View source: R/bt_estimate_squad.R
bt_estimate_squad | R Documentation |
Perform estimation of squad classifier
bt_estimate_squad(
detectives = NULL,
squad.level = 0.7,
train.share = 0.82,
squad.model = "XGB",
squad.robustness = 0,
detective.log = "content/0 core/Classifier statistics & history.Rdata",
squad.log = "content/0 core/Squad statistics & history.Rdata",
estimation.data = "data/classifier/training data.Rdata",
limit.classification = 50000
)
detectives |
the squad to be estimated |
squad.level |
min threshold for the squad score |
train.share |
train:test split ratio |
squad.model |
which model to use for the squad |
detective.log |
location of the log file |
squad.log |
location of the squad log file |
estimation.data |
the data to be used for testing |
limit.classification |
reduces the amount of rows of data to test to reduce comp load (my 4GB ram laptop tends to error above ~50k rows hence this being the default value) |
Several data frames about how the detective did.
Several data frames about how the detective did.
Johannes Fritz for GTA Bastiat, please estimate a new squad classifier based on the best detectives.
Johannes Fritz for GTA
www.globaltradealert.org
www.globaltradealert.org
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.