multi-class-text-classification-cnn-rnn
Classify San Francisco crime descriptions into 39 categories using a hybrid TextCNN and GRU model.
Classify text datasets with a TensorFlow 2.x Keras CNN and GRU architecture.
This open-source tool implements a hybrid TextCNN and GRU model using TensorFlow to classify text into multiple categories. It suits data scientists and machine learning engineers looking to experiment with deep learning text architectures.
What it is
It trains a neural network combining convolutional layers and gated recurrent units on word embeddings to map text descriptions to categorical labels. The pipeline handles data preprocessing, model training, artifact exportation, and batch prediction out of the box.
Who it's for
It suits machine learning practitioners working on multi-class text classification tasks who want a reference implementation using TensorFlow Keras. It does not suit non-technical users or those looking for a zero-code web application.
Key features
Hybrid TextCNN and GRU
Combines convolutional and recurrent neural network layers for sentence classification.
Configurable Training
Accepts external JSON configuration files to manage hyperparameters and sequence lengths.
Artifact Export
Writes out saved models, vocabulary indices, class labels, and trained parameters automatically.
Up and running in 3 steps
Our honest verdict
What's great
- Combines CNN and GRU architectures for sequence classification
- Includes bundled sample data and training configurations
- Exports structured model artifacts and vocabulary mappings
- Provides dedicated scripts for both training and inference
Where it falls short
- Requires manual Python virtual environment and dependency installation
- Command-line interface only with no graphical user interface
- Running deep learning training requires a suitable compute environment
Pricing & license
Free under the Apache-2.0 licence. There are no paid hosted options mentioned in the repository, so the only cost is running it yourself on your own hardware.
Open remote AI-training roles
870 open remote AI-training roles, updated November 14, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
What framework does the model use?
The model is built using TensorFlow 2.x Keras.
Where are trained model artifacts saved?
Artifacts are written to the trained_results directory, including the saved model and vocabulary mappings.
