This is a fast implementation of FastXML, PFastXML, and PFastreXML based on the following papers:

• "FastXML: A Fast, Accurate and Stable Tree-classifier for eXtreme Multi-label Learning" Paper
• "Extreme Multi-label Loss Functions for Recommendation, Tagging, Ranking & Other Missing Label Application" Paper
• "DiSMEC - Distributed Sparse Machines for Extreme Multi-label Classification" Paper Code

DiSMEC makes it's appearance via an L2 penalty rather than an L1 which, when set with a high alpha and sparsity eps of 0.01-0.05, also can produce sparse linear classifiers.

It's implemented in the quasi-familiar scikit-learn clf format.

• Version 2.0 is not backward compatible with 1.x
• User model.save(path) to save models instead of cPickle
• Rewrites data storage layer
• Uses 50% the memory, loads 30% faster, and is 40% faster to inference

This repo provides a simple script along with the library, fxml.py, which allows easy train / testing of simple datasets.

It takes two formats: a simple JSON format and the standard extreme multi label dataset format.

As an example, to train a standalone classifier against the Delicious-200K dataset:

fxml.py delicious.model deliciousLarge_train.txt --standard-dataset --verbose train --iters 5 --trees 20 --label-weight propensity --alpha 1e-4 --leaf-classifiers --no-remap-labels

To test:

fxml.py delicious.model deliciousLarge_test.txt --standard-dataset inference

As fxml.py is intended as an easy to understand example for setting up a FastXML classifier, the JSON format is very simple. It is newline delimited format.

train.json:

{"title": "red dresses", "tags": ["clothing", "women", "dresses"]}
{"title": "yellow dresses for sweet 16", "tags": ["yellow", "summer dresses", "occasionwear"]}
...

It can then be trained:

fxml.py my_json.model train.json --verbose train --iters 5 --trees 20 --label-weight propensity --alpha 1e-4 --leaf-classifiers

Not the omission of the flags "--standard-dataset" and "--no-remap-labels". Since the tags/classes provided are strings, fxml.py will remap them to an integer label space for training. During inference, it will map the label index back

from fastxml import Trainer, Inferencer

X = [Sparse or numpy arrays]
y = [[1, 3]] # Currently requires list[list[int]]

trainer = Trainer(n_trees=32, n_jobs=-1)

trainer.fit(X, y)

trainer.save(path)

clf = Inferencer(path)

clf.predict(X)
# or
clf.predict(X, fmt='dict')

#############
# PFastXML
#############

from fastxml.weights import propensity

weights = propensity(y)
trainer.fit(X, y, weights)

###############
# PFastreXML
###############
trainer = Trainer(n_trees=32, n_jobs=-1, leaf_classifiers=True)
trainer.fit(X, y, weights)

1. Run all the standard benchmark datasets against it.

2. Refactor. Most of the effort has been spent on speed and it needs to be cleaned up.