Upload P2 - Secom Notebook - Mercury.ipynb

P2 - Secom Notebook - Mercury.ipynb

@@ -26,7 +26,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 139,
+   "execution_count": 1,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -53,7 +53,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 140,
+   "execution_count": 2,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -64,7 +64,7 @@
      "data": {
       "application/mercury+json": {
        "allow_download": true,
-       "code_uid": "App.0.40.24.1-rand99a3439b",
+       "code_uid": "App.0.40.24.1-randc5ea2c84",
        "continuous_update": false,
        "description": "Recumpute everything dynamically",
        "full_screen": true,
@@ -96,7 +96,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 141,
+   "execution_count": 3,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -138,7 +138,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 142,
+   "execution_count": 4,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -195,7 +195,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 143,
+   "execution_count": 5,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -290,7 +290,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 144,
+   "execution_count": 6,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -341,7 +341,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 145,
+   "execution_count": 7,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -419,7 +419,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 146,
+   "execution_count": 8,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -499,7 +499,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 147,
+   "execution_count": 9,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -585,7 +585,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 148,
+   "execution_count": 10,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -648,7 +648,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 149,
+   "execution_count": 11,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -737,7 +737,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 150,
+   "execution_count": 12,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -818,7 +818,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 151,
+   "execution_count": 13,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -832,17 +832,17 @@
         "yes",
         "no"
        ],
-       "code_uid": "Select.0.40.16.25-rand28de2701",
+       "code_uid": "Select.0.40.16.25-randff0d51cb",
        "disabled": false,
        "hidden": false,
        "label": "Drop Duplicates",
-       "model_id": "1b5513f0c74f4b789b06e528c0702927",
+       "model_id": "cacee7b362e24d9491f76f9a6ef560cd",
        "url_key": "",
        "value": "yes",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "1b5513f0c74f4b789b06e528c0702927",
+       "model_id": "cacee7b362e24d9491f76f9a6ef560cd",
        "version_major": 2,
        "version_minor": 0
       },
@@ -856,18 +856,18 @@
     {
      "data": {
       "application/mercury+json": {
-       "code_uid": "Text.0.40.15.28-rand47e76187",
+       "code_uid": "Text.0.40.15.28-randf4e07b43",
        "disabled": false,
        "hidden": false,
        "label": "Missing Value Threeshold",
-       "model_id": "d938d6a0b2744021b8a2869fc3ed8d56",
+       "model_id": "8668847a73ba42bdaa31ab0b51777bd9",
        "rows": 1,
        "url_key": "",
        "value": "80",
        "widget": "Text"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "d938d6a0b2744021b8a2869fc3ed8d56",
+       "model_id": "8668847a73ba42bdaa31ab0b51777bd9",
        "version_major": 2,
        "version_minor": 0
       },
@@ -881,18 +881,18 @@
     {
      "data": {
       "application/mercury+json": {
-       "code_uid": "Text.0.40.15.31-randbeb3d20d",
+       "code_uid": "Text.0.40.15.31-rand9c2103a2",
        "disabled": false,
        "hidden": false,
        "label": "Variance Threshold",
-       "model_id": "7628bd791a434d4881994be8f0e7e104",
+       "model_id": "021b0af64c594ca9b5d0a00421dd0764",
        "rows": 1,
        "url_key": "",
        "value": "0",
        "widget": "Text"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "7628bd791a434d4881994be8f0e7e104",
+       "model_id": "021b0af64c594ca9b5d0a00421dd0764",
        "version_major": 2,
        "version_minor": 0
       },
@@ -906,18 +906,18 @@
     {
      "data": {
       "application/mercury+json": {
-       "code_uid": "Text.0.40.15.34-rand7204e09b",
+       "code_uid": "Text.0.40.15.34-rand8e9398fb",
        "disabled": false,
        "hidden": false,
        "label": "Correlation Threshold",
-       "model_id": "9c36001207a9406290a44dfbd27296e2",
+       "model_id": "cc8763e0552c4a75a8d722b9f933739c",
        "rows": 1,
        "url_key": "",
        "value": "1",
        "widget": "Text"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9c36001207a9406290a44dfbd27296e2",
+       "model_id": "cc8763e0552c4a75a8d722b9f933739c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -937,17 +937,17 @@
         4,
         5
        ],
-       "code_uid": "Select.0.40.16.38-rand6c036095",
+       "code_uid": "Select.0.40.16.38-rande38757e2",
        "disabled": false,
        "hidden": false,
        "label": "Outlier Removal Threshold",
-       "model_id": "deea9036e1dd45bdaf729893fb2c03ad",
+       "model_id": "f4da0f34706e406ab599a3292d095eb2",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "deea9036e1dd45bdaf729893fb2c03ad",
+       "model_id": "f4da0f34706e406ab599a3292d095eb2",
        "version_major": 2,
        "version_minor": 0
       },
@@ -963,22 +963,21 @@
       "application/mercury+json": {
        "choices": [
         "none",
-        "normal",
         "standard",
         "minmax",
         "robust"
        ],
-       "code_uid": "Select.0.40.16.46-rand6e19100d",
+       "code_uid": "Select.0.40.16.46-rand51c77a7b",
        "disabled": false,
        "hidden": false,
        "label": "Scaling Variables",
-       "model_id": "74c9a2bf7d774007a6e0aaee3c77b47a",
+       "model_id": "2acefeba429a46a5bd97a4664eda1f48",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "74c9a2bf7d774007a6e0aaee3c77b47a",
+       "model_id": "2acefeba429a46a5bd97a4664eda1f48",
        "version_major": 2,
        "version_minor": 0
       },
@@ -998,17 +997,17 @@
         "knn",
         "most_frequent"
        ],
-       "code_uid": "Select.0.40.16.50-rand44961a40",
+       "code_uid": "Select.0.40.16.50-rand6474c86d",
        "disabled": false,
        "hidden": false,
        "label": "Imputation Methods",
-       "model_id": "e7d32db61422400db77aed46104991be",
+       "model_id": "90be5bcc5dd54531b22d8b98bdad54f3",
        "url_key": "",
        "value": "mean",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e7d32db61422400db77aed46104991be",
+       "model_id": "90be5bcc5dd54531b22d8b98bdad54f3",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1029,17 +1028,17 @@
         "pca",
         "boruta"
        ],
-       "code_uid": "Select.0.40.16.55-rand17be4326",
+       "code_uid": "Select.0.40.16.55-rand84a19573",
        "disabled": false,
        "hidden": false,
        "label": "Feature Selection",
-       "model_id": "fecc32733d914aff9b0ad61cd4b7b6b5",
+       "model_id": "2ffbfd3253f24212ac14dbe0106109a1",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "fecc32733d914aff9b0ad61cd4b7b6b5",
+       "model_id": "2ffbfd3253f24212ac14dbe0106109a1",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1059,17 +1058,17 @@
         "undersampling",
         "rose"
        ],
-       "code_uid": "Select.0.40.16.59-rand8b476756",
+       "code_uid": "Select.0.40.16.59-rande70559fd",
        "disabled": false,
        "hidden": false,
        "label": "Imbalance Treatment",
-       "model_id": "e9479d12145f46009daeac5020fcea48",
+       "model_id": "e2d58738df2847d186095ca6c61f2a9c",
        "url_key": "",
        "value": "none",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "e9479d12145f46009daeac5020fcea48",
+       "model_id": "e2d58738df2847d186095ca6c61f2a9c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1092,17 +1091,17 @@
         "decision_tree",
         "xgboost"
        ],
-       "code_uid": "Select.0.40.16.64-randaa2cafdf",
+       "code_uid": "Select.0.40.16.64-rand19a17468",
        "disabled": false,
        "hidden": false,
        "label": "Model Selection",
-       "model_id": "a17088a739d847fcad51c1efc4aae6ff",
+       "model_id": "c5a9f64561944783aa2e71731fb67d6c",
        "url_key": "",
-       "value": "random_forest",
+       "value": "xgboost",
        "widget": "Select"
       },
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "a17088a739d847fcad51c1efc4aae6ff",
+       "model_id": "c5a9f64561944783aa2e71731fb67d6c",
        "version_major": 2,
        "version_minor": 0
       },
@@ -1179,7 +1178,7 @@
     "\n",
     "\n",
     "# input model\n",
-    "input_model = mr.Select(label=\"Model Selection\", value=\"random_forest\", choices=['random_forest', 'logistic_regression', 'knn', 'svm', 'naive_bayes','decision_tree','xgboost'])                      # 'all', 'random_forest', 'logistic_regression', 'knn', \n",
+    "input_model = mr.Select(label=\"Model Selection\", value=\"xgboost\", choices=['random_forest', 'logistic_regression', 'knn', 'svm', 'naive_bayes','decision_tree','xgboost'])                      # 'all', 'random_forest', 'logistic_regression', 'knn', \n",
     "                                        # 'svm', 'naive_bayes', # 'decision_tree', 'xgboost'\n",
     "input_model = str(input_model.value)\n"
    ]
@@ -1210,7 +1209,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 152,
+   "execution_count": 14,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -1291,7 +1290,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 153,
+   "execution_count": 15,
    "metadata": {
     "slideshow": {
      "slide_type": "skip"
@@ -1329,7 +1328,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 154,
+   "execution_count": 16,
    "metadata": {
     "slideshow": {
      "slide_type": "slide"
@@ -1367,7 +1366,7 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>random_forest</td>\n",
+       "      <td>xgboost</td>\n",
        "      <td>0.93</td>\n",
        "      <td>0.0</td>\n",
        "      <td>0.0</td>\n",
@@ -1378,8 +1377,8 @@
        "</div>"
       ],
       "text/plain": [
-       "           Model  Accuracy  Precision  Recall  F1-score\n",
-       "0  random_forest      0.93        0.0     0.0       0.0"
+       "     Model  Accuracy  Precision  Recall  F1-score\n",
+       "0  xgboost      0.93        0.0     0.0       0.0"
       ]
      },
      "metadata": {},
@@ -1416,9 +1415,9 @@
        "  <tbody>\n",
        "    <tr>\n",
        "      <th>0</th>\n",
-       "      <td>random_forest</td>\n",
-       "      <td>366</td>\n",
-       "      <td>0</td>\n",
+       "      <td>xgboost</td>\n",
+       "      <td>364</td>\n",
+       "      <td>2</td>\n",
        "      <td>26</td>\n",
        "      <td>0</td>\n",
        "    </tr>\n",
@@ -1427,11 +1426,8 @@
        "</div>"
       ],
       "text/plain": [
-       "           Model  True Negatives  False Positives  False Negatives  \\\n",
-       "0  random_forest             366                0               26   \n",
-       "\n",
-       "   True Positives  \n",
-       "0               0  "
+       "     Model  True Negatives  False Positives  False Negatives  True Positives\n",
+       "0  xgboost             364                2               26               0"
       ]
      },
      "metadata": {},
@@ -1439,7 +1435,7 @@
     },
     {
      "data": {
-      "image/png": "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",
+      "image/png": "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",
       "text/plain": [
        "<Figure size 500x500 with 1 Axes>"
       ]
@@ -1485,7 +1481,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 155,
+   "execution_count": 17,
    "metadata": {
     "slideshow": {
      "slide_type": "slide"
@@ -1513,7 +1509,7 @@
       "---------------------\n",
       "What is the imbalance treatment method? none\n",
       "---------------------\n",
-      "What is the model? random_forest\n"
+      "What is the model? xgboost\n"
      ]
     }
    ],