushes HF datasets, for the files and clusters, that can be directly used in the FW visualization space. In infer mode, we push the clusters dataset by default.

You can also change how the clusters are labeled (multiple topics (default) vs single topic with an educational score) using the flag --topic_mode.

Check the examples folder for an example of clustering and topic labeling applied to the AutoMathText dataset, utilizing Cosmopedia's web labeling approach.