model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class djura.edp_im.predict.EDPIMModel(**data)[source]¶

Bases: BaseModel

hysteresis: str¶

im_type: str | None¶

method: str | None¶

backbone: BackboneModel¶

classmethod str_to_lower(values)[source]¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class djura.edp_im.predict.EDPIMInfillModel(**data)[source]¶

Bases: BaseModel

period: float¶

c_y: float¶

c_rp: float¶

mu_h: float¶

mu_s: float¶

mu_rp: float¶

mu_ult: float¶

im_type: str | None¶

classmethod str_to_lower(values)[source]¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

class djura.edp_im.predict.EDPIMIsolModel(**data)[source]¶

Bases: BaseModel

R: float¶

mu: float¶

delta_max: float¶

model_config: ClassVar[ConfigDict] = {}¶: Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].

djura.edp_im.predict.edp_im(data)[source]¶

djura.edp_im.predict.edp_im_isol(data)[source]¶

djura.edp_im.predict.edp_im_infill(data)[source]¶

djura.edp_im.predict.edp_im_batch(data)[source]¶

class djura.edp_im.XGBPredict.XGBPredict(im_type, collapse)[source]¶

Bases: object

ductility_range = array([1.000e-02, 1.100e-01, 2.100e-01, 3.100e-01, 4.100e-01, 5.100e-01, 6.100e-01, 7.100e-01, 8.100e-01, 9.100e-01, 1.010e+00, 1.110e+00, 1.210e+00, 1.310e+00, 1.410e+00, 1.510e+00, 1.610e+00, 1.710e+00, 1.810e+00, 1.910e+00, 2.010e+00, 2.110e+00, 2.210e+00, 2.310e+00, 2.410e+00, 2.510e+00, 2.610e+00, 2.710e+00, 2.810e+00, 2.910e+00, 3.010e+00, 3.110e+00, 3.210e+00, 3.310e+00, 3.410e+00, 3.510e+00, 3.610e+00, 3.710e+00, 3.810e+00, 3.910e+00, 4.010e+00, 4.110e+00, 4.210e+00, 4.310e+00, 4.410e+00, 4.510e+00, 4.610e+00, 4.710e+00, 4.810e+00, 4.910e+00, 5.010e+00, 5.110e+00, 5.210e+00, 5.310e+00, 5.410e+00, 5.510e+00, 5.610e+00, 5.710e+00, 5.810e+00, 5.910e+00, 6.010e+00, 6.110e+00, 6.210e+00, 6.310e+00, 6.410e+00, 6.510e+00, 6.610e+00, 6.710e+00, 6.810e+00, 6.910e+00, 7.010e+00, 7.110e+00, 7.210e+00, 7.310e+00, 7.410e+00, 7.510e+00, 7.610e+00, 7.710e+00, 7.810e+00, 7.910e+00, 8.010e+00, 8.110e+00, 8.210e+00, 8.310e+00, 8.410e+00, 8.510e+00, 8.610e+00, 8.710e+00, 8.810e+00, 8.910e+00, 9.010e+00, 9.110e+00, 9.210e+00, 9.310e+00, 9.410e+00, 9.510e+00, 9.610e+00, 9.710e+00, 9.810e+00, 9.910e+00, 1.001e+01, 1.011e+01, 1.021e+01, 1.031e+01, 1.041e+01, 1.051e+01, 1.061e+01, 1.071e+01, 1.081e+01, 1.091e+01, 1.101e+01, 1.111e+01, 1.121e+01, 1.131e+01, 1.141e+01, 1.151e+01, 1.161e+01, 1.171e+01, 1.181e+01, 1.191e+01])¶

FEATURE_LOW_BOUND = array([[0.01 , 0.02 , 0.02 , 2. , 0.25669332]])¶

FEATURE_UP_BOUND = array([[ 3. , 0.2 , 0.1 , 8. , 19.28312206]])¶

FEATURE_LOW_BOUND_COLLAPSE = array([[0.01, 0.02, 0.02, 2. ]])¶

FEATURE_UP_BOUND_COLLAPSE = array([[3. , 0.2, 0.1, 8. ]])¶

FEATURE_ORDER = ['period', 'damping', 'hardening_ratio', 'ductility', 'actual_ductility_end']¶

generate_sr_for_ductility(scaler, model, disp_model, period, damping, hardening_ratio, ductility)[source]¶

estimate_ductility(medians, dispersions, strength_ratio)[source]¶

property model¶

make_prediction(scaler, model, dispersions, period, damping, hardening_ratio, ductility, dynamic_ductility=None)[source]¶

Make predictions using the XGB model

Parameters:

period (float) – Period
damping (float) – Damping ratio
hardening_ratio (float) – Hardening ratio
ductility (float) – Hardening ductility of system
dynamic_ductility (ndarray) – Ductility where the strength ratio is being predicted, required for non-collapse predictions, by default None
strength_ratio (float) – Strength ratio corresponding to which a ductility value is being estimated, by default, None

Returns:

Predictions in dict type:

{"median": float,  # R, ro_2, ro_3, or ductility
 "dispersion": float}

Return type:

PredictionSchema

class djura.edp_im.scaler.MinMaxScaler(lower_bound, upper_bound)[source]¶

Bases: object

transform(data, min=0, max=1)[source]¶