From 57f96f8a99b05b7ff941eec7073786ba3c528ef8 Mon Sep 17 00:00:00 2001 From: samyarpotlapalli Date: Fri, 18 Oct 2024 15:17:54 -0400 Subject: [PATCH] Updated example with score testing --- ...m_binary_classification_model_import.ipynb | 577 +++--------------- 1 file changed, 83 insertions(+), 494 deletions(-) diff --git a/examples/pzmm_binary_classification_model_import.ipynb b/examples/pzmm_binary_classification_model_import.ipynb index 304644e7..bfcb4b6a 100644 --- a/examples/pzmm_binary_classification_model_import.ipynb +++ b/examples/pzmm_binary_classification_model_import.ipynb @@ -44,7 +44,7 @@ }, { "cell_type": "code", - "execution_count": 116, + "execution_count": null, "metadata": { "Collapsed": "false" }, @@ -70,7 +70,7 @@ }, { "cell_type": "code", - "execution_count": 117, + "execution_count": null, "metadata": { "Collapsed": "false" }, @@ -94,22 +94,11 @@ }, { "cell_type": "code", - "execution_count": 118, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "data": { - "text/plain": [ - "(5960, 13)" - ] - }, - "execution_count": 118, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "hmeq_data = pd.read_csv(\"data/hmeq.csv\", sep= \",\")\n", "hmeq_data.shape" @@ -117,199 +106,33 @@ }, { "cell_type": "code", - "execution_count": 119, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
BADLOANMORTDUEVALUEREASONJOBYOJDEROGDELINQCLAGENINQCLNODEBTINC
002680046236.062711.0DebtConOffice17.00.00.0175.0750581.022.033.059934
102690074982.0126972.0DebtConOffice0.00.00.0315.8189110.023.038.325990
202690067144.092923.0DebtConOther16.00.00.089.1121731.017.032.791478
302690045763.073797.0DebtConOther23.0NaN0.0291.5916811.029.039.370858
4027000144901.0178093.0DebtConProfExe7.00.00.0331.1139720.034.040.566552
\n", - "
" - ], - "text/plain": [ - " BAD LOAN MORTDUE VALUE REASON JOB YOJ DEROG DELINQ \\\n", - "0 0 26800 46236.0 62711.0 DebtCon Office 17.0 0.0 0.0 \n", - "1 0 26900 74982.0 126972.0 DebtCon Office 0.0 0.0 0.0 \n", - "2 0 26900 67144.0 92923.0 DebtCon Other 16.0 0.0 0.0 \n", - "3 0 26900 45763.0 73797.0 DebtCon Other 23.0 NaN 0.0 \n", - "4 0 27000 144901.0 178093.0 DebtCon ProfExe 7.0 0.0 0.0 \n", - "\n", - " CLAGE NINQ CLNO DEBTINC \n", - "0 175.075058 1.0 22.0 33.059934 \n", - "1 315.818911 0.0 23.0 38.325990 \n", - "2 89.112173 1.0 17.0 32.791478 \n", - "3 291.591681 1.0 29.0 39.370858 \n", - "4 331.113972 0.0 34.0 40.566552 " - ] - }, - "execution_count": 119, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "hmeq_data.head()" ] }, { "cell_type": "code", - "execution_count": 120, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "hmeq_data.hist(figsize=(15,15), layout=(4, 4));" ] }, { "cell_type": "code", - "execution_count": 121, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "data": { - "text/plain": [ - "Index(['BAD', 'LOAN', 'MORTDUE', 'VALUE', 'REASON', 'JOB', 'YOJ', 'DEROG',\n", - " 'DELINQ', 'CLAGE', 'NINQ', 'CLNO', 'DEBTINC'],\n", - " dtype='object')" - ] - }, - "execution_count": 121, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "hmeq_data.columns" ] @@ -325,7 +148,7 @@ }, { "cell_type": "code", - "execution_count": 122, + "execution_count": null, "metadata": { "Collapsed": "false" }, @@ -342,7 +165,7 @@ }, { "cell_type": "code", - "execution_count": 123, + "execution_count": null, "metadata": { "Collapsed": "false" }, @@ -364,7 +187,7 @@ }, { "cell_type": "code", - "execution_count": 124, + "execution_count": null, "metadata": { "Collapsed": "false" }, @@ -382,7 +205,7 @@ }, { "cell_type": "code", - "execution_count": 125, + "execution_count": null, "metadata": { "Collapsed": "false" }, @@ -398,121 +221,11 @@ }, { "cell_type": "code", - "execution_count": 126, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
DecisionTreeRandomForestGradientBoosting
DEBTINC58.35%35.36%39.53%
DELINQ18.57%15.16%12.25%
CLAGE8.07%9.49%10.39%
DEROG4.86%9.28%6.28%
VALUE3.19%5.75%7.38%
YOJ2.7%3.59%3.77%
MORTDUE2.0%5.17%6.4%
CLNO1.2%5.52%5.9%
NINQ0.88%3.93%2.79%
LOAN0.17%6.74%5.32%
\n", - "
" - ], - "text/plain": [ - " DecisionTree RandomForest GradientBoosting\n", - "DEBTINC 58.35% 35.36% 39.53%\n", - "DELINQ 18.57% 15.16% 12.25%\n", - "CLAGE 8.07% 9.49% 10.39%\n", - "DEROG 4.86% 9.28% 6.28%\n", - "VALUE 3.19% 5.75% 7.38%\n", - "YOJ 2.7% 3.59% 3.77%\n", - "MORTDUE 2.0% 5.17% 6.4%\n", - "CLNO 1.2% 5.52% 5.9%\n", - "NINQ 0.88% 3.93% 2.79%\n", - "LOAN 0.17% 6.74% 5.32%" - ] - }, - "execution_count": 126, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "importances = pd.DataFrame.from_dict(sort_feature_importance(dtc, x_train), orient=\"index\").rename(columns={0: \"DecisionTree\"})\n", "importances[\"RandomForest\"] = pd.DataFrame.from_dict(sort_feature_importance(rfc, x_train), orient=\"index\")\n", @@ -522,30 +235,11 @@ }, { "cell_type": "code", - "execution_count": 127, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[[1428 13]\n", - " [ 279 68]]\n", - " precision recall f1-score support\n", - "\n", - " 0 0.84 0.99 0.91 1441\n", - " 1 0.84 0.20 0.32 347\n", - "\n", - " accuracy 0.84 1788\n", - " macro avg 0.84 0.59 0.61 1788\n", - "weighted avg 0.84 0.84 0.79 1788\n", - "\n", - "Decision Tree Model Accuracy = 83.67%\n" - ] - } - ], + "outputs": [], "source": [ "y_dtc_predict = dtc.predict(x_test)\n", "y_dtc_proba = dtc.predict_proba(x_test)\n", @@ -556,30 +250,11 @@ }, { "cell_type": "code", - "execution_count": 128, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[[1439 2]\n", - " [ 258 89]]\n", - " precision recall f1-score support\n", - "\n", - " 0 0.85 1.00 0.92 1441\n", - " 1 0.98 0.26 0.41 347\n", - "\n", - " accuracy 0.85 1788\n", - " macro avg 0.91 0.63 0.66 1788\n", - "weighted avg 0.87 0.85 0.82 1788\n", - "\n", - "Random Forest Model Accuracy = 85.46%\n" - ] - } - ], + "outputs": [], "source": [ "y_rfc_predict = rfc.predict(x_test)\n", "y_rfc_proba = rfc.predict_proba(x_test)\n", @@ -590,28 +265,9 @@ }, { "cell_type": "code", - "execution_count": 129, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "[[1436 5]\n", - " [ 202 145]]\n", - " precision recall f1-score support\n", - "\n", - " 0 0.88 1.00 0.93 1441\n", - " 1 0.97 0.42 0.58 347\n", - "\n", - " accuracy 0.88 1788\n", - " macro avg 0.92 0.71 0.76 1788\n", - "weighted avg 0.89 0.88 0.86 1788\n", - "\n", - "Gradient Boosting Model Accuracy = 88.42%\n" - ] - } - ], + "outputs": [], "source": [ "y_gbc_predict = gbc.predict(x_test)\n", "y_gbc_proba = gbc.predict_proba(x_test)\n", @@ -631,21 +287,11 @@ }, { "cell_type": "code", - "execution_count": 130, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Model DecisionTreeClassifier was successfully pickled and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\DecisionTreeClassifier.pickle.\n", - "Model RandomForest was successfully pickled and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\RandomForest.pickle.\n", - "Model GradientBoosting was successfully pickled and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\GradientBoosting.pickle.\n" - ] - } - ], + "outputs": [], "source": [ "# Model names within SAS Model Manager\n", "model_prefix = [\"DecisionTreeClassifier\", \"RandomForest\", \"GradientBoosting\"]\n", @@ -668,30 +314,11 @@ }, { "cell_type": "code", - "execution_count": 131, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "inputVar.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\inputVar.json\n", - "outputVar.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\outputVar.json\n", - "ModelProperties.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\ModelProperties.json\n", - "fileMetadata.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\fileMetadata.json\n", - "inputVar.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\inputVar.json\n", - "outputVar.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\outputVar.json\n", - "ModelProperties.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\ModelProperties.json\n", - "fileMetadata.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\fileMetadata.json\n", - "inputVar.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\inputVar.json\n", - "outputVar.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\outputVar.json\n", - "ModelProperties.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\ModelProperties.json\n", - "fileMetadata.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\fileMetadata.json\n" - ] - } - ], + "outputs": [], "source": [ "def write_json_files(data, predict, target, path, prefix): \n", " # Write input variable mapping to a json file\n", @@ -722,39 +349,10 @@ { "cell_type": "code", "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [] - }, - { - "cell_type": "code", - "execution_count": 132, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "dmcas_fitstat.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\dmcas_fitstat.json\n", - "dmcas_roc.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\dmcas_roc.json\n", - "dmcas_lift.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\dmcas_lift.json\n", - "dmcas_relativeimportance.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\dmcas_relativeimportance.json\n", - "dmcas_misc.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\dmcas_misc.json\n", - "dmcas_fitstat.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\dmcas_fitstat.json\n", - "dmcas_roc.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\dmcas_roc.json\n", - "dmcas_lift.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\dmcas_lift.json\n", - "dmcas_relativeimportance.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\dmcas_relativeimportance.json\n", - "dmcas_misc.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\dmcas_misc.json\n", - "dmcas_fitstat.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\dmcas_fitstat.json\n", - "dmcas_roc.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\dmcas_roc.json\n", - "dmcas_lift.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\dmcas_lift.json\n", - "dmcas_relativeimportance.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\dmcas_relativeimportance.json\n", - "dmcas_misc.json was successfully written and saved to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\dmcas_misc.json\n" - ] - } - ], + "outputs": [], "source": [ "import getpass\n", "def write_model_stats(x_train, y_train, test_predict, test_proba, y_test, model, path, prefix):\n", @@ -789,9 +387,9 @@ " selection_statistic='_RASE_',\n", " )\n", " \n", - "username = \"edmdev\"\n", - "password = \"Go4thsas\"\n", - "host = \"base.ingress-nginx.wenbao-gc1-m1.modelmanager.sashq-d.openstack.sas.com\" # Changes required by user\n", + "username = getpass.getpass()\n", + "password = getpass.getpass()\n", + "host = \"demo.sas.com\"# Changes required by user\n", "sess = Session(host, username, password, protocol=\"http\") # For TLS-enabled servers, change protocol value to \"https\"\n", "conn = sess.as_swat() # Connect to SWAT through the sasctl authenticated connection\n", "\n", @@ -803,69 +401,11 @@ }, { "cell_type": "code", - "execution_count": 133, + "execution_count": null, "metadata": { "Collapsed": "false" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Model score code was written successfully to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier\\score_DecisionTreeClassifier.py and uploaded to SAS Model Manager.\n", - "All model files were zipped to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\DecisionTreeClassifier.\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "c:\\Users\\sapotl\\AppData\\Local\\anaconda3\\Lib\\site-packages\\sasctl\\pzmm\\write_score_code.py:1500: UserWarning: Due to the ambiguity of the provided metrics and prediction return types, the score code assumes that a classification and the target event probability should be returned.\n", - " warn(\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Model was successfully imported into SAS Model Manager as DecisionTreeClassifier with the following UUID: 81806d99-6070-4bd0-8207-646e42eb797e.\n", - "Model score code was written successfully to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest\\score_RandomForest.py and uploaded to SAS Model Manager.\n", - "All model files were zipped to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\RandomForest.\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "c:\\Users\\sapotl\\AppData\\Local\\anaconda3\\Lib\\site-packages\\sasctl\\pzmm\\write_score_code.py:1500: UserWarning: Due to the ambiguity of the provided metrics and prediction return types, the score code assumes that a classification and the target event probability should be returned.\n", - " warn(\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Model was successfully imported into SAS Model Manager as RandomForest with the following UUID: 7bf4273f-d54b-4201-8107-f3503e12a43f.\n", - "Model score code was written successfully to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting\\score_GradientBoosting.py and uploaded to SAS Model Manager.\n", - "All model files were zipped to c:\\Users\\sapotl\\OneDrive - SAS\\Desktop\\sasctlofficial\\python-sasctl\\examples\\data\\hmeqModels\\GradientBoosting.\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "c:\\Users\\sapotl\\AppData\\Local\\anaconda3\\Lib\\site-packages\\sasctl\\pzmm\\write_score_code.py:1500: UserWarning: Due to the ambiguity of the provided metrics and prediction return types, the score code assumes that a classification and the target event probability should be returned.\n", - " warn(\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Model was successfully imported into SAS Model Manager as GradientBoosting with the following UUID: db434720-48ea-4fc2-95d2-b20e472f78e7.\n" - ] - } - ], + "outputs": [], "source": [ "for (prefix, path) in zip(model_prefix, zip_folder):\n", " pzmm.ImportModel.import_model(\n", @@ -885,13 +425,62 @@ " pzmm.ScoreCode.score_code = \"\"" ] }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Implementing Score Testing" + ] + }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ - "print(pzmm.ImportModel.get_model_properties(\"id\"))" + "import requests\n", + "import sys\n", + "\n", + "from sasctl._services.score_definitions import ScoreDefinitions as sd # Importing ScoreDefinitions service\n", + "from sasctl._services.score_execution import ScoreExecution as se # Importing ScoreExecution service" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Creating the score definition for this model using the model UUID generated two steps before\n", + "score_definition = sd.create_score_definition(\n", + " \"example_score_def_name\", # Name of the score_definition, which can be any string\n", + " \"model_id\", # Use Model UUID generated two steps before\n", + " \"table_name\", # Table name for input data, which must exist in host server or it will throw an HTTP error and prompt you to upload a data file\n", + " # True, # Uncomment 'True' if your Viya version is compatible with CAS Gateway\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Executing the score definition\n", + "score_execution = se.create_score_execution(\n", + " score_definition.get(\"id\") # Score definition id created in the previous cell\n", + ")\n", + "\n", + "# Prints score_execution_id\n", + "print(score_execution)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We can view our scored model information within Model Manager under Projects -> Choose your model -> Scoring. \n", + "***" ] } ], @@ -911,7 +500,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.7" + "version": "3.11.0" }, "latex_envs": { "LaTeX_envs_menu_present": true,