jcmoptimizer 3.0.7__tar.gz → 3.0.9__tar.gz

{jcmoptimizer-3.0.7 → jcmoptimizer-3.0.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jcmoptimizer
-Version: 3.0.7
+Version: 3.0.9
 Summary: The JCMoptimizer interface for Python
 Author-email: Philipp Schneider <philipp.schneider@jcmwave.com>
 License-Expression: MIT

{jcmoptimizer-3.0.7 → jcmoptimizer-3.0.9}/jcmoptimizer/drivers/active_learning_drivers.py

@@ -131,9 +131,10 @@ class ActiveLearningDriver(Driver):
                    and a tensor of shape ``(num_samples, num_points, output_dim)``
 
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
           quantiles: A list with quantiles. If not specified, the quantiles
               [0.16,0.5,0.84] are used.
           num_samples: Number of samples used for posteriors that have
@@ -177,9 +178,10 @@ class ActiveLearningDriver(Driver):
           num_samples: Number of initial start samples for optimization
              (default: automatic determination)
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
         """
         self._run_task(
             "optimize hyperparameters",
@@ -229,9 +231,10 @@ class ActiveLearningDriver(Driver):
           delta: step size parameter used for approximating second
               derivatives.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with information about local minima
             with lists of object values, uncertainties of the objective values,
@@ -302,9 +305,10 @@ class ActiveLearningDriver(Driver):
           max_samples: The Monte Carlo integration is stopped when the
               number of evaluated samples equals or exceeds the given value.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the entries
 
@@ -371,9 +375,10 @@ class ActiveLearningDriver(Driver):
           max_samples: The Monte Carlo integration is stopped when the
               number of evaluated samples equals or exceeds the given value.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the entries
 
@@ -458,7 +463,7 @@ class ActiveLearningDriver(Driver):
         correlation vanishes at a correlation time :math:`\tau`, :math:`N_{\rm ind} =
         N/\tau` of all :math:`N` MCMC samples are independent.  Note, that
         :math:`\tau` can only be estimated and thus the relative Monte-Carlo
-        integration error :math:`\delta_{\rm rel}=\tau/\sqrt{N}` can be under or
+        integration error :math:`\delta_{\rm rel}=\sqrt{\tau/N}` can be under or
         overestimated.
         
         Args:
@@ -483,9 +488,10 @@ class ActiveLearningDriver(Driver):
               estimate. E.g. ``max_sigma_dist=3.0`` means that only the 99.7% 
               probability region of each parameter is sampled. 
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the following entries:
 
@@ -600,9 +606,10 @@ class SingleGPSingleVariable(SpecializedActiveLearning):
           num_samples: Number of initial start samples for optimization
               (default: automatic determination)
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
         """
         self._ald.optimize_hyperparameters("driver.surrogates.0", num_samples, min_dist)
 
@@ -641,9 +648,10 @@ class SingleGPSingleVariable(SpecializedActiveLearning):
               derivatives. (default: 0.1)
           ftol: Precision goal for the minimum function value.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with information about local minima
             with lists of object values, uncertainties of the objective values,
@@ -702,9 +710,10 @@ class BayesianOptimization(SingleGPSingleVariable, Minimizer):
                  and a tensor of shape ``(num_samples, num_points, output_dim)``
 
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
           quantiles: A list with quantiles. If not specified, the quantiles
               [0.16,0.5,0.84] are used.
           num_samples: Number of samples used for posteriors that have
@@ -714,7 +723,7 @@ class BayesianOptimization(SingleGPSingleVariable, Minimizer):
               the minimum of num_samples and the ensemble size is used.
 
         Returns: A dictionary with the entries "mean" and "variance" if
-            ``output_type = "mean_var`` and "samples" if ``output_type = "samples"``
+            ``output_type = "mean_var"`` and "samples" if ``output_type = "samples"``
         """
         return self._ald.predict(
             points=points,
@@ -763,9 +772,10 @@ class BayesianOptimization(SingleGPSingleVariable, Minimizer):
           max_samples: The Monte Carlo integration is stopped when the
               number of evaluated samples equals or exceeds the given value.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the entries
 
@@ -852,9 +862,10 @@ class BayesianLeastSquares(SingleGPSingleVariable, LeastSquaresDriver):
                  and a tensor of shape ``(num_samples, num_points, output_dim)``
 
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
           quantiles: A list with quantiles. If not specified, the quantiles
               [0.16,0.5,0.84] are used.
           num_samples: Number of samples used for posteriors that have
@@ -864,7 +875,7 @@ class BayesianLeastSquares(SingleGPSingleVariable, LeastSquaresDriver):
               the minimum of num_samples and the ensemble size is used.
 
         Returns: A dictionary with the entries "mean" and "variance" if
-            ``output_type = "mean_var`` and "samples" if ``output_type = "samples"``
+            ``output_type = "mean_var"`` and "samples" if ``output_type = "samples"``
         """
 
         name = "model_vector" if object_type == "surrogate" else "objective"
@@ -922,9 +933,10 @@ class BayesianLeastSquares(SingleGPSingleVariable, LeastSquaresDriver):
           max_samples: The Monte Carlo integration is stopped when the
               number of evaluated samples equals or exceeds the given value.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the entries
 
@@ -1013,9 +1025,10 @@ class BayesianLeastSquares(SingleGPSingleVariable, LeastSquaresDriver):
                estimate. E.g. max_sigma_dist=3.0 means that only the 99.7% 
                probability region of each parameter is sampled. (default: inf)
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the following entries:
 
@@ -1086,9 +1099,10 @@ class BayesianReconstruction(SingleGPSingleVariable, Minimizer):
                  and a tensor of shape ``(num_samples, num_points, output_dim)``
 
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
           quantiles: A list with quantiles. If not specified, the quantiles
               [0.16,0.5,0.84] are used.
           num_samples: Number of samples used for posteriors that have
@@ -1098,7 +1112,7 @@ class BayesianReconstruction(SingleGPSingleVariable, Minimizer):
               the minimum of num_samples and the ensemble size is used.
 
         Returns: A dictionary with the entries "mean" and "variance" if
-            ``output_type = "mean_var`` and "samples" if ``output_type = "samples"``
+            ``output_type = "mean_var"`` and "samples" if ``output_type = "samples"``
         """
 
         name = "model_vector" if object_type == "surrogate" else "objective"
@@ -1155,9 +1169,10 @@ class BayesianReconstruction(SingleGPSingleVariable, Minimizer):
           max_samples: The Monte Carlo integration is stopped when the
               number of evaluated samples equals or exceeds the given value.
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the entries
 
@@ -1246,9 +1261,10 @@ class BayesianReconstruction(SingleGPSingleVariable, Minimizer):
                estimate. E.g. max_sigma_dist=3.0 means that only the 99.7% 
                probability region of each parameter is sampled. (default: inf)
           min_dist: In order to speed up the prediction, one can use a
-              sparsified version of the base surrogates where sampling
-              with a distance smaller than min_dist (in terms of the length
-              scales of the surrogate) are neglected.
+              sparsified version of the base surrogates where observations
+              used for training the surrogate are neglected if their distance
+              is smaller than min_dist (in terms of the length scales of the
+              surrogate).
 
         Returns: A dictionary with the following entries:
 

{jcmoptimizer-3.0.7 → jcmoptimizer-3.0.9}/jcmoptimizer/requestor.py

@@ -185,7 +185,7 @@ class Requestor:
         path: str,
         data: Optional[dict[str, Any]] = None,
     ) -> requests.Response:
-        """Make a request and return vaidated response.
+        """Make a request and return validated response.
         Args:
             method: HTTP method ("POST", "GET")
             purpose: purpose of the request

jcmoptimizer-3.0.9/jcmoptimizer/version.py

@@ -0,0 +1 @@
+__version__ = '3.0.9'

{jcmoptimizer-3.0.7 → jcmoptimizer-3.0.9}/jcmoptimizer.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: jcmoptimizer
-Version: 3.0.7
+Version: 3.0.9
 Summary: The JCMoptimizer interface for Python
 Author-email: Philipp Schneider <philipp.schneider@jcmwave.com>
 License-Expression: MIT

jcmoptimizer-3.0.7/jcmoptimizer/version.py

@@ -1 +0,0 @@
-__version__ = '3.0.7'