Monitoring data science models

The process for configuring monitoring for either a single (KServe) or a multi-model (ModelMesh) serving platform is the same.

Check the status of the trustyai-service-operator pod:

Use the Open Data Hub dashboard or the OpenShift command-line interface (CLI) to install an instance of the TrustyAI service.

When running curl commands, pass the token to the TrustyAI service using the Authorization header. For example:

Follow these steps to view cluster metrics and verify that TrustyAI is receiving data.

A "Feature and output name mapping successfully applied" message is displayed.

For information about the specific bias metrics, see Supported bias metrics.

For the complete list of TrustyAI metrics, see TrustyAI service API.

You can create a bias metric for a model by using the Open Data Hub dashboard or by using the OpenShift command-line interface (CLI).

Open Data Hub supports the following bias metrics:

For information about the specific data drift metrics, see Supported drift metrics.

For the complete list of TrustyAI metrics, see TrustyAI service API.

Use GET /metrics/{{metric}}/requests to list all currently active requests for the metric and verify the metric that you deleted is not shown. For example:

Open Data Hub supports the following data drift metrics:

For more information, see Supported explainers.

For more information, see Supported explainers.

Open Data Hub supports the following explainers:

LIME

Local Interpretable Model-agnostic Explanations (LIME) ^[1] is a saliency explanation method. LIME aims to explain a prediction 𝑝 = (𝑥, 𝑦) (an input-output pair) generated by a black-box model 𝑓 : ℝ^𝑑 → ℝ. The explanations come in the form of a "saliency" 𝑤_𝑖 attached to each feature 𝑥_𝑖 in the prediction 𝑥. LIME generates a local explanation ξ(𝑥) according to the following model:

In the original paper, 𝐺 is the class of linear models and 𝜋_𝑥 is an exponential kernel on a distance function 𝐷 (for example, cosine distance). LIME converts samples 𝑥_𝑖 from the original domain into interpretable samples as binary vectors 𝑥′_𝑖 ∈ 0,1. An encoded data set 𝐸 is built by taking nonzero elements of 𝑥′_𝑖 , recovering the original representation 𝑧 ∈ ℝ^𝑑 and then computing 𝑓(𝑧). A weighted linear model 𝑔 (with weights provided via 𝜋_𝑥) is then trained on the generated sparse data set 𝐸 and the model weights 𝑤 are used as feature weights for the final explanation ξ(𝑥).

SHAP

SHapley Additive exPlanations (SHAP), ^[2] seeks to unify several common explanation methods, notably LIME ^[1] and DeepLIFT, ^[3] under a common umbrella of additive feature attributions. These methods explain how an input 𝑥 = [𝑥₁, 𝑥₂, …​, 𝑥_𝑀] affects the output of some model 𝑓 by transforming 𝑥 ∈ ℝ^𝑀 into simplified inputs 𝑧′ ∈ 0, 1^𝑀 , such that 𝑧′_𝑖 indicates the inclusion or exclusion of feature 𝑖. The simplified inputs are then passed to an explanatory model 𝑔 that takes the following form:

In that form, each value 𝛷_𝑖 marks the contribution that feature 𝑖 had on the output model (called the attribution), 𝛷₀ marks the null output of the model; the model output when every feature is excluded. Therefore, this presents an easily interpretable explanation of the importance of each feature and a framework to permute the various input features to establish their collection contributions.

The final result of the algorithm are the Shapley values of each feature, which give an itemized "receipt" of all the contributing factors to the decision. For example, a SHAP explanation of a loan application might be as follows:

From this, the applicant can see that the biggest contributor to their denial was their home ownership status, which reduced their acceptance probability by 30 percentage points. Meanwhile, their number of children was of particular benefit, increasing their probability by 22 percentage points.

The service is built on top of two open-source projects:

The following information explains how to create an LMEvalJob custom resource (CR) to initiate an evaluation job and get the results.

Configurable global settings for LM-Eval services are stored in the TrustyAI operator global ConfigMap, named trustyai-service-operator-config. The global settings are located in the same namespace as the operator.

You can configure the following properties for LM-Eval:

After updating the settings in the ConfigMap, restart the operator to apply the new values.

To run an evaluation job, create an LMEvalJob object with the following information: model, model arguments, task, and secret.

After the LMEvalJob is created, the LM-Eval service runs the evaluation job. The status and results of the LMEvalJob object update when the information is available.

The sample LMEvalJob object contains the following features:

The following is an example of an LMEvalJob object:

After you apply the sample LMEvalJob, check its state by using the following command:

Output similar to the following appears: NAME: evaljob-sample STATE: Running

Evaluation results are available when the state of the object changes to Complete. Both the model and dataset in this example are small. The evaluation job should finish within 10 minutes on a CPU-only node.

Use the following command to get the results:

The command returns results similar to the following example:

The following table lists each property in the LMEvalJob and its usage:

Ensuring that your machine learning models are fair and unbiased is essential for building trust with your users. Although you can assess fairness during model training, it is only in deployment that your models use real-world data. Even if your models are unbiased on training data, they can exhibit serious biases in real-world scenarios. Therefore, it is crucial to monitor your models for fairness during their real-world deployment.

In this tutorial, you learn how to monitor models for bias. You will use two example models to complete the following tasks:

To set up your environment for this tutorial, complete the following tasks:

To deploy the models for this tutorial, run the following commands from the directory containing the downloaded tutorial files (odh-trustyai-demos-main).

Pass the training data through the models.

You can apply name mappings to your inputs and outputs for more meaningful field names by sending a POST request to the /info/names endpoint.

For this tutorial, run the following command from the directory containing the downloaded tutorial files (odh-trustyai-demos-main):

For general steps, see Labeling data fields.

To understand the payload structure, see the odh-trustyai-demos-main/2-BiasMonitoring/scripts/apply_name_mapping.sh file in your downloaded tutorial files.

Compute the model’s cumulative fairness up to this point.

The payload structure is as follows:

Confirm that you see outputs similar to the following examples:

The specificDefinition field is important in understanding the real-world interpretation of these metric values; you can see that both model Alpha and Beta are fair over the Is Male-Identifying field, with the two groups' rates of positive outcomes only differing by -0.3% for model Alpha and 2.8% for model Beta.

After you confirm that the models are fair over the training data, you want to ensure that they remain fair over real-world inference data. To monitor their fairness, you can schedule a metric request to compute at recurring intervals throughout deployment by passing the same payloads to the /metrics/group/fairness/spd/request endpoint.

These commands return the IDs of the created requests. Later, you can use these IDs to delete the scheduled requests.

You can monitor the average values of various data fields over time to see the average ratio of loan-payback to loan-default predictions and the average ratio of male-identifying to non-male-identifying applicants. To monitor the average values, you create an identity metric request by sending a POST request to the /metrics/identity/request endpoint.

The payload structure is as follows:

Now that you have scheduled your fairness and identify metric requests, you can simulate sending some "real world" data through your models to see if they remain fair.