v0.2.1

Author: bartzbeielstein

pyproject.toml

@@ -7,7 +7,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "spotPython"
-version = "0.2.0"
+version = "0.2.1"
 authors = [
   { name="T. Bartz-Beielstein", email="tbb@bartzundbartz.de" }
 ]

src/spotPython/data/torch_hyper_dict.json

@@ -33,10 +33,10 @@
                 "upper": 4},
         "k_folds": {
             "type": "int",
-            "default": 2,
+            "default": 1,
             "transform": "None",
-            "lower": 2,
-            "upper": 3},
+            "lower": 1,
+            "upper": 1},
         "patience": {
             "type": "int",
             "default": 5,
@@ -93,10 +93,10 @@
                 "upper": 4},
         "k_folds": {
             "type": "int",
-            "default": 2,
+            "default": 1,
             "transform": "None",
-            "lower": 2,
-            "upper": 3},
+            "lower": 1,
+            "upper": 1},
         "patience": {
             "type": "int",
             "default": 5,
@@ -157,10 +157,10 @@
                     "upper": 1},
             "k_folds": {
                 "type": "int",
-                "default": 2,
+                "default": 1,
                 "transform": "None",
-                "lower": 2,
-                "upper": 3
+                "lower": 1,
+                "upper": 1
             },
             "patience": {
                 "type": "int",

src/spotPython/data/torchdata.py

@@ -0,0 +1,12 @@
+from torchvision import datasets
+import torchvision.transforms as transforms
+
+
+def load_data_cifar10(data_dir="./data"):
+    transform = transforms.Compose([transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
+
+    trainset = datasets.CIFAR10(root=data_dir, train=True, download=True, transform=transform)
+
+    testset = datasets.CIFAR10(root=data_dir, train=False, download=True, transform=transform)
+
+    return trainset, testset

src/spotPython/hyperparameters/values.py

@@ -576,7 +576,7 @@ def get_values_from_dict(dictionary) -> np.array:
     return np.array(list(dictionary.values()))
 
 
-def add_core_model_to_fun_control(core_model, fun_control, hyper_dict, filename) -> dict:
+def add_core_model_to_fun_control(core_model, fun_control, hyper_dict, filename=None) -> dict:
     """Add the core model to the function control dictionary.
     Args:
         core_model (class): The core model.

src/spotPython/spot/spot.py

@@ -709,6 +709,18 @@ def plot_contour(
         if show:
             pylab.show()
 
+    def plot_important_hyperparameter_contour(self, threshold=0.025, filename=None):
+        impo = self.print_importance(threshold=threshold, print_screen=True)
+        var_plots = [i for i, x in enumerate(impo) if x[1] > threshold]
+        min_z = min(self.y)
+        max_z = max(self.y)
+        for i in var_plots:
+            for j in var_plots:
+                if j > i:
+                    if filename is not None:
+                        filename = filename + "_contour_" + str(i) + "_" + str(j) + ".png"
+                    self.plot_contour(i=i, j=j, min_z=min_z, max_z=max_z, filename=filename)
+
     def get_importance(self) -> list:
         """Get importance of each variable and return the results as a list.
         Returns:

src/spotPython/utils/init.py

@@ -48,5 +48,6 @@ def fun_control_init():
         "show_batch_interval": 1_000_000,
         "path": None,
         "save_model": False,
+        "weights": 1.0,
     }
     return fun_control