Merge pull request #79 from Hynn01/main

Fix hyperparameters-tensorflow checker with a library check and Update README

README.md

@@ -62,7 +62,7 @@ git submodule update --init --recursive
 ```
 2. `dslinter` uses `poetry` to manage dependencies, so you will need to install `poetry` first and then install the dependencies. 
 ```
-pip install poerty
+pip install poetry
 poetry install
 ```
 - To install `dslinter` from source for development purposes, install it with:

dslinter/checkers/hyperparameters_tensorflow.py

@@ -4,6 +4,7 @@
 from pylint.lint import PyLinter
 from dslinter.checkers.hyperparameters import HyperparameterChecker
 from dslinter.utils.exception_handler import ExceptionHandler
+from dslinter.utils.hyperparameters_helper import check_module_with_library
 
 
 class HyperparameterTensorflowChecker(HyperparameterChecker):
@@ -82,7 +83,8 @@ def visit_call(self, node: astroid.Call):
             ):
                 self.hyperparameter_in_class(node, node.func.name)
             if(
-                hasattr(node, "func")
+                check_module_with_library(node, "tensorflow")
+                and hasattr(node, "func")
                 and hasattr(node.func, "attrname")
                 and node.func.attrname == "fit"
             ):

dslinter/tests/checkers/test_hyperparameters_tensorflow.py

@@ -111,3 +111,26 @@ def test_learning_rate_not_set(self):
         with self.assertAddsMessages(pylint.testutils.MessageTest(msg_id = "hyperparameters-tensorflow", node = call_node)):
             self.checker.visit_importfrom(importfrom_node)
             self.checker.visit_call(call_node)
+
+    def test_sklearn_fit(self):
+        script = """
+            from sklearn.datasets import load_wine
+            from sklearn.model_selection import train_test_split
+            from sklearn.pipeline import make_pipeline
+            from sklearn.decomposition import PCA
+            from sklearn.naive_bayes import GaussianNB
+            from sklearn.metrics import accuracy_score
+
+            RANDOM_STATE = 42
+            features, target = load_wine(return_X_y=True)
+            X_train, X_test, y_train, y_test = train_test_split(
+                features, target, test_size=0.30, random_state=RANDOM_STATE
+            )
+            clf = make_pipeline(PCA(n_components=2), GaussianNB())
+            clf.fit(X_train, y_train) #@
+            pred_test = clf.predict(X_test)
+            ac = accuracy_score(y_test, pred_test)
+        """
+        call_node = astroid.extract_node(script)
+        with self.assertNoMessages():
+            self.checker.visit_call(call_node)

dslinter/utils/hyperparameters_helper.py

@@ -0,0 +1,15 @@
+import astroid
+
+
+def check_module_with_library(node, library_name: str):
+    while not isinstance(node.parent, astroid.Module):
+        node = node.parent
+    module = node
+
+    if isinstance(module, astroid.Module):
+        for node in module.body:
+            if isinstance(node, astroid.Import):
+                for name, _ in node.names:
+                    if name == library_name:
+                        return True
+    return False

pyproject.toml

@@ -22,7 +22,7 @@ skip = 'scripts'
 
 [tool.poetry]
 name = "dslinter"
-version = "2.0.4"
+version = "2.0.5"
 description = "`dslinter` is a pylint plugin for linting data science and machine learning code. We plan to support the following Python libraries: TensorFlow, PyTorch, Scikit-Learn, Pandas, NumPy and SciPy."
 
 license = "GPL-3.0 License"