Modeling GuidesΒΆ
The following documents describe the various modeling features provided by the MLTK.
For complete tutorials, please see the Tutorials.
| Name | Description |
|---|---|
| Model Profiler | Provides information about how efficiently a model may run on an embedded target |
| Model Profiler Utility | Standalone model profiler executable with webpage interface |
| Model Visualizer | View an ML model in an interactive webpage |
| Model Specification | Specify everything needed to create, train, and evaluate a machine learning model |
| Model Training | Train a machine learning model using the MLTK and Google Tensorflow |
| Model Training via SSH | Train a machine learning model on a cloud server via SSH |
| Model Training Monitor with Tensorboard | Monitor/profile the training of a model using Tensorboard |
| Model Evaluation | Evaluate the accuracy of a trained model |
| Model Quantization | Reduce memory usage by compressing model weights |
| Model Parameters | Embed custom parameters into the generated model file |
| Model Summary | Generate a text summary of a model |
| Model Archive | Store model information in a distributable archive |
| Model Search Path | Configure how models are found by the MLTK commands and APIs |