Skip to content
Snippets Groups Projects
Commit 9805e543 authored by kohlhaasrebecca's avatar kohlhaasrebecca
Browse files

[Fix] Remove parameters MetaModel, Model, ExpDesign from PostProcessing 

PostProcessing uses engine.* instead of self.* for MetaModel, Model and ExpDesign.
In addition it calls the engine properties instead of Model properties where possible
parent 4f71a743
No related branches found
No related tags found
1 merge request!37Fix/post processing
Hide whitespace changes
Inline Side-by-side
src/bayesvalidrox/post_processing/post_processing.py
+ 24
− 45
View file @ 9805e543
...@@ -35,9 +35,6 @@ class PostProcessing: ...@@ -35,9 +35,6 @@ class PostProcessing:
def __init__(self, engine, name='calib'): def __init__(self, engine, name='calib'):
self.engine = engine self.engine = engine
self.MetaModel = engine.MetaModel
self.ExpDesign = engine.ExpDesign
self.ModelObj = engine.Model
self.name = name self.name = name
# ------------------------------------------------------------------------- # -------------------------------------------------------------------------
...@@ -63,20 +60,16 @@ class PostProcessing: ...@@ -63,20 +60,16 @@ class PostProcessing:
""" """
bar_plot = True if plot_type == 'bar' else False bar_plot = True if plot_type == 'bar' else False
meta_model_type = self.MetaModel.meta_model_type meta_model_type = self.engine.MetaModel.meta_model_type
Model = self.ModelObj
# Read Monte-Carlo reference # Read Monte-Carlo reference
self.mc_reference = Model.read_observation('mc_ref') self.mc_reference = self.engine.Model.read_observation('mc_ref')
# Set the x values # Set the x values
x_values_orig = self.engine.ExpDesign.x_values x_values_orig = self.engine.ExpDesign.x_values
# Compute the moments with the PCEModel object # Compute the moments with the PCEModel object
self.means, self.stds = self.MetaModel.calculate_moments() self.means, self.stds = self.engine.MetaModel.compute_moments()
# Get the variables
out_names = Model.Output.names
# Open a pdf for the plots # Open a pdf for the plots
newpath = (f'Outputs_PostProcessing_{self.name}/') newpath = (f'Outputs_PostProcessing_{self.name}/')
...@@ -85,7 +78,7 @@ class PostProcessing: ...@@ -85,7 +78,7 @@ class PostProcessing:
# Plot the best fit line, set the linewidth (lw), color and # Plot the best fit line, set the linewidth (lw), color and
# transparency (alpha) of the line # transparency (alpha) of the line
for key in out_names: for key in self.engine.out_names:
fig, ax = plt.subplots(nrows=1, ncols=2) fig, ax = plt.subplots(nrows=1, ncols=2)
# Extract mean and std # Extract mean and std
...@@ -173,9 +166,6 @@ class PostProcessing: ...@@ -173,9 +166,6 @@ class PostProcessing:
None. None.
""" """
MetaModel = self.MetaModel
Model = self.ModelObj
if samples is None: if samples is None:
self.n_samples = n_samples self.n_samples = n_samples
samples = self._get_sample() samples = self._get_sample()
...@@ -189,12 +179,12 @@ class PostProcessing: ...@@ -189,12 +179,12 @@ class PostProcessing:
self.model_out_dict = model_out_dict self.model_out_dict = model_out_dict
else: else:
self.model_out_dict = self._eval_model(samples, key_str='valid') self.model_out_dict = self._eval_model(samples, key_str='valid')
self.pce_out_mean, self.pce_out_std = MetaModel.eval_metamodel(samples) self.pce_out_mean, self.pce_out_std = self.engine.eval_metamodel(samples)
try: try:
key = Model.Output.names[1] key = self.engine.out_names[1]
except IndexError: except IndexError:
key = Model.Output.names[0] key = self.engine.out_names[0]
n_obs = self.model_out_dict[key].shape[1] n_obs = self.model_out_dict[key].shape[1]
...@@ -232,9 +222,6 @@ class PostProcessing: ...@@ -232,9 +222,6 @@ class PostProcessing:
Validation error for each output. Validation error for each output.
""" """
MetaModel = self.MetaModel
Model = self.ModelObj
# Set the number of samples # Set the number of samples
if n_samples: if n_samples:
self.n_samples = n_samples self.n_samples = n_samples
...@@ -252,12 +239,12 @@ class PostProcessing: ...@@ -252,12 +239,12 @@ class PostProcessing:
outputs = self._eval_model(samples, key_str='validSet') outputs = self._eval_model(samples, key_str='validSet')
# Run the PCE model with the generated samples # Run the PCE model with the generated samples
pce_outputs, _ = MetaModel.eval_metamodel(samples) pce_outputs, _ = self.engine.eval_metamodel(samples)
self.rmse = {} self.rmse = {}
self.valid_error = {} self.valid_error = {}
# Loop over the keys and compute RMSE error. # Loop over the keys and compute RMSE error.
for key in Model.Output.names: for key in self.engine.out_names:
# Root mena square # Root mena square
self.rmse[key] = mean_squared_error(outputs[key], pce_outputs[key], self.rmse[key] = mean_squared_error(outputs[key], pce_outputs[key],
squared=False, squared=False,
...@@ -274,8 +261,8 @@ class PostProcessing: ...@@ -274,8 +261,8 @@ class PostProcessing:
in enumerate(zip(self.rmse[key], in enumerate(zip(self.rmse[key],
self.valid_error[key])))) self.valid_error[key]))))
# Save error dicts in PCEModel object # Save error dicts in PCEModel object
self.MetaModel.rmse = self.rmse self.engine.MetaModel.rmse = self.rmse
self.MetaModel.valid_error = self.valid_error self.engine.MetaModel.valid_error = self.valid_error
return return
...@@ -296,7 +283,6 @@ class PostProcessing: ...@@ -296,7 +283,6 @@ class PostProcessing:
""" """
engine = self.engine engine = self.engine
PCEModel = self.MetaModel
n_init_samples = engine.ExpDesign.n_init_samples n_init_samples = engine.ExpDesign.n_init_samples
n_total_samples = engine.ExpDesign.X.shape[0] n_total_samples = engine.ExpDesign.X.shape[0]
...@@ -564,13 +550,13 @@ class PostProcessing: ...@@ -564,13 +550,13 @@ class PostProcessing:
""" """
# This function currently only supports PCE/aPCE # This function currently only supports PCE/aPCE
if not hasattr(self.MetaModel, 'meta_model_type'): PCEModel = self.engine.MetaModel
if not hasattr(PCEModel, 'meta_model_type'):
raise AttributeError('Sobol indices currently only support PCE-type models!') raise AttributeError('Sobol indices currently only support PCE-type models!')
if self.MetaModel.meta_model_type.lower() not in ['pce', 'apce']: if PCEModel.meta_model_type.lower() not in ['pce', 'apce']:
raise AttributeError('Sobol indices currently only support PCE-type models!') raise AttributeError('Sobol indices currently only support PCE-type models!')
# Extract the necessary variables # Extract the necessary variables
PCEModel = self.MetaModel
basis_dict = PCEModel._basis_dict basis_dict = PCEModel._basis_dict
coeffs_dict = PCEModel._coeffs_dict coeffs_dict = PCEModel._coeffs_dict
n_params = PCEModel.ndim n_params = PCEModel.ndim
...@@ -583,7 +569,7 @@ class PostProcessing: ...@@ -583,7 +569,7 @@ class PostProcessing:
sobol_cell_, total_sobol_ = {}, {} sobol_cell_, total_sobol_ = {}, {}
for output in self.ModelObj.Output.names: for output in self.engine.out_names:
n_meas_points = len(coeffs_dict[f'b_{b_i+1}'][output]) n_meas_points = len(coeffs_dict[f'b_{b_i+1}'][output])
...@@ -771,7 +757,7 @@ class PostProcessing: ...@@ -771,7 +757,7 @@ class PostProcessing:
total_sobol_all[k][i] = v total_sobol_all[k][i] = v
self.total_sobol = {} self.total_sobol = {}
for output in self.ModelObj.Output.names: for output in self.engine.out_names:
self.total_sobol[output] = np.mean(total_sobol_all[output], axis=0) self.total_sobol[output] = np.mean(total_sobol_all[output], axis=0)
# ---------------- Plot ----------------------- # ---------------- Plot -----------------------
...@@ -784,7 +770,7 @@ class PostProcessing: ...@@ -784,7 +770,7 @@ class PostProcessing:
fig = plt.figure() fig = plt.figure()
for outIdx, output in enumerate(self.ModelObj.Output.names): for outIdx, output in enumerate(self.engine.out_names):
# Extract total Sobol indices # Extract total Sobol indices
total_sobol = self.total_sobol[output] total_sobol = self.total_sobol[output]
...@@ -859,8 +845,6 @@ class PostProcessing: ...@@ -859,8 +845,6 @@ class PostProcessing:
None. None.
""" """
MetaModel = self.MetaModel
if samples is None: if samples is None:
self.n_samples = n_samples self.n_samples = n_samples
samples = self._get_sample() samples = self._get_sample()
...@@ -869,7 +853,7 @@ class PostProcessing: ...@@ -869,7 +853,7 @@ class PostProcessing:
# Evaluate the original and the surrogate model # Evaluate the original and the surrogate model
y_val = self._eval_model(samples, key_str='valid') y_val = self._eval_model(samples, key_str='valid')
y_pce_val, _ = MetaModel.eval_metamodel(samples=samples) y_pce_val, _ = self.engine.eval_metamodel(samples=samples)
# Open a pdf for the plots # Open a pdf for the plots
newpath = f'Outputs_PostProcessing_{self.name}/' newpath = f'Outputs_PostProcessing_{self.name}/'
...@@ -968,8 +952,7 @@ class PostProcessing: ...@@ -968,8 +952,7 @@ class PostProcessing:
self.n_samples = 1000 self.n_samples = 1000
PCEModel = self.MetaModel PCEModel = self.engine.MetaModel
Model = self.ModelObj
n_samples = self.n_samples n_samples = self.n_samples
# Create 3D-Grid # Create 3D-Grid
...@@ -1006,7 +989,7 @@ class PostProcessing: ...@@ -1006,7 +989,7 @@ class PostProcessing:
pce_out_std[:, idx] = y_std pce_out_std[:, idx] = y_std
# Model evaluation # Model evaluation
model_out_dict, _ = Model.run_model_parallel(sample_mesh, model_out_dict, _ = self.engine.Model.run_model_parallel(sample_mesh,
key_str='Valid3D') key_str='Valid3D')
model_outs[:, idx] = model_out_dict[Outkey].T model_outs[:, idx] = model_out_dict[Outkey].T
...@@ -1066,7 +1049,7 @@ class PostProcessing: ...@@ -1066,7 +1049,7 @@ class PostProcessing:
""" """
if n_samples is None: if n_samples is None:
n_samples = self.n_samples n_samples = self.n_samples
self.samples = self.ExpDesign.generate_samples( self.samples = self.engine.ExpDesign.generate_samples(
n_samples, n_samples,
sampling_method='random') sampling_method='random')
return self.samples return self.samples
...@@ -1090,15 +1073,13 @@ class PostProcessing: ...@@ -1090,15 +1073,13 @@ class PostProcessing:
Dictionary of results. Dictionary of results.
""" """
Model = self.ModelObj
if samples is None: if samples is None:
samples = self._get_sample() samples = self._get_sample()
self.samples = samples self.samples = samples
else: else:
self.n_samples = len(samples) self.n_samples = len(samples)
model_outs, _ = Model.run_model_parallel(samples, key_str=key_str) model_outs, _ = self.engine.Model.run_model_parallel(samples, key_str=key_str)
return model_outs return model_outs
...@@ -1189,8 +1170,6 @@ class PostProcessing: ...@@ -1189,8 +1170,6 @@ class PostProcessing:
None. None.
""" """
Model = self.ModelObj
newpath = f'Outputs_PostProcessing_{self.name}/' newpath = f'Outputs_PostProcessing_{self.name}/'
if not os.path.exists(newpath): if not os.path.exists(newpath):
os.makedirs(newpath) os.makedirs(newpath)
...@@ -1201,7 +1180,7 @@ class PostProcessing: ...@@ -1201,7 +1180,7 @@ class PostProcessing:
fig = plt.figure() fig = plt.figure()
# Plot the model vs PCE model # Plot the model vs PCE model
for keyIdx, key in enumerate(Model.Output.names): for keyIdx, key in enumerate(self.engine.out_names):
y_pce_val = self.pce_out_mean[key] y_pce_val = self.pce_out_mean[key]
y_pce_val_std = self.pce_out_std[key] y_pce_val_std = self.pce_out_std[key]
...@@ -1247,4 +1226,4 @@ class PostProcessing: ...@@ -1247,4 +1226,4 @@ class PostProcessing:
plt.close() plt.close()
# Zip the subdirectories # Zip the subdirectories
Model.zip_subdirs(f'{Model.name}valid', f'{Model.name}valid_') self.engine.Model.zip_subdirs(f'{self.engine.Model.name}valid', f'{self.engine.Model.name}valid_')
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment