For AI agents: a documentation index is available at the root level at /llms.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
LogoLogoDocumentation
DiscordContact SupportStart for free
GuideExamplesDocsFAQReleases
GuideExamplesDocsFAQReleases
  • Predictive Query
    • Overview
  • Python SDK
    • kumoai
    • kumoai.connector
    • kumoai.graph
    • kumoai.pquery
    • kumoai.encoder
    • kumoai.trainer
    • kumoai.utils
    • kumoai.rfm
  • Model Plan
    • Column Processing
      • refit_full
      • refit_trainval
      • metrics
      • num_experiments
      • tune_metric
  • Overview
  • ASSUMING
  • CLASSIFY/RANK TOP K
  • FOR EACH
  • PREDICT
  • WHERE
  • AVG
  • COUNT
  • COUNT_DISTINCT
  • FIRST
  • LAST
  • LIST_DISTINCT
  • MAX
  • MIN
  • SUM
  • !=
  • <
  • <=
  • =
  • >
  • >=
  • AND
  • CONTAINS
  • ENDS WITH
  • IN
  • IS NOT NULL
  • IS NULL
  • LIKE
  • NOT LIKE
  • NOT CONTAINS
  • NOT
  • OR
  • STARTS WITH
  • kumoai
  • kumoai.connector
  • kumoai.graph
  • kumoai.pquery
  • kumoai.encoder
  • kumoai.trainer
  • kumoai.utils
  • kumoai.rfm
  • Column Processing
  • GNN architecture
  • aggregation
  • channels
  • Graph Transformer architecture
  • channels
  • num_layers
  • num_heads
  • dropout
  • positional_encodings
  • Selecting Link Prediction Model Architectures on Kumo
  • activation
  • distance_measure
  • handle_new_target_entities
  • module
  • normalization
  • num_post_message_passing_layers
  • num_pre_message_passing_layers
  • ranking_embedding_loss_coeff
  • output_embedding_dim
  • prediction_time_encodings
  • target_embedding_mode
  • use_seq_id
  • num_neighbors
  • sample_from_entity_table
  • base_lr
  • batch_size
  • early_stopping
  • lr_scheduler
  • loss
  • majority_sampling_ratio
  • max_epochs
  • max_steps_per_epoch
  • max_test_steps
  • max_val_steps
  • weight_decay
  • weight_mode
  • refit_full
  • refit_trainval
  • metrics
  • num_experiments
  • tune_metric
  • end_time
  • forecast_length
  • forecast_type
  • lag_timesteps
  • year_over_year
  • weight_col
  • split
  • start_time
  • timeframe_step
  • train_end_offset
  • train_start_offset
DiscordContact SupportStart for free
Model PlanTraining Job Plan

refit_full

||View as Markdown|
Something went wrong!
Previous

weight_mode

Next

refit_trainval

NVIDIANVIDIA
Developer-friendly docs for your API
Privacy Policy | Your Privacy Choices | Terms of Service | Accessibility | Corporate Policies | Product Security | Contact

Copyright © 2026, NVIDIA Corporation.