Converting the Scikit-learn model trained to the Core ML Format.

Author: DevOpsThinh

usecases/tabular_classifier/decision_tree_classifer/converting_scikit-learn_model_coreml.py

@@ -0,0 +1,25 @@
+# Author/ Learner: Nguyen Truong Thinh
+# Contact me: [email protected] || +84393280504
+#
+# Use case: Create a Decision Tree classification model that can be used to convert into the Core ML Format
+# via CoreML Tool .
+# The model will be trained & converted on the popular UCI ML Iris flowers dataset.
+
+import coremltools
+from iris_flowers_decision_tree_model import model_trained
+
+# Export the model to Core ML format
+coreml_model = coremltools.converters.sklearn.convert(model_trained, ["sepal length (cm)", "sepal width (cm)", "petal length (cm)", "petal width (cm)"], "target")
+coreml_model.author = 'Nguyen Truong Thinh'
+coreml_model.short_description = 'A Decision Tree model trained on the Iris flowers dataset.'
+# Features descriptions
+coreml_model.input_description['sepal length (cm)'] = 'Sepal length in cm.'
+coreml_model.input_description['sepal width (cm)'] = 'Sepal width in cm.'
+coreml_model.input_description['petal length (cm)'] = 'Petal length in cm.'
+coreml_model.input_description['petal width (cm)'] = 'Petal width in cm.'
+# Description of target variable
+coreml_model.output_description['target'] = 'A categorical value value, 0 = Iris-Setosa, 1 = Iris-Versicolour, 3 = Iris-Virginica'
+coreml_model.save("iris_flowers_dtm.mlpackage")
+
+
+

usecases/tabular_classifier/decision_tree_classifer/converting_skitlearn_model_coreml.py

@@ -1,8 +1,15 @@
+# Author/ Learner: Nguyen Truong Thinh
+# Contact me: [email protected] || +84393280504
+#
+# Use case: Create a Decision Tree classification model that can be used to convert into the Core ML Format
+# via CoreML Tool .
+# The model will be trained & converted on the popular UCI ML Iris flowers dataset.
+
 import numpy as np
 import matplotlib.pyplot as plt
 import pandas as pd
 import pydotplus
-
+# Scikit-learn: 1.1.2
 from sklearn import datasets
 from sklearn.metrics import accuracy_score
 from sklearn.model_selection import train_test_split
@@ -44,7 +51,7 @@
 # https:/scikit-learn.org/stable/modules;generated/sklearn.tree.DecisionTreeClassifier.html
 # Train a DTM
 model = DecisionTreeClassifier(random_state=17)
-model.fit(df_iris_features_train, df_iris_target_train.values.ravel())
+model_trained = model.fit(df_iris_features_train, df_iris_target_train.values.ravel())
 print(model.feature_importances_)
 # Get predictions from model, and compute accuracy
 predictions = model.predict(df_iris_features_test)
@@ -56,7 +63,7 @@
 export_graphviz(model, out_file=dot_data, filled=True, rounded=True,
                 special_characters=True, feature_names=df_iris_features.columns)
 graph = pydotplus.graph_from_dot_data(dot_data.getvalue())
-graph.write_png("iris_flow_dtm.png")
+graph.write_png("iris_flowers_dtm.png")
 Image(graph.create_png())
 
 

usecases/tabular_classifier/decision_tree_classifer/iris_flowers_decision_tree_model.py

@@ -0,0 +1,12 @@
+{
+    "fileFormatVersion": "1.0.0",
+    "itemInfoEntries": {
+        "7360340A-F6CF-46AC-99E4-3E2868896A7B": {
+            "author": "com.apple.CoreML",
+            "description": "CoreML Model Specification",
+            "name": "model.mlmodel",
+            "path": "com.apple.CoreML/model.mlmodel"
+        }
+    },
+    "rootModelIdentifier": "7360340A-F6CF-46AC-99E4-3E2868896A7B"
+}