generate_training_test_data_t: Generate training and prediction datasets for targeted...
In ginnyintifa/PTMscape:

Description Usage Arguments Details Examples

This function generates training and prediction datasets for whole proteome to predict 1/n_fold of the proteome by the other n-1 fold data.

1
2
3

generate_training_test_data_t(lower_bound = -1, upper_bound = 1,
  positive_training_feature_file, negative_training_feature_file,
  negative_predict_feature_file, output_label_training, output_label_predict)

`lower_bound`	The lower bound of the scaled data range, default to -1.
`upper_bound`	The upper bound of the scaled data range, default to 1.
`positive_training_feature_file`	A Rds file containing the positive training features(generated by g_feature_t()).
`negative_training_feature_file`	A Rds file containing the negative training features(generated by g_feature_t()).
`negative_predict_feature_file`	A Rds file containing the negative prediction features(generated by g_feature_t()).
`output_label_training`	The string to tag the output files associated with training proteins.
`output_label_predict`	The string to tag the output files associated with prediction proteins.

This function outputs formatted feature files ready for Liblinear training and prediction.

generate_training_test_data_t(lower_bound = -1, upper_bound = 1,
positive_training_feature_file ="ps_training_noc_not_na_pos_feature.Rds",
negative_training_feature_file = "ps_training_noc_not_na_candi_feature.Rds",
negative_predict_feature_file = "ps_predict_noc_not_na_candi_feature.Rds",
output_label_training = "ps_training",
output_label_predict = "ps_predict")