detoxify

library0.5.2pypypi✓ verified 85d ago

Detoxify is a Python library for detecting toxic comments using pre-trained transformer models. It provides a simple interface to classify text as toxic, severe toxic, obscene, threat, insult, identity hate, etc. The latest version is 0.5.2, with releases approximately every few months. It requires Python >=3.7 and supports Hugging Face transformers.

ai-ml llm-agents

pip install detoxify

github ↗package ↗

INSTALL

IMPORT

SIG · DETOXIFY

detoxify

ai-mlpythonv0.5.2

Install

75.2s avg

Import

8860ms

Disk

4992MB

Pass rate

4/ 10

Env Coverage4 / 10

glibc

3.9–3.13

musl

3.9–3.13

Install & Compatibility

Where this runs

tested against v0.5.2 · pip install

no network on importno background threads

Install × environment matrix

Each cell = how many times install + import succeeded across repeated harness runs. Partial = flaky.
glibc = Debian/Ubuntu slim · musl = Alpine Linux

musl

glibc

py 3.10

✕ build_error

✓ 86.35s

py 3.11

✕ build_error

✓ 80.05s

py 3.12

✕ build_error

✓ 69.55s

py 3.13

✕ build_error

✓ 65s

py 3.9

✕ build_error

✕ timeout

4992MB installed

● package 4992MB

Code

Verified usage

Verified import paths — ran on the pinned version, not inferred.

Detoxify

✓ from detoxify import Detoxify

✗ import Detoxify

Detoxify is a class inside the module, not the module itself.

Loads the 'original' toxic comment model and predicts toxicity scores for a sample sentence.

from detoxify import Detoxify # Load the model (chooses 'original' by default) model = Detoxify('original') result = model.predict('This is a terrible, horrible example!') print(result) # Example output: {'toxicity': 0.9, 'severe_toxicity': 0.1, ...}

Debug

Known issues

gotchaThe default model 'original' may be outdated. For better performance, use 'unbiased' (small) or 'multilingual' (larger) models. See https://github.com/unitaryai/detoxify#available-models.

fix

Specify a model name explicitly: Detoxify('unbiased')

affects: >=0.1.0

gotchaThe output scores are not probabilities; they are raw logits or sigmoid outputs. Thresholds are applied by the library (score >= 0.5 is considered toxic). Do not interpret as calibrated probabilities.

fix

Use the "toxicity" score directly; the library already applies a threshold for classification if needed. Check the `predict` method's threshold parameter.

affects: all

gotchaFirst run will download the model weights (~500 MB) from Hugging Face, which may take time and require an internet connection.

fix

Pre-download models using `detoxify.download_model('original')` before offline use, or set a local cache directory.

affects: all

deprecatedThe 'original' model is based on a deprecated BERT checkpoint. It is no longer recommended for new projects; use 'unbiased' or 'multilingual' instead.

fix

Switch to Detoxify('unbiased') or Detoxify('multilingual').

affects: >=0.4.0

Errors

Common errors & fixes

ModuleNotFoundError: No module named 'detoxify'

The package is not installed or the module name is misspelled.

fix

Run 'pip install detoxify' in your environment.

OSError: Can't load the model 'original' from HuggingFace. File not found or no internet.

The model is not cached locally and there is no internet connection to download it.

fix

Ensure internet is available on first run, or pre-download the model using detoxify.download_model('original') while connected.

AttributeError: module 'detoxify' has no attribute 'Detoxify'

Incorrect import pattern: used 'import detoxify' instead of 'from detoxify import Detoxify'.

fix

Use 'from detoxify import Detoxify' to import the class.

Upgrade

Version history

0.5.2latest on PyPI · released Feb 1, 2024

Audit

Dependencies

transformersrequiredCore dependency for loading and running pre-trained models.

torchoptionalPyTorch backend for the transformer models. Optional if using TensorFlow.

tensorflowoptionalTensorFlow backend option. Not required if using PyTorch.

Agent activity

15 hits · last 30 days

node

OpenAI (training)

Resources

githubhttps://github.com/unitaryai/detoxify ↗

packagehttps://pypi.org/project/detoxify/ ↗