# qubvel/efficientnet

Implementation of EfficientNet model. Keras and TensorFlow Keras.

Repository: https://github.com/qubvel/efficientnet
Canonical: https://ross.abutalabs.com/products/efficientnet
Homepage: https://arxiv.org/abs/1905.11946
Language: Python
License: Apache-2.0
License Family: permissive
Topics: classification, imagenet, efficientnet, pretrained-models, image-classification, efficient, nasnetmobile, mobilenet, deep-learning
Last push: 2024-01-24T14:27:34+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2652, "days_push": 952, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2100, forks 457 (observed 2026-08-28T04:06:13.579842+00:00)

## What it is
A Keras and TensorFlow Keras reimplementation of the EfficientNet convolutional neural network family (B0-B7), including ImageNet-pretrained weights. It enables lightweight, state-of-the-art image classification and transfer learning.

## Use cases
- classify images with a pretrained EfficientNet model in Keras
- load EfficientNet weights for transfer learning on my own dataset
- use EfficientNet-B0 to B7 with tensorflow.keras
- build an image classifier with fewer parameters than ResNet
- fine-tune EfficientNet on ImageNet-style data
- run efficient CNN inference on CPU

## When to choose
- you work in Keras or TensorFlow Keras and need EfficientNet architectures with pretrained weights
- you want high-accuracy image classification with a lightweight model
- you need a simple pip-installable model zoo for transfer learning

## When to avoid
- you use PyTorch or JAX instead of Keras/TensorFlow
- you need the latest architectures like EfficientNetV2 or ConvNeXt
- you need a project under active development with frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, image-processing, computer-vision
- domain: deep-learning, computer-vision, image-processing, machine-learning
- platform: python
- tags: efficientnet, keras, tensorflow, pretrained-models, imagenet, image-classification, transfer-learning, convolutional-neural-networks

## Member repositories
- qubvel/efficientnet (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:13.579842+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 2026-08-30T02:54:21.104617+00:00, confidence not recorded.
  - readme: https://github.com/qubvel/efficientnet (fetched 2026-08-28T04:06:13.579842+00:00, sha 372d540be4bb)
  - homepage: https://arxiv.org/abs/1905.11946 (fetched 2026-08-29T10:34:47.061754+00:00, sha c7b0709df1df)
  - site_page: https://info.arxiv.org/about/donate.html (fetched 2026-08-29T10:34:47.071628+00:00, sha cca9c3a11c56)
  - site_page: https://info.arxiv.org/about/ourmembers.html (fetched 2026-08-29T10:34:47.075619+00:00, sha 47cbc55ff1de)
  - site_page: https://info.arxiv.org/about (fetched 2026-08-29T10:34:47.077498+00:00, sha a1f16f915a9a)
  - registry_pypi: https://pypi.org/pypi/efficientnet/json (fetched 2026-08-29T10:34:47.079050+00:00, sha 0e85ec5b2fe6)
  - site_page: https://info.arxiv.org/labs/index.html (fetched 2026-08-29T10:34:47.073864+00:00, sha b14a8d05a0ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.