# shenweichen/DeepCTR-Torch

【PyTorch】Easy-to-use,Modular and Extendible package of deep-learning based CTR models.

Repository: https://github.com/shenweichen/DeepCTR-Torch
Canonical: https://ross.abutalabs.com/products/deepctr-torch
Homepage: https://deepctr-torch.readthedocs.io/en/latest/index.html
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deepctr, ctr-models, deep-learning, deepctr-pytorch, deepfm, fibinet, xdeepfm, torchrec, deeprec
Last push: 2026-07-09T15:07:55+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 91, release rhythm 48, longevity 100
- inputs: {"age_days": 2553, "days_push": 55, "days_rel": 137, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3450, forks 735 (observed 2026-08-28T04:08:05.248869+00:00)

## What it is
DeepCTR-Torch is a PyTorch library providing easy-to-use, modular, and extendable implementations of deep-learning-based CTR (click-through rate) prediction models such as DeepFM, xDeepFM, and Wide & Deep. It exposes a scikit-learn-like fit/predict API and reusable component layers for building custom models.

## Use cases
- train a DeepFM model for click-through rate prediction
- build a custom CTR model in PyTorch from modular layers
- compare deep learning CTR models like xDeepFM and FiBiNET
- predict user ad click probability from multi-field categorical data
- prototype deep learning recommendation ranking models

## When to choose
- you need ready-made implementations of many published CTR models in PyTorch
- you want a simple fit/predict workflow for CTR experiments
- you want modular layers to compose your own custom CTR architecture

## When to avoid
- you need a full production recommender system rather than model implementations
- you prefer TensorFlow, where the original DeepCTR may fit better
- your task is not click-through-rate or user response prediction

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning
- domain: machine-learning, deep-learning, data-science
- platform: python
- tags: ctr, click-through-rate-prediction, recommender-systems, pytorch, deepfm, xdeepfm, torchrec

## Member repositories
- shenweichen/DeepCTR-Torch (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:05.248869+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:37:35.319088+00:00, confidence not recorded.
  - readme: https://github.com/shenweichen/DeepCTR-Torch (fetched 2026-08-28T04:08:05.248869+00:00, sha abbc5b5a27d1)
  - registry_pypi: https://pypi.org/pypi/deepctr-torch/json (fetched 2026-08-29T09:32:01.753675+00:00, sha 2bb193c162d6)
- Data as of 2026-08-30T08:39:29.467469+00:00.