Train MNIST Model Repository
This graph trains a simple neuronal network based on the tensorflow MNIST example below and stores the model packed as a tar file in the blob repository. Additionally, the training status can be tracked via a Terminal operator.
- Training operator [TF Train MNIST Data]: The training uses a Python script that contains calls to the tensorflow API.
- Repository Model Production Pipeline: The model .tar.gz-file is produced by the training operator and first stored on blob repository.
- Tracking UI Pipeline: The training operator streams the current training status (process of the current training epoch) over its outport. This information is displayed by the Terminal operator.
- (Only if run in local mode) Operators in this graph use the following
Configure and Run the Graph
- In the left panel, select the Graphs tab and navigate to com.sap.ml/tensorflow/trainMnist.
- After the graph loads, click the Model Producer operator.
- In the properties panel, set the blobName and blobVersion fields to the name and version of the model to be saved.
- Click on the Metadata Generator operator.
- Edit the content field to set the model metadata.
- In the tool bar, select “Save Graph” (disk button) and then Run Graph (play button).
- The Status panel indicates if the graph is running.
- Open the Terminal UI to see the accuracy and model save progress.