fix issue regarding tests and search space from yaml file

docs/pipeline_space.md

@@ -7,27 +7,25 @@ In NePS, defining the Search Space is one of two essential tasks. You can define
 
 To define the Search Space using a Python dictionary, follow these steps:
 
-
 Create a Python dictionary that specifies the parameters and their respective ranges. For example:
 
 ```python
 search_space = {
     "learning_rate": neps.FloatParameter(lower=0.00001, upper=0.1, log=True),
     "num_epochs": neps.IntegerParameter(lower=3, upper=30, is_fidelity=True),
     "optimizer": neps.CategoricalParameter(choices=["adam", "sgd", "rmsprop"]),
-    "dropout_rate": neps.FloatParameter(value=0.5)
+    "dropout_rate": neps.FloatParameter(value=0.5),
 }
-
 ```
 
 ## Option 2: Using a YAML File
-Create a YAML file (e.g., search_space.yaml) with the parameter definitions following this structure.
 
+Create a YAML file (e.g., search_space.yaml) with the parameter definitions following this structure.
 
 ```yaml
 search_space: # important to start with
   learning_rate:
-    lower: 0.00001
+    lower: 2e-3 # or 2*10^-3
     upper: 0.1
     log: true
 
@@ -43,50 +41,69 @@ search_space: # important to start with
     value: 0.5
 ...
 ```
+
 Ensure your YAML file starts with `search_space:`.
 This is the root key under which all parameter configurations are defined.
 
 ## Option 3: Using ConfigSpace
+
 For users familiar with the ConfigSpace library, can also define the Search Space through
 ConfigurationSpace()
+
 ```python
 from configspace import ConfigurationSpace, UniformFloatHyperparameter
 
 configspace = ConfigurationSpace()
-configspace.add_hyperparameter(UniformFloatHyperparameter("learning_rate", 0.00001, 0.1, log=True))
+configspace.add_hyperparameter(
+    UniformFloatHyperparameter("learning_rate", 0.00001, 0.1, log=True)
+)
 ```
-Link: https://github.com/automl/ConfigSpace
+
+For additional information on ConfigSpace and its features, please visit the following link:
+https://github.com/automl/ConfigSpace
 
 # Supported HyperParameter Types
 
 ### FloatParameter and IntegerParameter
+
 - **Expected Arguments:**
   - `lower`: The minimum value of the parameter.
   - `upper`: The maximum value of the parameter.
 - **Optional Arguments:**
+  - `type`: Specifies the data type of the parameter. Accepted Values: 'int', 'integer', or 'float'.
   - `log`: Indicates if the parameter uses a logarithmic scale (default: False).
   - `is_fidelity`: Marks the parameter as a fidelity parameter (default: False).
-  - `default`: Sets a prior central value for the parameter (default: None.
+  - `default`: Sets a prior central value for the parameter (default: None).
   - `default_confidence`: Specifies the confidence level of the default value,
-  indicating how strongly the prior
-   should be considered default: "low".
+    indicating how strongly the prior
+    should be considered default: "low".
 
 ### Categorical Parameter
+
 - **Expected Arguments:**
   - `choices`: A list of discrete options that the parameter can take.
 - **Optional Arguments:**
+  - `type`: Specifies the data type of the parameter. Accepted Values: 'cat' or 'categorical'.
   - `is_fidelity`: Marks the parameter as a fidelity parameter (default: False).
   - `default`: Sets a prior central value for the parameter (default: None.
   - `default_confidence`: Specifies the confidence level of the default value,
-  indicating how strongly the prior
+    indicating how strongly the prior
     should be considered default: "low".
 
 ### ConstantParameter
+
 - **Expected Arguments:**
   - `value`: The fixed value for the parameter.
 - **Optional Arguments:**
+  - `type`: Specifies the data type of the parameter. Accepted Values: 'const' or 'constant'.
   - `is_fidelity`: Marks the parameter as a fidelity parameter (default: False).
 
 # Supported ArchitectureParameter Types
 
+**Note**: The definition of Search Space from a YAML file is limited to supporting only Hyperparameter Types.
+
+If you are interested in exploring Architecture, particularly Hierarchical parameters, you can find detailed examples and usage in the following resources:
+
+- [Basic Usage Examples](https://github.com/automl/neps/tree/master/neps_examples/basic_usage) - This link provides basic usage examples that can help you understand the fundamentals of Architecture parameters.
 
+- [Experimental Examples](https://github.com/automl/neps/tree/master/neps_examples/experimental) - For more advanced and experimental use cases, including Hierarchical parameters, check out this collection of examples.

src/neps/search_spaces/search_space.py

@@ -146,18 +146,19 @@ def pipeline_space_from_yaml(yaml_file_path):
                     if not isinstance(details["lower"], int) or not isinstance(
                         details["upper"], int
                     ):
-                        # for numbers like 1e2
-                        details["lower"] = int(
-                            convert_scientific_notation(details["lower"])
-                        )
-                        details["upper"] = int(
-                            convert_scientific_notation(details["upper"])
-                        )
-                    else:
-                        raise TypeError(
-                            f"'lower' and 'upper' must be integer for "
-                            f"integer parameter '{name}'."
-                        )
+                        try:
+                            # for numbers like 1e2 and 10^
+                            details["lower"] = int(
+                                convert_scientific_notation(details["lower"])
+                            )
+                            details["upper"] = int(
+                                convert_scientific_notation(details["upper"])
+                            )
+                        except ValueError as e:
+                            raise TypeError(
+                                f"'lower' and 'upper' must be integer for "
+                                f"integer parameter '{name}'."
+                            ) from e
 
                 pipeline_space[name] = IntegerParameter(
                     lower=details["lower"],
@@ -178,14 +179,19 @@ def pipeline_space_from_yaml(yaml_file_path):
                     if not isinstance(details["lower"], float) or not isinstance(
                         details["upper"], float
                     ):
-                        # for numbers like 1e-5
-                        details["lower"] = convert_scientific_notation(details["lower"])
-                        details["upper"] = convert_scientific_notation(details["upper"])
-                    else:
-                        raise TypeError(
-                            f"'lower' and 'upper' must be integer for "
-                            f"integer parameter '{name}'."
-                        )
+                        try:
+                            # for numbers like 1e-5 and 10^
+                            details["lower"] = convert_scientific_notation(
+                                details["lower"]
+                            )
+                            details["upper"] = convert_scientific_notation(
+                                details["upper"]
+                            )
+                        except ValueError as e:
+                            raise TypeError(
+                                f"'lower' and 'upper' must be integer for "
+                                f"integer parameter '{name}'."
+                            ) from e
 
                 pipeline_space[name] = FloatParameter(
                     lower=details["lower"],
@@ -243,7 +249,7 @@ def convert_scientific_notation(value, show_usage_flag=False):
 
     e_notation_pattern = r"^-?\d+(\.\d+)?[eE]-?\d+$"
     # Pattern for '10^' style notation, with optional base and multiplication symbol
-    ten_power_notation_pattern = r"^(-?\d+)?(\.\d+)?\*?10\^(-?\d+)$"
+    ten_power_notation_pattern = r"^(-?\d+)?(\.\d+)?[xX*]?10\^(-?\d+)$"
 
     if isinstance(value, str):
         # Remove all whitespace from the string
@@ -260,7 +266,7 @@ def convert_scientific_notation(value, show_usage_flag=False):
                 if decimal:
                     base = base + decimal
                 base = float(base) if base else 1  # Default to 1 if base is empty
-                value = base * (10 ** float(exponent))
+                value = format(base * (10 ** float(exponent)), "e")
                 if show_usage_flag is True:
                     return float(value), True
                 else:

tests/test_yaml_search_space/config_including_wrong_types.yaml

@@ -6,7 +6,7 @@ search_space:
     log: true
 
   num_epochs:
-    type: float
+    type: cat
     lower: 3
     upper: 30
     is_fidelity: True

tests/test_yaml_search_space/correct_config.yaml

@@ -9,6 +9,20 @@ search_space:
     upper: 30
     log: false
     is_fidelity: True
+  batch_size:
+    type: int
+    lower: 10^2
+    upper: 3 * 10^4
+    log: true
+    is_fidelity: false
+
+  sec_learning_rate:
+    lower: 3.3e-5
+    upper: 1E-1
+
+  parameter_ex:
+    lower: 3.3 x 10^-5
+    upper: 3.2*10^1
 
   optimizer:
     choices: ["adam", "sgd", "rmsprop"]

tests/test_yaml_search_space/correct_config_including_types.yaml

@@ -6,11 +6,28 @@ search_space:
     log: true
 
   num_epochs:
-    type: int
+    type: integer
     lower: 3
     upper: 30
     is_fidelity: True
 
+  batch_size:
+    type: "int"
+    lower: 10^2
+    upper: 3 * 10^4
+    log: true
+    is_fidelity: false
+
+  sec_learning_rate:
+    type: "float"
+    lower: 3.3e-5
+    upper: 1E-1
+
+  parameter_ex:
+    type: float
+    lower: 3.3 x 10^-5
+    upper: 3.2*10^1
+
   optimizer:
     type: cat
     choices: ["adam", "sgd", "rmsprop"]

tests/test_yaml_search_space/test_search_space.py

@@ -27,6 +27,27 @@ def test_correct_yaml_file(path):
         assert pipeline_space["num_epochs"].is_fidelity is True
         assert pipeline_space["num_epochs"].default is None
         assert pipeline_space["num_epochs"].default_confidence_score == 0.5
+        assert isinstance(pipeline_space["batch_size"], IntegerParameter)
+        assert pipeline_space["batch_size"].lower == 100
+        assert pipeline_space["batch_size"].upper == 30000
+        assert pipeline_space["batch_size"].log is True
+        assert pipeline_space["batch_size"].is_fidelity is False
+        assert pipeline_space["batch_size"].default is None
+        assert pipeline_space["batch_size"].default_confidence_score == 0.5
+        assert isinstance(pipeline_space["sec_learning_rate"], FloatParameter)
+        assert pipeline_space["sec_learning_rate"].lower == 3.3e-5
+        assert pipeline_space["sec_learning_rate"].upper == 0.1
+        assert pipeline_space["sec_learning_rate"].log is False
+        assert pipeline_space["sec_learning_rate"].is_fidelity is False
+        assert pipeline_space["sec_learning_rate"].default is None
+        assert pipeline_space["sec_learning_rate"].default_confidence_score == 0.5
+        assert isinstance(pipeline_space["parameter_ex"], FloatParameter)
+        assert pipeline_space["parameter_ex"].lower == 3.3e-5
+        assert pipeline_space["parameter_ex"].upper == 32.0
+        assert pipeline_space["parameter_ex"].log is False
+        assert pipeline_space["parameter_ex"].is_fidelity is False
+        assert pipeline_space["parameter_ex"].default is None
+        assert pipeline_space["parameter_ex"].default_confidence_score == 0.5
         assert isinstance(pipeline_space["optimizer"], CategoricalParameter)
         assert pipeline_space["optimizer"].choices == ["adam", "sgd", "rmsprop"]
         assert pipeline_space["optimizer"].is_fidelity is False