OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Fixy

Fixy is an open-source Turkish writing-support and checking project that uses deep learning for spelling correction and semantic analysis to identify and correct context-sensitive writing errors. Its documented focus includes the separate or joined spelling of -da/-de, -ki and -mi, with model weights, training notebooks and a related DeepChecker Python package.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Fixy — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Fixy when investigating or building Turkish writing-check workflows, especially for context-sensitive separate-versus-joined suffix errors. The repository is also relevant when you want to inspect the project's models, training notebooks and data.

Who it is for

  • Developers exploring Turkish spelling correction and NLP.
  • Researchers or practitioners interested in context-sensitive correction of Turkish suffix and conjunction forms.
  • Python users who want to try the documented DeepChecker functions or train a corrector using repository materials.

Documented uses

  • Correct documented Turkish -de/-da, -ki and -mi writing cases through DeepChecker.
  • Check whether one of those forms should be written separately or attached.
  • Load provided model weights or use training notebooks and datasets as a starting point for custom corrector work.
Before you choose

Requirements & limits

The project is described as open source and is published under the MIT license. The README lists Python 3.6+ as a requirement.

  • The documented DeepChecker functions target -de/-da, -ki and -mi rather than presenting a general-purpose editor interface.
  • The README describes a two-layer bidirectional LSTM as the selected model for its semantic spelling task and names subtitle, Wikipedia and newspaper corpora in its data approach.
  • The repository provides separate routes for using pre-trained weights and for training with notebooks and data.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Prepare the Fixy repository for model work

Clone the project repository and identify its documented model and training materials.

Before you start

  • Python 3.6 or later

Steps

  1. Clone https://github.com/Fixy-TR/fixy.git.

  2. Use the repository's pre_trained_weights for the documented direct model-loading route, or consult its training notebooks and data for a custom training workflow.

Expected outcome

A local copy of the repository with the documented next paths for loading weights or working with training materials.

Guide limitations

  • This setup guidance does not provide a complete environment installation or a full end-to-end run procedure.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
Fixy-TR/fixy · Official homepage title · retained in its source language
How we review sourcesSuggest a correction