Crate lgbm_sys

Constants

• C_API_DTYPE_FLOAT32
• C_API_DTYPE_FLOAT64
• C_API_DTYPE_INT32
• C_API_DTYPE_INT64
• C_API_FEATURE_IMPORTANCE_GAIN
• C_API_FEATURE_IMPORTANCE_SPLIT
• C_API_MATRIX_TYPE_CSC
• C_API_MATRIX_TYPE_CSR
• C_API_PREDICT_CONTRIB
• C_API_PREDICT_LEAF_INDEX
• C_API_PREDICT_NORMAL
• C_API_PREDICT_RAW_SCORE

Functions

• LGBM_BoosterAddValidData^⚠
  Add new validation data to booster.
• LGBM_BoosterCalcNumPredict^⚠
  Get number of predictions.
• LGBM_BoosterCreate^⚠
  Create a new boosting learner.
• LGBM_BoosterCreateFromModelfile^⚠
  Load an existing booster from model file.
• LGBM_BoosterDumpModel^⚠
  Dump model to JSON.
• LGBM_BoosterFeatureImportance^⚠
  Get model feature importance.
• LGBM_BoosterFree^⚠
  Free space for booster.
• LGBM_BoosterFreePredictSparse^⚠
  Method corresponding to LGBM_BoosterPredictSparseOutput to free the allocated data.
• LGBM_BoosterGetCurrentIteration^⚠
  Get index of the current boosting iteration.
• LGBM_BoosterGetEval^⚠
  Get evaluation for training data and validation data.
• LGBM_BoosterGetEvalCounts^⚠
  Get number of evaluation metrics.
• LGBM_BoosterGetEvalNames^⚠
  Get names of evaluation metrics.
• LGBM_BoosterGetFeatureNames^⚠
  Get names of features.
• LGBM_BoosterGetLeafValue^⚠
  Get leaf value.
• LGBM_BoosterGetLinear^⚠
  Get int representing whether booster is fitting linear trees.
• LGBM_BoosterGetLoadedParam^⚠
  Get parameters as JSON string.
• LGBM_BoosterGetLowerBoundValue^⚠
  Get model lower bound value.
• LGBM_BoosterGetNumClasses^⚠
  Get number of classes.
• LGBM_BoosterGetNumFeature^⚠
  Get number of features.
• LGBM_BoosterGetNumPredict^⚠
  Get number of predictions for training data and validation data (this can be used to support customized evaluation functions).
• LGBM_BoosterGetPredict^⚠
  Get prediction for training data and validation data.
• LGBM_BoosterGetUpperBoundValue^⚠
  Get model upper bound value.
• LGBM_BoosterLoadModelFromString^⚠
  Load an existing booster from string.
• LGBM_BoosterMerge^⚠
  Merge model from other_handle into handle.
• LGBM_BoosterNumModelPerIteration^⚠
  Get number of trees per iteration.
• LGBM_BoosterNumberOfTotalModel^⚠
  Get number of weak sub-models.
• LGBM_BoosterPredictForCSC^⚠
  Make prediction for a new dataset in CSC format.
• LGBM_BoosterPredictForCSR^⚠
  Make prediction for a new dataset in CSR format.
• LGBM_BoosterPredictForCSRSingleRow^⚠
  Make prediction for a new dataset in CSR format. This method re-uses the internal predictor structure from previous calls and is optimized for single row invocation.
• LGBM_BoosterPredictForCSRSingleRowFast^⚠
  Faster variant of LGBM_BoosterPredictForCSRSingleRow.
• LGBM_BoosterPredictForCSRSingleRowFastInit^⚠
  Initialize and return a FastConfigHandle for use with LGBM_BoosterPredictForCSRSingleRowFast.
• LGBM_BoosterPredictForFile^⚠
  Make prediction for file.
• LGBM_BoosterPredictForMat^⚠
  Make prediction for a new dataset.
• LGBM_BoosterPredictForMatSingleRow^⚠
  Make prediction for a new dataset. This method re-uses the internal predictor structure from previous calls and is optimized for single row invocation.
• LGBM_BoosterPredictForMatSingleRowFast^⚠
  Faster variant of LGBM_BoosterPredictForMatSingleRow.
• LGBM_BoosterPredictForMatSingleRowFastInit^⚠
  Initialize and return a FastConfigHandle for use with LGBM_BoosterPredictForMatSingleRowFast.
• LGBM_BoosterPredictForMats^⚠
  Make prediction for a new dataset presented in a form of array of pointers to rows.
• LGBM_BoosterPredictSparseOutput^⚠
  Make sparse prediction for a new dataset in CSR or CSC format. Currently only used for feature contributions.
• LGBM_BoosterRefit^⚠
  Refit the tree model using the new data (online learning).
• LGBM_BoosterResetParameter^⚠
  Reset config for booster.
• LGBM_BoosterResetTrainingData^⚠
  Reset training data for booster.
• LGBM_BoosterRollbackOneIter^⚠
  Rollback one iteration.
• LGBM_BoosterSaveModel^⚠
  Save model into file.
• LGBM_BoosterSaveModelToString^⚠
  Save model to string.
• LGBM_BoosterSetLeafValue^⚠
  Set leaf value.
• LGBM_BoosterShuffleModels^⚠
  Shuffle models.
• LGBM_BoosterUpdateOneIter^⚠
  Update the model for one iteration.
• LGBM_BoosterUpdateOneIterCustom^⚠
  Update the model by specifying gradient and Hessian directly (this can be used to support customized loss functions).
• LGBM_BoosterValidateFeatureNames^⚠
  Check that the feature names of the data match the ones used to train the booster.
• LGBM_ByteBufferFree^⚠
  Free space for byte buffer.
• LGBM_ByteBufferGetAt^⚠
  Get a ByteBuffer value at an index.
• LGBM_DatasetAddFeaturesFrom^⚠
  Add features from source to target.
• LGBM_DatasetCreateByReference^⚠
  Allocate the space for dataset and bucket feature bins according to reference dataset.
• LGBM_DatasetCreateFromCSC^⚠
  Create a dataset from CSC format.
• LGBM_DatasetCreateFromCSR^⚠
  Create a dataset from CSR format.
• LGBM_DatasetCreateFromCSRFunc^⚠
  Create a dataset from CSR format through callbacks.
• LGBM_DatasetCreateFromFile^⚠
  Load dataset from file (like LightGBM CLI version does).
• LGBM_DatasetCreateFromMat^⚠
  Create dataset from dense matrix.
• LGBM_DatasetCreateFromMats^⚠
  Create dataset from array of dense matrices.
• LGBM_DatasetCreateFromSampledColumn^⚠
  Allocate the space for dataset and bucket feature bins according to sampled data.
• LGBM_DatasetCreateFromSerializedReference^⚠
  Allocate the space for dataset and bucket feature bins according to serialized reference dataset.
• LGBM_DatasetDumpText^⚠
  Save dataset to text file, intended for debugging use only.
• LGBM_DatasetFree^⚠
  Free space for dataset.
• LGBM_DatasetGetFeatureNames^⚠
  Get feature names of dataset.
• LGBM_DatasetGetFeatureNumBin^⚠
  Get number of bins for feature.
• LGBM_DatasetGetField^⚠
  Get info vector from dataset.
• LGBM_DatasetGetNumData^⚠
  Get number of data points.
• LGBM_DatasetGetNumFeature^⚠
  Get number of features.
• LGBM_DatasetGetSubset^⚠
  Create subset of a data.
• LGBM_DatasetInitStreaming^⚠
  Initialize the Dataset for streaming.
• LGBM_DatasetMarkFinished^⚠
  Mark the Dataset as complete by calling dataset->FinishLoad.
• LGBM_DatasetPushRows^⚠
  Push data to existing dataset, if nrow + start_row == num_total_row, will call dataset->FinishLoad.
• LGBM_DatasetPushRowsByCSR^⚠
  Push data to existing dataset, if nrow + start_row == num_total_row, will call dataset->FinishLoad.
• LGBM_DatasetPushRowsByCSRWithMetadata^⚠
  Push CSR data to existing dataset. (See LGBM_DatasetPushRowsWithMetadata for more details.)
• LGBM_DatasetPushRowsWithMetadata^⚠
  Push data to existing dataset. The general flow for a streaming scenario is:
• LGBM_DatasetSaveBinary^⚠
  Save dataset to binary file.
• LGBM_DatasetSerializeReferenceToBinary^⚠
  Create a dataset schema representation as a binary byte array (excluding data).
• LGBM_DatasetSetFeatureNames^⚠
  Save feature names to dataset.
• LGBM_DatasetSetField^⚠
  Set vector to a content in info.
• LGBM_DatasetSetWaitForManualFinish^⚠
  Set whether or not the Dataset waits for a manual MarkFinished call or calls FinishLoad on itself automatically. Set to 1 for streaming scenario, and use LGBM_DatasetMarkFinished to manually finish the Dataset.
• LGBM_DatasetUpdateParamChecking^⚠
  Raise errors for attempts to update dataset parameters.
• LGBM_DumpParamAliases^⚠
  Dump all parameter names with their aliases to JSON.
• LGBM_FastConfigFree^⚠
  Release FastConfig object.
• LGBM_GetLastError^⚠
  Get string message of the last error.
• LGBM_GetSampleCount^⚠
  Get number of samples based on parameters and total number of rows of data.
• LGBM_NetworkFree^⚠
  Finalize the network.
• LGBM_NetworkInit^⚠
  Initialize the network.
• LGBM_NetworkInitWithFunctions^⚠
  Initialize the network with external collective functions.
• LGBM_RegisterLogCallback^⚠
  Register a callback function for log redirecting.
• LGBM_SampleIndices^⚠
  Create sample indices for total number of rows.
• LGBM_SetLastError^⚠
  Set string message of the last error.

Type Definitions

• BoosterHandle
• ByteBufferHandle
• DatasetHandle
• FastConfigHandle