The NeuNetS graphical tool in IBM Watson Studio synthesizes a neural network and trains it on your training data without you having to design or build anything by hand. In the NeuNetS tool, you can view or download performance metrics, including statistics about classes and a confusion matrix showing how well the model is performing. NeuNetS is currently in beta.
Attention: In 2020, the Synthesized Neural Networks (NeuNetS) model building tool will be merged with AutoAI for a unified, automated model-building experience. Starting on December 6, 2019, the NeuNetS tool will be removed from the Watson Studio interface until the merge is complete. Please remove your NeuNets models prior to that date and migrate them to newer versions of Keras models. For details on the merge of NeuNets with AutoAI, see this blog post.
- Required service
- Watson OpenScale service
- Data format
- Textual: CSV files with labeled textual data (UTF-8 encoded and English-only)
Image: Image files in a compressed file plus a CSV file that labels the image files
- Data size
- Extremely large data sets
- How you can build models
- Create a deep learning flow to design and run experiments
Use built-in training data
Automatically test a series of algorithm and optimization options
Track, audit, and tune the model in production on a Watson OpenScale dashboard
For more information on choosing the right tool for your data and use case, see Choosing a tool
- Use built-in sample data
- Synthesize a neural network
- View performance metrics and statistics
- Deploy or download the trained model
Use built-in sample data
To try out the NeuNetS tool, you can create a model using sample data that is available in to the tool:
You do not need to download the sample data to be able to create one of the sample models, but you can explore the sample training data for your interest:
Synthesize a neural network
The NeuNetS tools synthesizes a neural network and trains it on your training data without you having to design or build anything by hand.
- If using your own training data, upload training data to cloud object storage
- Create a bucket in cloud object storage for training results
- Open the NeuNetS tool
- Follow the NeuNetS tool prompts:
- Name the model and provide a description
- Specify the training data type:
- Specify the location of your training data (if using your own data) and results bucket
View performance metrics and statistics
The NeuNetS tool divides your training data into two sets:
- A small number of samples are set aside to be used as validation data (for measuring performance)
- The rest is used to train the neural network
You can monitor performance metrics while training runs.
You can view or download training data and validation data statistics.
You can assess model performance using a confusion matrix.
See also: Confusion matrix
Deploy or download the trained model
You can deploy the trained model to your IBM Watson Machine Learning service with a single click in the NeuNetS tool.
You can also download the trained model to examine, reuse, or deploy in your own environment.
Example downloaded model