kumoai.trainer.OptimizationPlan#

class kumoai.trainer.OptimizationPlan[source]#

Configuration parameters that define how columns are encoded in the training and batch prediction pipelines. Please see the Kumo documentation for more information.

Variables:

  • max_epochs – (int) The maximum number of epochs to train a model for (default: run_mode-dependent). Supported Task Types: All

  • min_steps_per_epoch – (int) The minimum number of steps to be included in an epoch; one step corresponds to one forward pass of a mini-batch (default: 30). Supported Task Types: All

  • max_steps_per_epoch – (int) The maximum number of steps to be included in an epoch; one step corresponds to one forward pass of a mini-batch (default: run_mode-dependent). Supported Task Types: All

  • max_val_steps – (int) The maximum number of steps to be included in a validation pass; one step corresponds to one forward pass of a mini-batch (default: run_mode-dependent). Supported Task Types: All

  • max_test_steps – (int) The maximum number of steps to be included in a test pass; one step corresponds to one forward pass of a mini-batch (default: run_mode-dependent). Supported Task Types: All

  • loss – (list[str]) The loss type to use in the model optimizer (default: task_type-dependent). Supported Task Types: All

  • base_lr – (list[float]) The base learning rate (pre-decay) to be used in the model optimizer. (default: [1e-4, 5e-4, 1e-3, 5e-3, 1e-2, 5e-2]). Supported Task Types: All

  • weight_decay – (list[float]) A list of potential weight decay options in the model optimizer. (default: [0.0, 5e-8, 5e-7, 5e-6]). Supported Task Types: All

  • batch_size – (list[int]) The number of examples to be included in one mini-batch. (default: [512, 1024]). Supported Task Types: All

  • early_stopping – (list[EarlyStoppingConfig]) A list of potential early stopping strategies EarlyStoppingConfig for model optimization (default: [{min_delta=0.0, patience=3}]). Supported Task Types: All

  • lr_scheduler – (list[LRSchedulerConfig]) A list of potential learning rate schedulers LRSchedulerConfig for model optimization (default: [ {name="cosine_with_warmup_restarts", interval="step"}, {name="constant_with_warmup", interval="step"}, {name="linear_with_warmup", interval="step"}, {name="csoine_with_warmup", interval="step"}]). Supported Task Types: All

  • majority_sampling_ratio – (list[float | None]) A ratio to specify how examples are smapled from the majority class (default: [None]). Supported Task Types: Binary Classification

  • weight_mode – (list[float | None]) If majority_sampling_ratio is given, this option specifies how to weigh majority vs. minority classes during training (default: ["sample"]). Supported Task Types: Binary Classification