title: Generate Metrics and Compare Models in SAP AI Core

description: Explore different ways of logging metrics during training and compare generated models.

- You have an understanding of using data and generating models in SAP AI Core, from [this tutorial](https://developers.sap.com/tutorials/ai-core-data.html/#)

In this tutorial, you will use SAP AI Launchpad to compare two models that have been generated using SAP AI Core. This tutorial builds on the previous tutorials on house price prediction and ingesting data.

from ai_core_sdk.tracking import Tracking

aic_connection = Tracking(base_url='') # DO NOT Change, the value is intentionally passed as empty string, When this code will run inside SAP AI Core then the values will be auto-populated

[ACCORDION-BEGIN [Step 3: ](Add basic metric logging)]

After execution, this is shown in SAP AI Launchpad. You can zoom in for details.

Add the following snippet to store metrics for step information. This snippet is also useful for tracking the metrics on epochs of the training process.

Add the following snippet to store metrics for step information.

The parameter `value="housepricemodel"` refers to the artifact name, which references the model that will be stored in AWS S3. The name of this parameter must match the name that you defined in your YAML workflow.

After execution, this is shown in SAP AI Launchpad.

[ACCORDION-BEGIN [Step 6: ](Custom metrics for model inspection)]

After execution, this is shown in SAP AI Launchpad.

[ACCORDION-BEGIN [Step 6: ](Add tags for execution meta after training)]

After execution, this is shown in SAP AI Launchpad.

[ACCORDION-BEGIN [Step 7: ](Complete files)]

Use the following commands to build your Docker image and push the contents to the cloud.

Paste and edit, then execute the following snippet:

- How to store model accuracy data using Postman or Python in SAP AI Core

In SAP AI Core, you use an **Execution** to train an ML model and even do batch inferencing on data. In this tutorial, you will learn to store the performance indicators for a model after the execution has been completed. The steps demonstrate how to store any type of indicator, such as accuracy, mean-squared-loss, or your own custom information about the execution.

The steps demonstrated in the tutorial, send requests to store metrics about an execution from your local system to SAP AI Core. Therefore, these steps need to be performed after the execution is completed.

Install `SAP AI Core SDK` to use APIs available for SAP AI Core functions.

Replace with your SAP AI Core credentials and execute. This creates a client to make API calls to SAP AI Core.

List all executions for your **resource group**.

Alternatively, you can view the tags using SAP AI Launchpad.

A basic metric has a **name**, **value** and **timestamp**.

The code creates an object of class **Metric** with metric name of `Accuracy`. Then passes this object to `metric` parameter of metric of the `.metrics.modify` function to update metrics information in your SAP AI Core instance.

You may either supply a list of multiple metrics in a single function call or in separate function calls. In either case, the update will append over existing metrics information.

Alternatively, check the metric data using SAP AI Launchpad. The **Timestamp** is always converted to the local time of the user.

You may call the function `ai_core_client.metrics.modify` multiple times after each training epoch, each call will append new metrics to previous information.

Any custom information structure stored as **Character Large Object (CLOB)** and hence must be converted to string `str` type while passing as argument to function call. Here you will store and retrieve a snippet of [classification report](https://scikit-learn.org/stable/modules/generated/sklearn.metrics.classification_report.html) for your model.

Alternatively, you can view the metric data using SAP AI Launchpad.

This step is similar to storing custom metric information. Here you will generate a demo visualization, save the visualization locally in **Portable Network Graphics (PNG)** format, which you will load (as bytes) and encode to base64 then convert to string to store it SAP AI Core.

[ACCORDION-BEGIN [Step 2: ](Upload model files to AWS S3)]

> Note that depending on your region, your AWS endpoint syntax may differ from the example above. In the event of an error, try this step again with alternative syntax. For available syntaxes, please see the [AWS documentation](https://docs.aws.amazon.com/AmazonS3/latest/userguide/WebsiteEndpoints.html)