DeepTrain
Industry’s first, low-code, no-code, cloud based deep learning training platform
Multiple task types
Model architectures available for Classification, Segmentation and Object Detection. New architecture support will be available with the new releases.
Multiple frameworks
Implementations available in multiple DL frameworks - Tensorflow, PyTorch.
Hyperparameter tuning
Configure hyperparameters. Detects the best combination of hyperparameters. Create multiple experiments from a tuning run.
Experiment tracking
Track and compare experiments. Visualise losses, metrics and resources. Visualise model predictions.
Export to different format
Export final model in multiple framework specific and generic formats
Transfer learning and fine tuning
Load weights from standard pre-trained models or similar previously run experiments
Developer and lite version
Switch between basic and advanced experiment configuration options
Experiment comparison and similarity
Identify previously run experiments with similar configurations
