Skip to content

Trainer

Overview

The trainer section of the configuration lets you specify parameters that configure the training process, like the number of epochs or the learning rate. By default, the ECD trainer is used.

trainer:
    early_stop: 5
    learning_rate: 0.001
    epochs: 100
    batch_size: auto
    optimizer:
        type: adam
        betas:
        - 0.9
        - 0.999
        eps: 1.0e-08
        weight_decay: 0.0
        amsgrad: false
    regularization_type: l2
    use_mixed_precision: false
    compile: false
    checkpoints_per_epoch: 0
    effective_batch_size: auto
    gradient_accumulation_steps: auto
    regularization_lambda: 0.0
    enable_gradient_checkpointing: false
    validation_field: null
    validation_metric: null
    train_steps: null
    steps_per_checkpoint: 0
    max_batch_size: 1099511627776
    eval_batch_size: null
    evaluate_training_set: false
    should_shuffle: true
    increase_batch_size_on_plateau: 0
    increase_batch_size_on_plateau_patience: 5
    increase_batch_size_on_plateau_rate: 2.0
    increase_batch_size_eval_metric: loss
    increase_batch_size_eval_split: training
    gradient_clipping:
        clipglobalnorm: 0.5
        clipnorm: null
        clipvalue: null
    learning_rate_scaling: linear
    bucketing_field: null
    skip_all_evaluation: false
    enable_profiling: false
    profiler:
        wait: 1
        warmup: 1
        active: 3
        repeat: 5
        skip_first: 0
    learning_rate_scheduler:
        decay: null
        decay_rate: 0.96
        decay_steps: 10000
        staircase: false
        reduce_on_plateau: 0
        reduce_on_plateau_patience: 10
        reduce_on_plateau_rate: 0.1
        warmup_evaluations: 0
        warmup_fraction: 0.0
        reduce_eval_metric: loss
        reduce_eval_split: training
        t_0: null
        t_mult: 1
        eta_min: 0

trainer:
    type: finetune
    early_stop: 5
    learning_rate: 0.001
    epochs: 100
    batch_size: auto
    optimizer:
        type: adam
        betas:
        - 0.9
        - 0.999
        eps: 1.0e-08
        weight_decay: 0.0
        amsgrad: false
    regularization_type: l2
    use_mixed_precision: false
    compile: false
    checkpoints_per_epoch: 0
    effective_batch_size: auto
    gradient_accumulation_steps: auto
    regularization_lambda: 0.0
    enable_gradient_checkpointing: false
    validation_field: null
    validation_metric: null
    train_steps: null
    steps_per_checkpoint: 0
    max_batch_size: 1099511627776
    evaluate_training_set: false
    should_shuffle: true
    increase_batch_size_on_plateau: 0
    increase_batch_size_on_plateau_patience: 5
    increase_batch_size_on_plateau_rate: 2.0
    increase_batch_size_eval_metric: loss
    increase_batch_size_eval_split: training
    gradient_clipping:
        clipglobalnorm: 0.5
        clipnorm: null
        clipvalue: null
    learning_rate_scaling: linear
    bucketing_field: null
    skip_all_evaluation: false
    enable_profiling: false
    profiler:
        wait: 1
        warmup: 1
        active: 3
        repeat: 5
        skip_first: 0
    learning_rate_scheduler:
        decay: null
        decay_rate: 0.96
        decay_steps: 10000
        staircase: false
        reduce_on_plateau: 0
        reduce_on_plateau_patience: 10
        reduce_on_plateau_rate: 0.1
        warmup_evaluations: 0
        warmup_fraction: 0.0
        reduce_eval_metric: loss
        reduce_eval_split: training
        t_0: null
        t_mult: 1
        eta_min: 0
    eval_batch_size: 2
    base_learning_rate: 0.0

trainer:
    early_stop: 5
    learning_rate: 0.03
    max_depth: 18
    boosting_type: gbdt
    bagging_fraction: 0.8
    feature_fraction: 0.75
    extra_trees: false
    lambda_l1: 0.25
    lambda_l2: 0.2
    drop_rate: 0.1
    tree_learner: serial
    boosting_rounds_per_checkpoint: 50
    num_boost_round: 1000
    num_leaves: 82
    min_data_in_leaf: 20
    pos_bagging_fraction: 1.0
    neg_bagging_fraction: 1.0
    bagging_freq: 1
    bagging_seed: 3
    feature_fraction_bynode: 1.0
    feature_fraction_seed: 2
    extra_seed: 6
    linear_lambda: 0.0
    max_drop: 50
    skip_drop: 0.5
    uniform_drop: false
    drop_seed: 4
    validation_field: null
    validation_metric: null
    eval_batch_size: 1048576
    evaluate_training_set: false
    min_sum_hessian_in_leaf: 0.001
    max_delta_step: 0.0
    min_gain_to_split: 0.03
    xgboost_dart_mode: false
    top_rate: 0.2
    other_rate: 0.1
    min_data_per_group: 100
    max_cat_threshold: 32
    cat_l2: 10.0
    cat_smooth: 10.0
    max_cat_to_onehot: 4
    cegb_tradeoff: 1.0
    cegb_penalty_split: 0.0
    path_smooth: 0.0
    verbose: -1
    max_bin: 255
    feature_pre_filter: true
    skip_all_evaluation: false
    enable_profiling: false
    profiler:
        wait: 1
        warmup: 1
        active: 3
        repeat: 5
        skip_first: 0

Trainer parameters

  • early_stop (default: 5) : Number of consecutive rounds of evaluation without any improvement on the validation_metric that triggers training to stop. Can be set to -1, which disables early stopping entirely.
  • learning_rate (default: 0.001) : Controls how much to change the model in response to the estimated error each time the model weights are updated. If 'auto', the optimal learning rate is estimated by choosing the learning rate that produces the smallest non-diverging gradient update.
  • epochs (default: 100) : Number of epochs the algorithm is intended to be run over. Overridden if train_steps is set
  • batch_size (default: auto) : The number of training examples utilized in one training step of the model. If ’auto’, the batch size that maximized training throughput (samples / sec) will be used. For CPU training, the tuned batch size is capped at 128 as throughput benefits of large batch sizes are less noticeable without a GPU.
  • optimizer (default: {"type": "adam"}) : Optimizer type and its parameters. The optimizer is responsble for applying the gradients computed from the loss during backpropagation as updates to the model weights. See Optimizer parameters for details.
  • regularization_type (default: l2) : Type of regularization. Options: l1, l2, l1_l2, null.
  • use_mixed_precision (default: false) : Enable automatic mixed-precision (AMP) during training. Options: true, false.
  • compile (default: false) : Whether to compile the model before training. Options: true, false.
  • checkpoints_per_epoch (default: 0): Number of checkpoints per epoch. For example, 2 -> checkpoints are written every half of an epoch. Note that it is invalid to specify both non-zero steps_per_checkpoint and non-zero checkpoints_per_epoch.
  • effective_batch_size (default: auto): The effective batch size is the total number of samples used to compute a single gradient update to the model weights. This differs from batch_size by taking gradient_accumulation_steps and number of training worker processes into account. In practice, effective_batch_size = batch_size * gradient_accumulation_steps * num_workers. If 'auto', the effective batch size is derivied implicitly from batch_size, but if set explicitly, then one of batch_size or gradient_accumulation_steps must be set to something other than 'auto', and consequently will be set following the formula given above.
  • gradient_accumulation_steps (default: auto): Number of steps to accumulate gradients over before performing a weight update.
  • regularization_lambda (default: 0.0): Strength of the regularization.
  • enable_gradient_checkpointing (default: false): Whether to enable gradient checkpointing, which trades compute for memory.This is useful for training very deep models with limited memory. Options: true, false.
  • validation_field (default: null): The field for which the validation_metric is used for validation-related mechanics like early stopping, parameter change plateaus, as well as what hyperparameter optimization uses to determine the best trial. If unset (default), the first output feature is used. If explicitly specified, neither validation_field nor validation_metric are overwritten.
  • validation_metric (default: null): Metric from validation_field that is used. If validation_field is not explicitly specified, this is overwritten to be the first output feature type's default_validation_metric, consistent with validation_field. If the validation_metric is specified, then we will use the first output feature that produces this metric as the validation_field.
  • train_steps (default: null): Maximum number of training steps the algorithm is intended to be run over. Unset by default. If set, will override epochs and if left unset then epochs is used to determine training length.
  • steps_per_checkpoint (default: 0): How often the model is checkpointed. Also dictates maximum evaluation frequency. If 0 the model is checkpointed after every epoch.
  • max_batch_size (default: 1099511627776): Auto batch size tuning and increasing batch size on plateau will be capped at this value. The default value is 2^40.
  • eval_batch_size (default: null): Size of batch to pass to the model for evaluation. If it is 0 or None, the same value of batch_size is used. This is useful to speedup evaluation with a much bigger batch size than training, if enough memory is available. If ’auto’, the biggest batch size (power of 2) that can fit in memory will be used.
  • evaluate_training_set (default: false): Whether to evaluate on the entire training set during evaluation. By default, training metrics will be computed at the end of each training step, and accumulated up to the evaluation phase. In practice, computing training set metrics during training is up to 30% faster than running a separate evaluation pass over the training set, but results in more noisy training metrics, particularly during the earlier epochs. It's recommended to only set this to True if you need very exact training set metrics, and are willing to pay a significant performance penalty for them. Options: true, false.
  • should_shuffle (default: true): Whether to shuffle batches during training when true. Options: true, false.
  • increase_batch_size_on_plateau (default: 0): The number of times to increase the batch size on a plateau.
  • increase_batch_size_on_plateau_patience (default: 5): How many epochs to wait for before increasing the batch size.
  • increase_batch_size_on_plateau_rate (default: 2.0): Rate at which the batch size increases.
  • increase_batch_size_eval_metric (default: loss): Which metric to listen on for increasing the batch size.
  • increase_batch_size_eval_split (default: training): Which dataset split to listen on for increasing the batch size.
  • gradient_clipping : Parameter values for gradient clipping.
  • gradient_clipping.clipglobalnorm (default: 0.5): Maximum allowed norm of the gradients
  • gradient_clipping.clipnorm (default: null): Maximum allowed norm of the gradients
  • gradient_clipping.clipvalue (default: null): Maximum allowed value of the gradients
  • learning_rate_scaling (default: linear): Scale by which to increase the learning rate as the number of distributed workers increases. Traditionally the learning rate is scaled linearly with the number of workers to reflect the proportion by which the effective batch size is increased. For very large batch sizes, a softer square-root scale can sometimes lead to better model performance. If the learning rate is hand-tuned for a given number of workers, setting this value to constant can be used to disable scale-up. Options: constant, sqrt, linear.
  • bucketing_field (default: null): Feature to use for bucketing datapoints
  • skip_all_evaluation (default: false): Whether to skip evaluation entirely. If you are training a model with a well-known configuration on a well-known dataset and are confident about the expected results, you might skip all evaluation. Moreover, evaluating a model, especially on large validation or test sets, can be time-consuming. Options: true, false.
  • enable_profiling (default: false): Whether to enable profiling of the training process using torch.profiler.profile. Options: true, false.
  • profiler : Parameter values for profiling config.
  • profiler.wait (default: 1): The number of steps to wait profiling.
  • profiler.warmup (default: 1): The number of steps for profiler warmup after waiting finishes.
  • profiler.active (default: 3): The number of steps that are actively recorded. Values more than 10 wil dramatically slow down tensorboard loading.
  • profiler.repeat (default: 5): The optional number of profiling cycles. Use 0 to profile the entire training run.
  • profiler.skip_first (default: 0): The number of steps to skip in the beginning of training.
  • learning_rate_scheduler : Parameter values for learning rate scheduler.
  • learning_rate_scheduler.decay (default: null) : Turn on decay of the learning rate. Options: linear, exponential, cosine, null.
  • learning_rate_scheduler.decay_rate (default: 0.96): Decay per epoch (%): Factor to decrease the Learning rate.
  • learning_rate_scheduler.decay_steps (default: 10000): The number of steps to take in the exponential learning rate decay.
  • learning_rate_scheduler.staircase (default: false): Decays the learning rate at discrete intervals. Options: true, false.
  • learning_rate_scheduler.reduce_on_plateau (default: 0) : How many times to reduce the learning rate when the algorithm hits a plateau (i.e. the performance on the training set does not improve)
  • learning_rate_scheduler.reduce_on_plateau_patience (default: 10): How many evaluation steps have to pass before the learning rate reduces when reduce_on_plateau > 0.
  • learning_rate_scheduler.reduce_on_plateau_rate (default: 0.1): Rate at which we reduce the learning rate when reduce_on_plateau > 0.
  • learning_rate_scheduler.warmup_evaluations (default: 0): Number of evaluation steps to warmup the learning rate for.
  • learning_rate_scheduler.warmup_fraction (default: 0.0): Fraction of total training steps to warmup the learning rate for.
  • learning_rate_scheduler.reduce_eval_metric (default: loss): Metric plateau used to trigger when we reduce the learning rate when reduce_on_plateau > 0.
  • learning_rate_scheduler.reduce_eval_split (default: training): Which dataset split to listen on for reducing the learning rate when reduce_on_plateau > 0.
  • learning_rate_scheduler.t_0 (default: null): Number of steps before the first restart for cosine annealing decay. If not specified, it will be set to steps_per_checkpoint.
  • learning_rate_scheduler.t_mult (default: 1): Period multiplier after each restart for cosine annealing decay. Defaults to 1, i.e., restart every t_0 steps. If set to a larger value, the period between restarts increases by that multiplier. For e.g., if t_mult is 2, then the periods would be: t_0, 2t_0, 2^2t_0, 2^3*t_0, etc.
  • learning_rate_scheduler.eta_min (default: 0): Minimum learning rate allowed for cosine annealing decay. Default: 0.
  • type (default: finetune): Options: finetune.
  • early_stop (default: 5) : Number of consecutive rounds of evaluation without any improvement on the validation_metric that triggers training to stop. Can be set to -1, which disables early stopping entirely.
  • learning_rate (default: 0.001) : Controls how much to change the model in response to the estimated error each time the model weights are updated. If 'auto', the optimal learning rate is estimated by choosing the learning rate that produces the smallest non-diverging gradient update.
  • epochs (default: 100) : Number of epochs the algorithm is intended to be run over. Overridden if train_steps is set
  • batch_size (default: auto) : The number of training examples utilized in one training step of the model. If ’auto’, the batch size that maximized training throughput (samples / sec) will be used. For CPU training, the tuned batch size is capped at 128 as throughput benefits of large batch sizes are less noticeable without a GPU.
  • optimizer (default: {"type": "adam"}) : Optimizer type and its parameters. The optimizer is responsble for applying the gradients computed from the loss during backpropagation as updates to the model weights. See Optimizer parameters for details.
  • regularization_type (default: l2) : Type of regularization. Options: l1, l2, l1_l2, null.
  • use_mixed_precision (default: false) : Enable automatic mixed-precision (AMP) during training. Options: true, false.
  • compile (default: false) : Whether to compile the model before training. Options: true, false.
  • checkpoints_per_epoch (default: 0): Number of checkpoints per epoch. For example, 2 -> checkpoints are written every half of an epoch. Note that it is invalid to specify both non-zero steps_per_checkpoint and non-zero checkpoints_per_epoch.
  • effective_batch_size (default: auto): The effective batch size is the total number of samples used to compute a single gradient update to the model weights. This differs from batch_size by taking gradient_accumulation_steps and number of training worker processes into account. In practice, effective_batch_size = batch_size * gradient_accumulation_steps * num_workers. If 'auto', the effective batch size is derivied implicitly from batch_size, but if set explicitly, then one of batch_size or gradient_accumulation_steps must be set to something other than 'auto', and consequently will be set following the formula given above.
  • gradient_accumulation_steps (default: auto): Number of steps to accumulate gradients over before performing a weight update.
  • regularization_lambda (default: 0.0): Strength of the regularization.
  • enable_gradient_checkpointing (default: false): Whether to enable gradient checkpointing, which trades compute for memory.This is useful for training very deep models with limited memory. Options: true, false.
  • validation_field (default: null): The field for which the validation_metric is used for validation-related mechanics like early stopping, parameter change plateaus, as well as what hyperparameter optimization uses to determine the best trial. If unset (default), the first output feature is used. If explicitly specified, neither validation_field nor validation_metric are overwritten.
  • validation_metric (default: null): Metric from validation_field that is used. If validation_field is not explicitly specified, this is overwritten to be the first output feature type's default_validation_metric, consistent with validation_field. If the validation_metric is specified, then we will use the first output feature that produces this metric as the validation_field.
  • train_steps (default: null): Maximum number of training steps the algorithm is intended to be run over. Unset by default. If set, will override epochs and if left unset then epochs is used to determine training length.
  • steps_per_checkpoint (default: 0): How often the model is checkpointed. Also dictates maximum evaluation frequency. If 0 the model is checkpointed after every epoch.
  • max_batch_size (default: 1099511627776): Auto batch size tuning and increasing batch size on plateau will be capped at this value. The default value is 2^40.
  • evaluate_training_set (default: false): Whether to evaluate on the entire training set during evaluation. By default, training metrics will be computed at the end of each training step, and accumulated up to the evaluation phase. In practice, computing training set metrics during training is up to 30% faster than running a separate evaluation pass over the training set, but results in more noisy training metrics, particularly during the earlier epochs. It's recommended to only set this to True if you need very exact training set metrics, and are willing to pay a significant performance penalty for them. Options: true, false.
  • should_shuffle (default: true): Whether to shuffle batches during training when true. Options: true, false.
  • increase_batch_size_on_plateau (default: 0): The number of times to increase the batch size on a plateau.
  • increase_batch_size_on_plateau_patience (default: 5): How many epochs to wait for before increasing the batch size.
  • increase_batch_size_on_plateau_rate (default: 2.0): Rate at which the batch size increases.
  • increase_batch_size_eval_metric (default: loss): Which metric to listen on for increasing the batch size.
  • increase_batch_size_eval_split (default: training): Which dataset split to listen on for increasing the batch size.
  • gradient_clipping : Parameter values for gradient clipping.
  • gradient_clipping.clipglobalnorm (default: 0.5): Maximum allowed norm of the gradients
  • gradient_clipping.clipnorm (default: null): Maximum allowed norm of the gradients
  • gradient_clipping.clipvalue (default: null): Maximum allowed value of the gradients
  • learning_rate_scaling (default: linear): Scale by which to increase the learning rate as the number of distributed workers increases. Traditionally the learning rate is scaled linearly with the number of workers to reflect the proportion by which the effective batch size is increased. For very large batch sizes, a softer square-root scale can sometimes lead to better model performance. If the learning rate is hand-tuned for a given number of workers, setting this value to constant can be used to disable scale-up. Options: constant, sqrt, linear.
  • bucketing_field (default: null): Feature to use for bucketing datapoints
  • skip_all_evaluation (default: false): Whether to skip evaluation entirely. If you are training a model with a well-known configuration on a well-known dataset and are confident about the expected results, you might skip all evaluation. Moreover, evaluating a model, especially on large validation or test sets, can be time-consuming. Options: true, false.
  • enable_profiling (default: false): Whether to enable profiling of the training process using torch.profiler.profile. Options: true, false.
  • profiler : Parameter values for profiling config.
  • profiler.wait (default: 1): The number of steps to wait profiling.
  • profiler.warmup (default: 1): The number of steps for profiler warmup after waiting finishes.
  • profiler.active (default: 3): The number of steps that are actively recorded. Values more than 10 wil dramatically slow down tensorboard loading.
  • profiler.repeat (default: 5): The optional number of profiling cycles. Use 0 to profile the entire training run.
  • profiler.skip_first (default: 0): The number of steps to skip in the beginning of training.
  • learning_rate_scheduler : Parameter values for learning rate scheduler.
  • learning_rate_scheduler.decay (default: null) : Turn on decay of the learning rate. Options: linear, exponential, cosine, null.
  • learning_rate_scheduler.decay_rate (default: 0.96): Decay per epoch (%): Factor to decrease the Learning rate.
  • learning_rate_scheduler.decay_steps (default: 10000): The number of steps to take in the exponential learning rate decay.
  • learning_rate_scheduler.staircase (default: false): Decays the learning rate at discrete intervals. Options: true, false.
  • learning_rate_scheduler.reduce_on_plateau (default: 0) : How many times to reduce the learning rate when the algorithm hits a plateau (i.e. the performance on the training set does not improve)
  • learning_rate_scheduler.reduce_on_plateau_patience (default: 10): How many evaluation steps have to pass before the learning rate reduces when reduce_on_plateau > 0.
  • learning_rate_scheduler.reduce_on_plateau_rate (default: 0.1): Rate at which we reduce the learning rate when reduce_on_plateau > 0.
  • learning_rate_scheduler.warmup_evaluations (default: 0): Number of evaluation steps to warmup the learning rate for.
  • learning_rate_scheduler.warmup_fraction (default: 0.0): Fraction of total training steps to warmup the learning rate for.
  • learning_rate_scheduler.reduce_eval_metric (default: loss): Metric plateau used to trigger when we reduce the learning rate when reduce_on_plateau > 0.
  • learning_rate_scheduler.reduce_eval_split (default: training): Which dataset split to listen on for reducing the learning rate when reduce_on_plateau > 0.
  • learning_rate_scheduler.t_0 (default: null): Number of steps before the first restart for cosine annealing decay. If not specified, it will be set to steps_per_checkpoint.
  • learning_rate_scheduler.t_mult (default: 1): Period multiplier after each restart for cosine annealing decay. Defaults to 1, i.e., restart every t_0 steps. If set to a larger value, the period between restarts increases by that multiplier. For e.g., if t_mult is 2, then the periods would be: t_0, 2t_0, 2^2t_0, 2^3*t_0, etc.
  • learning_rate_scheduler.eta_min (default: 0): Minimum learning rate allowed for cosine annealing decay. Default: 0.
  • eval_batch_size (default: 2): Batch size used for evaluation in the LLM trainer.
  • base_learning_rate (default: 0.0): Base learning rate used for training in the LLM trainer.

See the LightGBM documentation for more details about the available parameters.

  • early_stop (default: 5) : Number of consecutive rounds of evaluation without any improvement on the validation_metric that triggers training to stop. Can be set to -1, which disables early stopping entirely.
  • learning_rate (default: 0.03) : Controls how much to change the model in response to the estimated error each time the model weights are updated.
  • max_depth (default: 18) : Maximum depth of a tree in the GBM trainer. A negative value means no limit.
  • boosting_type (default: gbdt) : Type of boosting algorithm to use with GBM trainer. Options: gbdt, dart.
  • bagging_fraction (default: 0.8) : Fraction of data to use for bagging with GBM trainer.
  • feature_fraction (default: 0.75) : Fraction of features to use in the GBM trainer.
  • extra_trees (default: false) : Whether to use extremely randomized trees in the GBM trainer. Options: true, false.
  • lambda_l1 (default: 0.25) : L1 regularization factor for the GBM trainer.
  • lambda_l2 (default: 0.2) : L2 regularization factor for the GBM trainer.
  • drop_rate (default: 0.1): Dropout rate for the GBM trainer. Used only with boosting_type 'dart'.
  • tree_learner (default: serial): Type of tree learner to use with GBM trainer. Options: serial, feature, data, voting.
  • boosting_rounds_per_checkpoint (default: 50): Number of boosting rounds per checkpoint / evaluation round.
  • num_boost_round (default: 1000): Number of boosting rounds to perform with GBM trainer.
  • num_leaves (default: 82): Number of leaves to use in the tree with GBM trainer.
  • min_data_in_leaf (default: 20): Minimum number of data points in a leaf with GBM trainer.
  • pos_bagging_fraction (default: 1.0): Fraction of positive data to use for bagging with GBM trainer.
  • neg_bagging_fraction (default: 1.0): Fraction of negative data to use for bagging with GBM trainer.
  • bagging_freq (default: 1): Frequency of bagging with GBM trainer.
  • bagging_seed (default: 3): Random seed for bagging with GBM trainer.
  • feature_fraction_bynode (default: 1.0): Fraction of features to use for each tree node with GBM trainer.
  • feature_fraction_seed (default: 2): Random seed for feature fraction with GBM trainer.
  • extra_seed (default: 6): Random seed for extremely randomized trees in the GBM trainer.
  • linear_lambda (default: 0.0): Linear tree regularization in the GBM trainer.
  • max_drop (default: 50): Maximum number of dropped trees during one boosting iteration. Used only with boosting_type 'dart'. A negative value means no limit.
  • skip_drop (default: 0.5): Probability of skipping the dropout during one boosting iteration. Used only with boosting_type 'dart'.
  • uniform_drop (default: false): Whether to use uniform dropout in the GBM trainer. Used only with boosting_type 'dart'. Options: true, false.
  • drop_seed (default: 4): Random seed to choose dropping models in the GBM trainer. Used only with boosting_type 'dart'.
  • validation_field (default: null): First output feature, by default it is set as the same field of the first output feature.
  • validation_metric (default: null): Metric used on validation_field, set by default to the output feature type's default_validation_metric.
  • eval_batch_size (default: 1048576): Size of batch to pass to the model for evaluation.
  • evaluate_training_set (default: false): Whether to evaluate on the entire training set during evaluation. By default, training metrics will be computed at the end of each training step, and accumulated up to the evaluation phase. In practice, computing training set metrics during training is up to 30% faster than running a separate evaluation pass over the training set, but results in more noisy training metrics, particularly during the earlier epochs. It's recommended to only set this to True if you need very exact training set metrics, and are willing to pay a significant performance penalty for them. Options: true, false.
  • min_sum_hessian_in_leaf (default: 0.001): Minimum sum of hessians in a leaf with GBM trainer.
  • max_delta_step (default: 0.0): Used to limit the max output of tree leaves in the GBM trainer. A negative value means no constraint.
  • min_gain_to_split (default: 0.03): Minimum gain to split a leaf in the GBM trainer.
  • xgboost_dart_mode (default: false): Whether to use xgboost dart mode in the GBM trainer. Used only with boosting_type 'dart'. Options: true, false.
  • top_rate (default: 0.2): The retain ratio of large gradient data in the GBM trainer. Used only with boosting_type 'goss'.
  • other_rate (default: 0.1): The retain ratio of small gradient data in the GBM trainer. Used only with boosting_type 'goss'.
  • min_data_per_group (default: 100): Minimum number of data points per categorical group for the GBM trainer.
  • max_cat_threshold (default: 32): Number of split points considered for categorical features for the GBM trainer.
  • cat_l2 (default: 10.0): L2 regularization factor for categorical split in the GBM trainer.
  • cat_smooth (default: 10.0): Smoothing factor for categorical split in the GBM trainer.
  • max_cat_to_onehot (default: 4): Maximum categorical cardinality required before one-hot encoding in the GBM trainer.
  • cegb_tradeoff (default: 1.0): Cost-effective gradient boosting multiplier for all penalties in the GBM trainer.
  • cegb_penalty_split (default: 0.0): Cost-effective gradient boosting penalty for splitting a node in the GBM trainer.
  • path_smooth (default: 0.0): Smoothing factor applied to tree nodes in the GBM trainer.
  • verbose (default: -1): Verbosity level for GBM trainer. Options: -1, 0, 1, 2.
  • max_bin (default: 255): Maximum number of bins to use for discretizing features with GBM trainer.
  • feature_pre_filter (default: true): Whether to ignore features that are unsplittable based on min_data_in_leaf in the GBM trainer. Options: true, false.
  • skip_all_evaluation (default: false): Whether to skip evaluation entirely. If you are training a model with a well-known configuration on a well-known dataset and are confident about the expected results, you might skip all evaluation. Moreover, evaluating a model, especially on large validation or test sets, can be time-consuming. Options: true, false.
  • enable_profiling (default: false): Whether to enable profiling of the training process using torch.profiler.profile. Options: true, false.
  • profiler : Parameter values for profiling config.
  • profiler.wait (default: 1): The number of steps to wait profiling.
  • profiler.warmup (default: 1): The number of steps for profiler warmup after waiting finishes.
  • profiler.active (default: 3): The number of steps that are actively recorded. Values more than 10 wil dramatically slow down tensorboard loading.
  • profiler.repeat (default: 5): The optional number of profiling cycles. Use 0 to profile the entire training run.
  • profiler.skip_first (default: 0): The number of steps to skip in the beginning of training.

Optimizer parameters

The available optimizers wrap the ones available in PyTorch. For details about the parameters that can be used to configure different optimizers, please refer to the PyTorch documentation.

The learning_rate parameter used by the optimizer comes from the trainer section. Other optimizer specific parameters, shown with their Ludwig default settings, follow:

sgd

optimizer:
    type: sgd
    momentum: 0.0
    weight_decay: 0.0
    dampening: 0.0
    nesterov: false
  • momentum (default: 0.0): Momentum factor.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • dampening (default: 0.0): Dampening for momentum.
  • nesterov (default: false): Enables Nesterov momentum. Options: true, false.

sgd_8bit

optimizer:
    type: sgd_8bit
    momentum: 0.0
    weight_decay: 0.0
    dampening: 0.0
    nesterov: false
    block_wise: false
    percentile_clipping: 100
  • momentum (default: 0.0): Momentum factor.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • dampening (default: 0.0): Dampening for momentum.
  • nesterov (default: false): Enables Nesterov momentum. Options: true, false.
  • block_wise (default: false): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

lbfgs

optimizer:
    type: lbfgs
    max_iter: 20
    max_eval: null
    tolerance_grad: 1.0e-07
    tolerance_change: 1.0e-09
    history_size: 100
    line_search_fn: null
  • max_iter (default: 20): Maximum number of iterations per optimization step.
  • max_eval (default: null): Maximum number of function evaluations per optimization step. Default: max_iter * 1.25.
  • tolerance_grad (default: 1e-07): Termination tolerance on first order optimality.
  • tolerance_change (default: 1e-09): Termination tolerance on function value/parameter changes.
  • history_size (default: 100): Update history size.
  • line_search_fn (default: null): Line search function to use. Options: strong_wolfe, null.

adam

optimizer:
    type: adam
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.

adam_8bit

optimizer:
    type: adam_8bit
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    block_wise: true
    percentile_clipping: 100
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

paged_adam

optimizer:
    type: paged_adam
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    block_wise: true
    percentile_clipping: 100
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

paged_adam_8bit

optimizer:
    type: paged_adam_8bit
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    block_wise: true
    percentile_clipping: 100
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

adamw

optimizer:
    type: adamw
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.

adamw_8bit

optimizer:
    type: adamw_8bit
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    block_wise: true
    percentile_clipping: 100
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

paged_adamw

optimizer:
    type: paged_adamw
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    block_wise: true
    percentile_clipping: 100
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

paged_adamw_8bit

optimizer:
    type: paged_adamw_8bit
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    block_wise: true
    percentile_clipping: 100
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

adadelta

optimizer:
    type: adadelta
    rho: 0.9
    eps: 1.0e-06
    weight_decay: 0.0
  • rho (default: 0.9): Coefficient used for computing a running average of squared gradients.
  • eps (default: 1e-06): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).

adagrad

optimizer:
    type: adagrad
    initial_accumulator_value: 0
    lr_decay: 0
    weight_decay: 0
    eps: 1.0e-10
  • initial_accumulator_value (default: 0):
  • lr_decay (default: 0): Learning rate decay.
  • weight_decay (default: 0): Weight decay ($L2$ penalty).
  • eps (default: 1e-10): Term added to the denominator to improve numerical stability.

adagrad_8bit

optimizer:
    type: adagrad_8bit
    initial_accumulator_value: 0
    lr_decay: 0
    weight_decay: 0
    eps: 1.0e-10
    block_wise: true
    percentile_clipping: 100
  • initial_accumulator_value (default: 0):
  • lr_decay (default: 0): Learning rate decay.
  • weight_decay (default: 0): Weight decay ($L2$ penalty).
  • eps (default: 1e-10): Term added to the denominator to improve numerical stability.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

adamax

optimizer:
    type: adamax
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).

nadam

optimizer:
    type: nadam
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    momentum_decay: 0.004
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • momentum_decay (default: 0.004): Momentum decay.

rmsprop

optimizer:
    type: rmsprop
    momentum: 0.0
    alpha: 0.99
    eps: 1.0e-08
    centered: false
    weight_decay: 0.0
  • momentum (default: 0.0): Momentum factor.
  • alpha (default: 0.99): Smoothing constant.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • centered (default: false): If True, computes the centered RMSProp, and the gradient is normalized by an estimation of its variance. Options: true, false.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).

rmsprop_8bit

optimizer:
    type: rmsprop_8bit
    momentum: 0.0
    alpha: 0.99
    eps: 1.0e-08
    centered: false
    weight_decay: 0.0
    block_wise: true
    percentile_clipping: 100
  • momentum (default: 0.0): Momentum factor.
  • alpha (default: 0.99): Smoothing constant.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • centered (default: false): If True, computes the centered RMSProp, and the gradient is normalized by an estimation of its variance. Options: true, false.
  • weight_decay (default: 0.0): Weight decay ($L2$ penalty).
  • block_wise (default: true): Whether to use block wise update. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.

lamb

optimizer:
    type: lamb
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    bias_correction: true
    adam_w_mode: true
    percentile_clipping: 100
    block_wise: false
    max_unorm: 1.0
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • bias_correction (default: true): Options: true, false.
  • adam_w_mode (default: true): Whether to use the AdamW mode of this algorithm from the paper 'Decoupled Weight Decay Regularization'. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.
  • block_wise (default: false): Whether to use block wise update. Options: true, false.
  • max_unorm (default: 1.0):

lamb_8bit

optimizer:
    type: lamb_8bit
    betas:
    - 0.9
    - 0.999
    eps: 1.0e-08
    weight_decay: 0.0
    amsgrad: false
    bias_correction: true
    adam_w_mode: true
    percentile_clipping: 100
    block_wise: false
    max_unorm: 1.0
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • eps (default: 1e-08): Term added to the denominator to improve numerical stability.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • amsgrad (default: false): Whether to use the AMSGrad variant of this algorithm from the paper 'On the Convergence of Adam and Beyond'. Options: true, false.
  • bias_correction (default: true): Options: true, false.
  • adam_w_mode (default: true): Whether to use the AdamW mode of this algorithm from the paper 'Decoupled Weight Decay Regularization'. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.
  • block_wise (default: false): Whether to use block wise update. Options: true, false.
  • max_unorm (default: 1.0):

lars

optimizer:
    type: lars
    momentum: 0.9
    dampening: 0.0
    weight_decay: 0.0
    nesterov: false
    percentile_clipping: 100
    max_unorm: 1.0
  • momentum (default: 0.9): Momentum factor.
  • dampening (default: 0.0): Dampening for momentum.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • nesterov (default: false): Enables Nesterov momentum. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.
  • max_unorm (default: 1.0):

lars_8bit

optimizer:
    type: lars_8bit
    momentum: 0.9
    dampening: 0.0
    weight_decay: 0.0
    nesterov: false
    percentile_clipping: 100
    max_unorm: 1.0
  • momentum (default: 0.9): Momentum factor.
  • dampening (default: 0.0): Dampening for momentum.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • nesterov (default: false): Enables Nesterov momentum. Options: true, false.
  • percentile_clipping (default: 100): Percentile clipping.
  • max_unorm (default: 1.0):

lion

optimizer:
    type: lion
    betas:
    - 0.9
    - 0.999
    weight_decay: 0.0
    percentile_clipping: 100
    block_wise: true
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • percentile_clipping (default: 100): Percentile clipping.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.

lion_8bit

optimizer:
    type: lion_8bit
    betas:
    - 0.9
    - 0.999
    weight_decay: 0.0
    percentile_clipping: 100
    block_wise: true
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • percentile_clipping (default: 100): Percentile clipping.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.

paged_lion

optimizer:
    type: paged_lion
    betas:
    - 0.9
    - 0.999
    weight_decay: 0.0
    percentile_clipping: 100
    block_wise: true
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • percentile_clipping (default: 100): Percentile clipping.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.

paged_lion_8bit

optimizer:
    type: paged_lion_8bit
    betas:
    - 0.9
    - 0.999
    weight_decay: 0.0
    percentile_clipping: 100
    block_wise: true
  • betas (default: [0.9, 0.999]): Coefficients used for computing running averages of gradient and its square.
  • weight_decay (default: 0.0): Weight decay (L2 penalty).
  • percentile_clipping (default: 100): Percentile clipping.
  • block_wise (default: true): Whether to use block wise update. Options: true, false.

Note

Gradient clipping is also configurable, through optimizers, with the following parameters:

clip_global_norm: 0.5
clipnorm: null
clip_value: null

No optimizer parameters are available for the LightGBM trainer.

Training length

The length of the training process is configured by:

  • epochs (default: 100): One epoch is one pass through the entire dataset. By default, epochs is 100 which means that the training process will run for a maximum of 100 epochs before terminating.
  • train_steps (default: None): The maximum number of steps to train for, using one mini-batch per step. By default this is unset, and epochs will be used to determine training length.
  • num_boost_round (default: 100): The number of boosting iterations. By default, num_boost_round is 100 which means that the training process will run for a maximum of 100 boosting iterations before terminating.

Tip

In general, it's a good idea to set up a long training runway, relying on early stopping criteria (early_stop) to stop training when there hasn't been any improvement for a long time.

Early stopping

Machine learning models, when trained for too long, are often prone to overfitting. It's generally a good policy to set up some early stopping criteria as it's not useful to have a model train after it's maximized what it can learn, as to retain it's ability to generalize to new data.

How early stopping works in Ludwig

By default, Ludwig sets trainer.early_stop=5, which means that if there have been 5 consecutive rounds of evaluation where there hasn't been any improvement on the validation subset, then training will terminate.

Ludwig runs evaluation once per checkpoint, which by default is once per epoch. Checkpoint frequency can be configured using checkpoints_per_epoch (default: 1) or steps_per_checkpoint (default: 0, disabled). See this section for more details.

Changing the metric early stopping metrics

The metric that dictates early stopping is trainer.validation_field and trainer.validation_metric. By default, early stopping uses the combined loss on the validation subset.

trainer:
    validation_field: combined
    validation_metric: loss

However, this can be configured to use other metrics. For example, if we had an output feature called recommended, then we can configure early stopping on the output feature accuracy like so:

trainer:
    validation_field: recommended
    validation_metric: accuracy

Disabling early stopping

trainer.early_stop can be set to -1, which disables early stopping entirely.

Checkpoint-evaluation frequency

Evaluation is run every time the model is checkpointed.

By default, checkpoint-evaluation will occur once every epoch.

The frequency of checkpoint-evaluation can be configured using:

  • steps_per_checkpoint (default: 0): every n training steps
  • checkpoints_per_epoch (default: 0): n times per epoch

Note

It is invalid to specify both non-zero steps_per_checkpoint and non-zero checkpoints_per_epoch.

Tip

Running evaluation once per epoch is an appropriate fit for small datasets that fit in memory and train quickly. However, this can be a poor fit for unstructured datasets, which tend to be much larger, and train more slowly due to larger models.

Running evaluation too frequently can be wasteful while running evaluation not frequently enough can be uninformative. In large-scale training runs, it's common for evaluation to be configured to run on a sub-epoch time scale, or every few thousand steps.

We recommend configuring evaluation such that new evaluation results are available at least several times an hour. In general, it is not necessary for models to train over the entirety of a dataset, nor evaluate over the entirety of a test set, to produce useful monitoring metrics and signals to indicate model performance.

Increasing throughput on GPUs

Increase batch size

trainer:
    batch_size: auto

Users training on GPUs can often increase training throughput by increasing the batch_size so that more examples are computed every training step. Set batch_size to auto to use the largest batch size that can fit in memory.

Use mixed precision

trainer:
    use_mixed_precision: true

Speeds up training by using float16 parameters where it makes sense. Mixed precision training on GPU can dramatically speedup training, with some risks to model convergence. In practice, it works particularly well when fine-tuning a pretrained model like a HuggingFace transformer. See blog here for more details.