Mli configuration

h2o_mli_nthreads

h2o_mli_nthreads (Number)

Default value 8

Number of threads for H2O instance for use by MLI.

mli_sample_above_for_scoring

mli_sample_above_for_scoring (Number)

Default value 1000000

When number of rows are above this limit sample for MLI for scoring UI data.

mli_sample_above_for_training

mli_sample_above_for_training (Number)

Default value 100000

When number of rows are above this limit sample for MLI for training surrogate models.

mli_interpreter_status_cache_size

mli_interpreter_status_cache_size (Number)

Default value 1000

Maximum number of interpreters status cache entries.

mli_sample_training

mli_sample_training (Boolean)

Default value True

not only sample training, but also sample scoring.

mli_strict_version_check

mli_strict_version_check (Boolean)

Default value True

Strict version check for MLI

mli_cloud_name

mli_cloud_name (String)

Default value 'H2O-MLI-DAI'

MLI cloud name

mli_ice_per_bin_strategy

mli_ice_per_bin_strategy (Boolean)

Default value False

Compute original model ICE using per feature’s bin predictions (true) or use “one frame” strategy (false).

mli_dia_default_max_cardinality

mli_dia_default_max_cardinality (Number)

Default value 10

By default DIA will run for categorical columns with cardinality <= mli_dia_default_max_cardinality.

mli_dia_default_min_cardinality

mli_dia_default_min_cardinality (Number)

Default value 2

By default DIA will run for categorical columns with cardinality >= mli_dia_default_min_cardinality.

enable_mli_keeper

enable_mli_keeper (Boolean)

Default value True

Enable MLI keeper which ensures efficient use of filesystem/memory/DB by MLI.

enable_mli_sa

enable_mli_sa (Boolean)

Default value True

Enable MLI Sensitivity Analysis

enable_mli_priority_queues

enable_mli_priority_queues (Boolean)

Default value True

Enable priority queues based explainers execution. Priority queues restrict available system resources and prevent system over-utilization. Interpretation execution time might be (significantly) slower.

mli_sequential_task_execution

mli_sequential_task_execution (Boolean)

Default value True

Explainers are run sequentially by default. This option can be used to run all explainers in parallel which can - depending on hardware strength and the number of explainers - decrease interpretation duration. Consider explainer dependencies, random explainers order and hardware over utilization.

mli_dia_sample_size

Sample size for Disparate Impact Analysis (Number)

Default value 100000

When number of rows are above this limit, then sample for Disparate Impact Analysis.

mli_pd_sample_size

Sample size for Partial Dependence Plot (Number)

Default value 25000

When number of rows are above this limit, then sample for Partial Dependence Plot.

mli_pd_numcat_num_chart

Unique feature values count driven Partial Dependence Plot binning and chart selection (Boolean)

Default value True

Use dynamic switching between Partial Dependence Plot numeric and categorical binning and UI chart selection in case of features which were used both as numeric and categorical by experiment.

mli_pd_numcat_threshold

Threshold for Partial Dependence Plot binning and chart selection (<=threshold categorical, >threshold numeric) (Number)

Default value 11

If ‘mli_pd_numcat_num_chart’ is enabled, then use numeric binning and chart if feature unique values count is bigger than threshold, else use categorical binning and chart.

new_mli_list_only_explainable_datasets

new_mli_list_only_explainable_datasets (Boolean)

Default value False

In New Interpretation screen show only datasets which can be used to explain a selected model. This can slow down the server significantly.

enable_mli_async_api

enable_mli_async_api (Boolean)

Default value True

Enable async/await-based non-blocking MLI API

enable_mli_sa_main_chart_aggregator

enable_mli_sa_main_chart_aggregator (Boolean)

Default value True

Enable main chart aggregator in Sensitivity Analysis

mli_sa_sampling_limit

Sample size for SA (Number)

Default value 500000

When to sample for Sensitivity Analysis (number of rows after sampling).

mli_sa_main_chart_aggregator_limit

mli_sa_main_chart_aggregator_limit (Number)

Default value 1000

Run main chart aggregator in Sensitivity Analysis when the number of dataset instances is bigger than given limit.

mli_predict_safe

mli_predict_safe (Boolean)

Default value False

Use predict_safe() (true) or predict_base() (false) in MLI (PD, ICE, SA, …).

mli_max_surrogate_retries

mli_max_surrogate_retries (Number)

Default value 5

Number of max retries should the surrogate model fail to build.

h2o_mli_fraction_memory

h2o_mli_fraction_memory (Float)

Default value 0.45

Fraction of memory to allocate for h2o MLI jar

excluded_mli_explainers

Exclude specific explainers by explainer ID (List)

Default value []

To exclude e.g. Sensitivity Analysis explainer use: excluded_mli_explainers=[‘h2oaicore.mli.byor.recipes.sa_explainer.SaExplainer’].

enable_ws_perfmon

enable_ws_perfmon (Boolean)

Default value False

Enable RPC API performance monitor.

mli_kernel_explainer_workers

mli_kernel_explainer_workers (Number)

Default value 4

Number of parallel workers when scoring using MOJO in Kernel Explainer.

mli_run_kernel_explainer

Use Kernel Explainer to obtain Shapley values for original features (Boolean)

Default value False

Use Kernel Explainer to obtain Shapley values for original features.

mli_kernel_explainer_sample

Sample input dataset for Kernel Explainer (Boolean)

Default value True

Sample input dataset for Kernel Explainer.

mli_kernel_explainer_sample_size

Sample size for input dataset passed to Kernel Explainer (Number)

Default value 1000

Sample size for input dataset passed to Kernel Explainer.

mli_kernel_explainer_nsamples

Number of times to re-evaluate the model when explaining each prediction with Kernel Explainer. Default is determined internally (String)

Default value 'auto'

‘auto’ or int. Number of times to re-evaluate the model when explaining each prediction. More samples lead to lower variance estimates of the SHAP values. The ‘auto’ setting uses nsamples = 2 * X.shape[1] + 2048. This setting is disabled by default and DAI determines the right number internally.

mli_kernel_explainer_l1_reg

L1 regularization for Kernel Explainer (String)

Default value 'aic'

‘num_features(int)’, ‘auto’ (default for now, but deprecated), ‘aic’, ‘bic’, or float. The l1 regularization to use for feature selection (the estimation procedure is based on a debiased lasso). The ‘auto’ option currently uses aic when less that 20% of the possible sample space is enumerated, otherwise it uses no regularization. THE BEHAVIOR OF ‘auto’ WILL CHANGE in a future version to be based on ‘num_features’ instead of AIC. The aic and bic options use the AIC and BIC rules for regularization. Using ‘num_features(int)’ selects a fix number of top features. Passing a float directly sets the alpha parameter of the sklearn.linear_model.Lasso model used for feature selection.

mli_kernel_explainer_max_runtime

Max runtime for Kernel Explainer in seconds (Number)

Default value 900

Max runtime for Kernel Explainer in seconds. Default is 900, which equates to 15 minutes. Setting this parameter to -1 means to honor the Kernel Shapley sample size provided regardless of max runtime.

mli_nlp_tokenizer

mli_nlp_tokenizer (String)

Default value 'tfidf'

Tokenizer used to extract tokens from text columns for MLI.

mli_image_enable

mli_image_enable (Boolean)

Default value True

Enable MLI for image experiments.