LifeWatch ERIC Metadata Catalogue
elapid_object.ela
This a serialized python object resulting of the execution of Maxent in case user needs it for later. It represents an instance of the class MaxentModel used in Elapid package. This package is the main library used in Maxent application.
Default
Identification
- Alternate Identifier
- fc8868f9-aec5-4188-b0c2-a277c5ef59a9
- Publication Date
- 2023-06-28
- Title
- elapid_object.ela
- Abstract
- This a serialized python object resulting of the execution of Maxent in case user needs it for later. It represents an instance of the class MaxentModel used in Elapid package. This package is the main library used in Maxent application.
- Dataset Language
- eng
- Dataset Creator
- Dataset Contact
- Keywords ()
-
- Serialized python object format
- Elapid object
- Maxent Model
License Information
- Intellectual Rights
Resource License
- License Name
Distribution
No information provided.
Projects
Projects
• Project
No information provided.
Methods
Methods
• Method
No information provided.
Data Tables
Data Tables
• Data Table
- Name
- elapid_object
Physicaldocument
Data Format
Externally Defined Format
- Format Name
- ela
Attribute List
Attribute
- Name
- feature_types
- Label
- Definition
- Maxent feature types to fit. Must be in string "lqphta" or list ["linear", "quadratic", "product", "hinge", "threshold", "auto"].
Attribute
- Name
- tau
- Label
- Definition
- Maxent prevalence value for scaling logistic output transform: maxent model transformation type. Select from ["raw", "logistic", "cloglog"].
Attribute
- Name
- clamp
- Label
- Definition
- Set features to min/max range from training during prediction.
Attribute
- Name
- scorer
- Label
- Definition
- Sklearn scoring function for model training.
Attribute
- Name
- beta_multiplier
- Label
- Definition
- Scaler for all regularization parameters. Higher values drop more coeffiecients.
Attribute
- Name
- beta_lqp
- Label
- Definition
- Linear, quadratic and product feature regularization scaler.
Attribute
- Name
- beta_hinge
- Label
- Definition
- Hinge feature regularization scaler.
Attribute
- Name
- beta_threshold
- Label
- Definition
- Threshold feature regularization scaler.
Attribute
- Name
- beta_categorical
- Label
- Definition
- Categorical feature regularization scaler.
Attribute
- Name
- n_hinge_features
- Label
- Definition
- The number of hinge features to fit in feature transformation.
Attribute
- Name
- n_threshold_features
- Label
- Definition
- The number of thresholds to fit in feature transformation.
Attribute
- Name
- convergence_tolerance
- Label
- Definition
- Model convergence tolerance level.
Attribute
- Name
- use_lambdas
- Label
- Definition
- Guide for which model lambdas to select (either "best" or "last").
Attribute
- Name
- n_lambdas
- Label
- Definition
- Number of lamba values to fit models with.
Attribute
- Name
- class_weights
- Label
- Definition
- Strategy for weighting presence samples. Pass "balanced" to compute the ratio based on sample frequency or pass a float for the presence:background weight ratio R maxnet package uses a value of 100 as default. Set to None to ignore.
Attribute
- Name
- n_cpus
- Label
- Definition
- Threads to use during model training.
Attribute
- Name
- use_sklearn
- Label
- Definition
- Force using "sklearn" for fitting logistic regression. Turned off by default to use "glmnet" for fitting. This feature was turned on to support Windows users install the package without a fortran compiler.
