# emedvedev/attention-ocr

A Tensorflow model for text recognition (CNN + seq2seq with visual attention) available as a Python package and compatible with Google Cloud ML Engine.

Repository: https://github.com/emedvedev/attention-ocr
Canonical: https://ross.abutalabs.com/products/emedvedev-attention-ocr
Language: Python
License: MIT
License Family: permissive
Topics: tensorflow, ocr, ocr-recognition, machine-learning, ml, cnn, seq2seq, google-cloud-ml, google-cloud, image-recognition, hacktoberfest
Last push: 2023-10-20T17:48:54+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": 3330, "days_push": 1048, "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 1086, forks 248 (observed 2026-08-28T04:03:32.173267+00:00)

## What it is
A TensorFlow-based OCR model using a CNN plus seq2seq architecture with visual attention for text recognition in images. It ships as a Python package with a CLI for building datasets, training, testing, and exporting trained models.

## Use cases
- recognize text in images with a deep learning OCR model
- train a custom OCR model on my own labeled image dataset
- export a trained OCR model as a SavedModel or frozen graph
- run OCR training on Google Cloud ML Engine
- visualize attention weights during OCR testing

## When to choose
- you need trainable, customizable text recognition rather than off-the-shelf OCR
- you want a CNN + attention seq2seq model in TensorFlow 1.x
- you plan to train on your own image/label pairs and export for deployment

## When to avoid
- you need TensorFlow 2.x support
- you want a ready-to-use pretrained OCR without training
- you need actively maintained software with recent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: ocr, machine-learning, image-processing, deep-learning
- domain: computer-vision, machine-learning, image-processing
- platform: python, cloud
- tags: tensorflow, seq2seq, attention, cnn, text-recognition, google-cloud-ml, cli, linux, macos

## Member repositories
- emedvedev/attention-ocr (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:32.173267+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 2026-08-30T06:49:31.631496+00:00, confidence not recorded.
  - readme: https://github.com/emedvedev/attention-ocr (fetched 2026-08-28T04:03:32.173267+00:00, sha 4c0c4be2055c)
- Data as of 2026-08-30T08:39:29.467469+00:00.