## Feature considerations {: #feature-considerations }

Consider the following when working with Scoring Code:

* Using Scoring Code in production requires additional development efforts to implement model management and model monitoring, which the DataRobot API provides out of the box.

* Exportable Java Scoring Code requires extra RAM during model building. As a result, to use this feature, you should keep your training dataset under 8GB. Projects larger than 8GB may fail due to memory issues. If you get an out-of-memory error, decrease the sample size and try again. The memory requirement _does not apply during model scoring_. During scoring, the only limitation on the dataset is the RAM of the machine on which the Scoring Code is run.

### Model support {: #model-support }

* Scoring Code is available for models containing only _supported_ built-in tasks. It is not available for [custom models](custom-inf-model) or models containing one or more [custom tasks](cml-custom-tasks).

* Scoring Code is not supported in multilabel projects.

*  Keras models do not support Scoring Code by default; however, support can be enabled by having an administrator activate the Enable Scoring Code Support for Keras Models feature flag. If enabled, note that these models are not compatible with Scoring Code for Android and Snowflake.

Additional instances in which Scoring Code generation is not available include:

* Naive Bayes

* Text tokenization involving the MeCab tokenizer

* Visual AI and Location AI

### Time series support {: #time-series-support }

* The following time series capabilities are not supported for Scoring Code:

    * Row-based / irregular data
    * Nowcasting (single forecast point)
    * Intramonth seasonality
    * Time series blenders
    * Autoexpansion
    * EWMA (Exponentially Weighted Moving Average)

* Scoring Code is not supported in time series binary classification projects.

* Scoring Code is not typically supported in time series anomaly detection models; however, it is supported for IsolationForest and some XGBoost-based anomaly detection model blueprints. For a list of supported time series blueprints, see the [Time series blueprints with Scoring Code support](#ts-sc-blueprint-support) note.

{% include 'includes/scoring-code-consider-ts.md' %}

### Prediction Explanations support {: #prediction-explanations-support }

Consider the following when working with Prediction Explanations for Scoring Code:

* To download Prediction Explanations with Scoring Code, you _must_ select **Include Prediction Explanations** during [Leaderboard download](sc-download-leaderboard#leaderboard-download) or [Deployment download](sc-download-deployment#deployment-download). This option is _not_ available for [Legacy download](sc-download-legacy).

* Scoring Code _doesn't_ support Prediction Explanations for time series models.

* Scoring Code _only_ supports [XEMP-based](xemp-pe) prediction explanations. [SHAP-based](shap-pe) prediction explanations aren't supported.
