exalate-issue-sync[bot]
commented
1 year ago Arno Candel commented: offset only applies to regression and GLM binomial.
For DL regression, it's a per-observation bias term outside of the non-linearity: For 1 hidden layer, it's y = tanh(Ax+b) + o. It's the same as subtracting offset from the response. It's a convenience function.
For DL classification, I don't know how offset would be defined. Offset for which class??
DinukaH2O
parse attached file with first column as enum run -
buildModel 'deeplearning', {"model_id":"deeplearning-f274f771-9708-4248-9170-2aff509f4fc8","training_frame":"Key_Frame__hdf.hex","ignored_columns":[],"ignore_const_cols":true,"response_column":"Incident","offset_column":"ofset","activation":"Tanh","hidden":[20],"epochs":"10000","variable_importances":false,"balance_classes":false,"checkpoint":"","use_all_factor_levels":true,"train_samples_per_iteration":-2,"adaptive_rate":true,"input_dropout_ratio":0,"l1":0,"l2":0,"loss":"Automatic","score_interval":5,"score_training_samples":10000,"score_duty_cycle":0.1,"autoencoder":false,"overwrite_with_best_model":true,"target_ratio_comm_to_comp":0.02,"seed":-5298722591438265000,"rho":0.99,"epsilon":1e-8,"max_w2":"Infinity","initial_weight_distribution":"UniformAdaptive","classification_stop":0,"diagnostics":true,"fast_mode":true,"force_load_balance":true,"single_node_mode":false,"shuffle_training_data":false,"missing_values_handling":"MeanImputation","quiet_mode":false,"sparse":false,"col_major":false,"average_activation":0,"sparsity_beta":0,"max_categorical_features":2147483647,"reproducible":false,"export_weights_and_biases":false}
get - Error calling POST /3/ModelBuilders/deeplearning with opts {"model_id":"deeplearning-f274f771-9708-4248-9170-2aff509f4fc8","training_frame":"Key_Frame__hdf.hex","ignored_columns":null,"ignore_const_cols":true,"response_column":"Incident","offset_column":"ofset","activation":"Tanh","hidden":"[20]","epochs":"10000","variable_importances":false,"balance_classes":false,"checkpoint":"","use_all_factor_levels":true,"train_samples_per_iteration":-2,"adaptive_rate":true,"input_dropout_ratio":0,"l1":0,"l2":0,"loss":"Automatic","score_interval":5,"score_training_samples":10000,"score_duty_cycle":0.1,"autoencoder":false,"overwrite_with_best_model":true,"target_ratio_comm_to_comp":0.02,"seed":-5298722591438265000,"rho":0.99,"epsilon":1e-8,"max_w2":"Infinity","initial_weight_distribution":"UniformAdaptive","classification_stop":0,"diagnostics":true,"fast_mode":true,"force_load_balance":true,"single_node_mode":false,"shuffle_training_data":false,"missing_values_handling":"MeanImputation","quiet_mode":false,"sparse":false,"col_major":false,"average_activation":0,"sparsity_beta":0,"max_categorical_features":2147483647,"reproducible":false,"export_weights_and_biases":false}
Illegal argument(s) for deeplearning model: deeplearning-f274f771-9708-4248-9170-2aff509f4fc8. Details: ERROR on field: _offset: Offset only applies to regression and logistic regression.