bump to python 3.12.3, deprecated knn gpu because of simsig dependency. minor changes and refactoring.

Author: SamoraHunter

.github/workflows/docs.yml

@@ -23,14 +23,42 @@ jobs:
   # Build job: builds the documentation and uploads it as an artifact.
   build:
     runs-on: ubuntu-latest
+    # Set ENV vars at the job level to be inherited by all steps, including actions
+    env:
+      NODE_EXTRA_CA_CERTS: /etc/ssl/certs/ca-certificates.crt
+      SSL_CERT_FILE: /etc/ssl/certs/ca-certificates.crt
+      REQUESTS_CA_BUNDLE: /etc/ssl/certs/ca-certificates.crt
+    
     steps:
       - name: Checkout repository
         uses: actions/checkout@v4
 
+      # Install Act dependencies FIRST (before any apt-get commands)
+      - name: Install Act dependencies 
+        if: ${{ env.ACT }} 
+        run: | 
+          # Ensure apt-get is available and install sudo for 'act' environments
+          if ! command -v apt-get &> /dev/null; then
+            echo "apt-get not found, cannot proceed"
+            exit 1
+          fi
+          # Install apt-utils first to avoid debconf warnings, then sudo
+          DEBIAN_FRONTEND=noninteractive apt-get update && \
+          DEBIAN_FRONTEND=noninteractive apt-get install -y apt-utils sudo
+
+      - name: Install Runner Dependencies & Configure CA
+        run: |
+          sudo apt-get update
+          # Install lsb-release, Node.js, and ca-certificates all at once
+          DEBIAN_FRONTEND=noninteractive sudo apt-get install -y lsb-release nodejs ca-certificates
+          
+          # Trust your mounted certificate (from the act command)
+          sudo update-ca-certificates
+
       - name: Set up Python
         uses: actions/setup-python@v5
         with:
-          python-version: '3.10'
+          python-version: '3.12.3'
           cache: 'pip'
 
       - name: Install dependencies
@@ -52,7 +80,21 @@ jobs:
       name: github-pages
       url: ${{ steps.deployment.outputs.page_url }}
     runs-on: ubuntu-latest
+    env:
+      NODE_EXTRA_CA_CERTS: /etc/ssl/certs/ca-certificates.crt
+      
     steps:
+      - name: Install Act dependencies 
+        if: ${{ env.ACT }} 
+        run: | 
+          apt-get update && apt-get install -y sudo
+
+      - name: Install Runner Dependencies & Configure CA
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y nodejs ca-certificates
+          sudo update-ca-certificates
+          
       - name: Deploy to GitHub Pages
         id: deployment
         uses: actions/deploy-pages@v4
@@ -62,7 +104,21 @@ jobs:
     needs: build
     if: ${{ vars.IS_GITEA == 'true' }}
     runs-on: ubuntu-latest
+    env:
+      NODE_EXTRA_CA_CERTS: /etc/ssl/certs/ca-certificates.crt
+
     steps:
+      - name: Install Act dependencies 
+        if: ${{ env.ACT }} 
+        run: | 
+          apt-get update && apt-get install -y sudo
+
+      - name: Install Runner Dependencies & Configure CA
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y nodejs ca-certificates
+          sudo update-ca-certificates
+
       - name: Download artifact
         uses: actions/download-artifact@v4
         with:

.github/workflows/notebook-test.yml

@@ -5,6 +5,12 @@ on: [push, pull_request]
 jobs:
   test:
     runs-on: ubuntu-latest
+    
+    
+    env:
+      NODE_EXTRA_CA_CERTS: /etc/ssl/certs/ca-certificates.crt
+      SSL_CERT_FILE: /etc/ssl/certs/ca-certificates.crt
+      REQUESTS_CA_BUNDLE: /etc/ssl/certs/ca-certificates.crt
 
     steps:
       - name: Check out repository
@@ -25,12 +31,15 @@ jobs:
       - name: Install Act dependencies 
         if: ${{ env.ACT }} 
         run: | 
-          apt-get update && apt-get install sudo -y 
+          apt-get update && apt-get install -y sudo
 
-      - name: Install CA cert tools and trust internal CA
+      - name: Install Runner Dependencies & Configure CA
         run: |
           sudo apt-get update
-          sudo apt-get install -y ca-certificates
+          # Install Node.js (for JS-based actions) & ca-certificates
+          sudo apt-get install -y nodejs ca-certificates
+          
+          # This trusts your mounted certificate (from the act command)
           sudo update-ca-certificates
 
       - name: Install ping utility
@@ -40,7 +49,6 @@ jobs:
 
       - name: Set timezone to UTC
         run: |
-          # Use sudo for all commands to ensure permissions in all runner environments
           sudo ln -fs /usr/share/zoneinfo/Etc/UTC /etc/localtime
           echo "Etc/UTC" | sudo tee /etc/timezone
           sudo apt-get update
@@ -57,37 +65,37 @@ jobs:
           sudo apt-get update
           sudo apt-get install -y default-jre
           
-      - name: Install Python 3.10, Git, and set CA environment
+      - name: Install build dependencies
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y build-essential    
+
+      - name: Install Python 3.12 and Git
         run: |
           sudo apt-get update
           sudo apt-get install -y lsb-release software-properties-common gnupg curl git
 
-          # Properly and securely add the GPG key for the deadsnakes PPA
+          # --- THIS IS THE GPG KEY FIX ---
+          
           sudo gpg --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys BA6932366A755776
           sudo gpg --export BA6932366A755776 | sudo gpg --dearmor -o /usr/share/keyrings/deadsnakes-archive-keyring.gpg
+          # --- END FIX ---
 
-          # Add the PPA repository, signed by the new key
           echo "deb [signed-by=/usr/share/keyrings/deadsnakes-archive-keyring.gpg] http://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu $(lsb_release -cs) main" | sudo tee /etc/apt/sources.list.d/deadsnakes-ppa.list
           
           sudo apt-get update
-          sudo apt-get install -y python3.10 python3.10-venv python3.10-distutils
+          sudo apt-get install -y python3.12 python3.12-venv python3.12-dev
 
-          # Set Python 3.10 as default using symbolic links
-          sudo ln -sf /usr/bin/python3.10 /usr/local/bin/python
-          sudo ln -sf /usr/bin/python3.10 /usr/local/bin/python3
+          sudo ln -sf /usr/bin/python3.12 /usr/local/bin/python
+          sudo ln -sf /usr/bin/python3.12 /usr/local/bin/python3
 
           python --version
-          
-          # Set the certificate path for all subsequent steps in the job
-          echo "SSL_CERT_FILE=/etc/ssl/certs/ca-certificates.crt" >> $GITHUB_ENV
-          echo "REQUESTS_CA_BUNDLE=/etc/ssl/certs/ca-certificates.crt" >> $GITHUB_ENV
 
       - name: Setup ML project
         run: |
           cd $GITHUB_WORKSPACE
-          # Make install.sh executable if it isn't already
           chmod +x install.sh
-          # Run the installation script
+          # This 'pip' command will now inherit the job-level 'REQUESTS_CA_BUNDLE'
           ./install.sh
 
       - name: Debug virtual environment

config_single_run.yml

@@ -45,6 +45,7 @@ models:
   H2O_XGBoost_class: true
   H2O_StackedEnsemble_class: false
   H2O_GAM_class: true
+  knn__gpu_wrapper_class: false
 
 # This section defines a single set of parameters for a standalone run.
 run_params:

ml_grid/model_classes/knn_gpu_classifier_class.py

@@ -11,7 +11,7 @@ class for the KNNWrapper (GPU-accelerated KNN). It provides parameter spaces for
 import pandas as pd
 from skopt.space import Categorical, Integer
 
-from ml_grid.model_classes.knn_wrapper_class import KNNWrapper
+#from ml_grid.model_classes.knn_wrapper_class import KNNWrapper
 from ml_grid.util import param_space
 from ml_grid.util.global_params import global_parameters
 
@@ -45,7 +45,7 @@ def __init__(
         self.y: Optional[pd.Series] = y
 
         # Initialize KNNWrapper for GPU support
-        self.algorithm_implementation: KNNWrapper = KNNWrapper()
+        #self.algorithm_implementation: KNNWrapper = KNNWrapper()
         self.method_name: str = "knn__gpu"
 
         # Define the parameter vector space

ml_grid/pipeline/hyperparameter_search.py

@@ -22,7 +22,7 @@
 )
 from ml_grid.model_classes.H2OXGBoostClassifier import H2OXGBoostClassifier
 from ml_grid.model_classes.keras_classifier_class import KerasClassifierClass
-from ml_grid.model_classes.knn_wrapper_class import KNNWrapper
+#from ml_grid.model_classes.knn_wrapper_class import KNNWrapper
 from ml_grid.model_classes.NeuralNetworkKerasClassifier import NeuralNetworkClassifier
 from ml_grid.util.global_params import global_parameters
 from ml_grid.util.validate_parameters import validate_parameters_helper
@@ -99,7 +99,7 @@ def __init__(
 
         # Custom wrappers that might not be recognized by is_classifier
         custom_classifier_types = (
-            KNNWrapper,
+            #KNNWrapper,
             H2OAutoMLClassifier,
             H2OGBMClassifier,
             H2ODRFClassifier,
@@ -192,7 +192,7 @@ def run_search(self, X_train: pd.DataFrame, y_train: pd.Series) -> BaseEstimator
 
         # Also limit n_jobs for Bayesian search and other specific wrappers to avoid issues.
         is_single_threaded_search = isinstance(
-            self.algorithm, (KNNWrapper, KerasClassifierClass, NeuralNetworkClassifier)
+            self.algorithm, ( KerasClassifierClass, NeuralNetworkClassifier) #KNNWrapper,
         )
 
         if is_h2o_model or is_single_threaded_search or bayessearch:

ml_grid/pipeline/model_class_list.py

@@ -35,7 +35,7 @@
 )
 from ml_grid.model_classes.h2o_xgboost_classifier_class import H2O_XGBoost_class
 from ml_grid.model_classes.knn_classifier_class import KNeighborsClassifierClass
-from ml_grid.model_classes.knn_gpu_classifier_class import KNNGpuWrapperClass
+#from ml_grid.model_classes.knn_gpu_classifier_class import KNNGpuWrapperClass
 from ml_grid.model_classes.keras_classifier_class import KerasClassifierClass
 from ml_grid.model_classes.light_gbm_class import LightGBMClassifierWrapper
 from ml_grid.model_classes.logistic_regression_class import LogisticRegressionClass
@@ -78,7 +78,7 @@
     "NeuralNetworkClassifier_class": NeuralNetworkClassifier_class, # Corrected mapping
     # GPU specific
     "KerasClassifierClass": KerasClassifierClass,
-    "KNNGpuWrapperClass": KNNGpuWrapperClass,
+    #"KNNGpuWrapperClass": KNNGpuWrapperClass, #deprecated by python 3.12 and simsig dependency
     # H2O Models
     "H2O_class": H2OAutoMLClass,  # Alias for AutoML
     "H2OAutoMLClass": H2OAutoMLClass,
@@ -147,7 +147,7 @@ def get_model_class_list(ml_grid_object: pipe) -> List[Any]:
             "LightGBMClassifierWrapper": True,
             "AdaBoostClassifierClass": True,
             "KerasClassifierClass": gpu_available,
-            "KNNGpuWrapperClass": gpu_available,
+            #"KNNGpuWrapperClass": gpu_available, #deprecated
             "NeuralNetworkClassifier_class": False,  # NNI based,
             "TabTransformerClass": False,  # PyTorch based
             "H2OAutoMLClass": False,  # H2O AutoML
@@ -169,7 +169,7 @@ def get_model_class_list(ml_grid_object: pipe) -> List[Any]:
         )
         models_to_disable = [
             "KerasClassifierClass",
-            "KNNGpuWrapperClass",
+            #"KNNGpuWrapperClass",
             "H2OAutoMLClass",
             "H2O_GBM_class",
             "H2O_DRF_class",

ml_grid/util/validate_parameters.py

@@ -6,8 +6,8 @@
 from sklearn.neighbors import KNeighborsClassifier
 from xgboost import XGBClassifier
 
-from ml_grid.model_classes.knn_gpu_classifier_class import KNNGpuWrapperClass
-from ml_grid.model_classes.knn_wrapper_class import KNNWrapper
+#from ml_grid.model_classes.knn_gpu_classifier_class import KNNGpuWrapperClass
+#from ml_grid.model_classes.knn_wrapper_class import KNNWrapper
 
 
 def validate_knn_parameters(
@@ -108,17 +108,17 @@ def validate_parameters_helper(
 
         return parameters
 
-    elif type(algorithm_implementation) == KNNWrapper:
+    # elif type(algorithm_implementation) == KNNWrapper:
 
-        parameters = validate_knn_parameters(parameters, ml_grid_object)
+    #     parameters = validate_knn_parameters(parameters, ml_grid_object)
 
-        return parameters
+    #     return parameters
 
-    elif isinstance(algorithm_implementation, KNNGpuWrapperClass):
+    # elif isinstance(algorithm_implementation, KNNGpuWrapperClass):
 
-        parameters = validate_knn_parameters(parameters, ml_grid_object)
+    #     parameters = validate_knn_parameters(parameters, ml_grid_object)
 
-        return parameters
+    #     return parameters
 
     elif isinstance(algorithm_implementation, XGBClassifier):
         parameters = validate_XGB_parameters(parameters, ml_grid_object)

notebooks/unit_test_synthetic.ipynb

@@ -1498,7 +1498,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
+   "version": "3.12.3"
   }
  },
  "nbformat": 4,