fix data loading

MLBazaar · Sep 25, 2023 · 68538ac · 68538ac
1 parent df295cf
commit 68538ac
Show file tree

Hide file tree

Showing 11 changed files with 81,562 additions and 47 deletions.
diff --git a/README.md b/README.md
@@ -87,10 +87,15 @@ pipeline which combines primitives from [MLPrimitives](https://github.com/MLBaza
 
 ```python3
 from mlblocks import MLPipeline
-from mlprimitives.datasets import load_dataset
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score
 
-dataset = load_dataset('census')
-X_train, X_test, y_train, y_test = dataset.get_splits(1)
+import pandas as pd
+
+dataset = pd.read_csv('http://mlblocks.s3.amazonaws.com/census.csv')
+label = dataset.pop('label')
+
+X_train, X_test, y_train, y_test = train_test_split(dataset, label, stratify=label)
 
 primitives = [
     'mlprimitives.custom.preprocessing.ClassEncoder',
@@ -104,7 +109,7 @@ pipeline = MLPipeline(primitives)
 pipeline.fit(X_train, y_train)
 predictions = pipeline.predict(X_test)
 
-dataset.score(y_test, predictions)
+accuracy_score(y_test, predictions)
 ```
 
 # What's Next?

diff --git a/examples/tutorials/1. Using and MLPipeline.ipynb b/examples/tutorials/1. Using and MLPipeline.ipynb
@@ -33,9 +33,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')"
+    "dataset = load_census()"
    ]
   },
   {
@@ -528,7 +528,16 @@
    "cell_type": "code",
    "execution_count": 13,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
    "source": [
     "pipeline.fit(X_train, y_train)"
    ]
@@ -546,9 +555,7 @@
   {
    "cell_type": "code",
    "execution_count": 14,
-   "metadata": {
-    "scrolled": false
-   },
+   "metadata": {},
    "outputs": [],
    "source": [
     "predictions = pipeline.predict(X_test)"
@@ -611,7 +618,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -625,7 +632,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.9"
+   "version": "3.8.16"
   }
  },
  "nbformat": 4,

diff --git a/examples/tutorials/3. Setting MLPipeline Hyperparameters.ipynb b/examples/tutorials/3. Setting MLPipeline Hyperparameters.ipynb
@@ -37,9 +37,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')\n",
+    "dataset = load_census()\n",
     "X_train, X_test, y_train, y_test = dataset.get_splits(1)"
    ]
   },
@@ -268,6 +268,14 @@
    "execution_count": 7,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    },
     {
      "data": {
       "text/plain": [
@@ -394,6 +402,14 @@
    "execution_count": 11,
    "metadata": {},
    "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    },
     {
      "data": {
       "text/plain": [
@@ -415,7 +431,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -429,7 +445,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.9"
+   "version": "3.8.16"
   }
  },
  "nbformat": 4,

diff --git a/examples/tutorials/4. Saving and Loading a Pipeline.ipynb b/examples/tutorials/4. Saving and Loading a Pipeline.ipynb
@@ -35,9 +35,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')"
+    "dataset = load_census()"
    ]
   },
   {
@@ -71,7 +71,16 @@
    "cell_type": "code",
    "execution_count": 4,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
    "source": [
     "pipeline.fit(X_train, y_train)"
    ]
@@ -166,7 +175,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -180,7 +189,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.9"
+   "version": "3.8.16"
   }
  },
  "nbformat": 4,

diff --git a/examples/tutorials/5. Partial execution and pipeline debugging.ipynb b/examples/tutorials/5. Partial execution and pipeline debugging.ipynb
@@ -36,9 +36,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')"
+    "dataset = load_census()"
    ]
   },
   {
@@ -430,7 +430,16 @@
    "cell_type": "code",
    "execution_count": 11,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
    "source": [
     "fit_context = pipeline.fit(start_=1, output_=2, **fit_context)"
    ]
@@ -690,7 +699,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -704,7 +713,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.9"
+   "version": "3.8.16"
   }
  },
  "nbformat": 4,

diff --git a/examples/tutorials/6. Flexible outputs specification.ipynb b/examples/tutorials/6. Flexible outputs specification.ipynb
@@ -37,9 +37,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')"
+    "dataset = load_census()"
    ]
   },
   {
@@ -420,7 +420,16 @@
    "cell_type": "code",
    "execution_count": 10,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
    "source": [
     "output_spec = [\n",
     "    'sklearn.impute.SimpleImputer#1.X',\n",
@@ -441,7 +450,16 @@
    "cell_type": "code",
    "execution_count": 11,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/Users/sarah/anaconda3/envs/mlp/lib/python3.8/site-packages/sklearn/impute/_base.py:382: FutureWarning: The 'verbose' parameter was deprecated in version 1.1 and will be removed in 1.3. A warning will always be raised upon the removal of empty columns in the future version.\n",
+      "  warnings.warn(\n"
+     ]
+    }
+   ],
    "source": [
     "output_spec = [\n",
     "    'mlprimitives.custom.feature_extraction.CategoricalEncoder#1.X',\n",
@@ -495,7 +513,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -509,7 +527,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.9"
+   "version": "3.8.16"
   }
  },
  "nbformat": 4,

diff --git a/examples/tutorials/7. Tuning a Pipeline.ipynb b/examples/tutorials/7. Tuning a Pipeline.ipynb
@@ -34,9 +34,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')"
+    "dataset = load_census()"
    ]
   },
   {

diff --git a/examples/tutorials/8. Searching for the best pipeline with BTBSession.ipynb b/examples/tutorials/8. Searching for the best pipeline with BTBSession.ipynb
@@ -37,9 +37,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from mlprimitives.datasets import load_dataset\n",
+    "from utils import load_census\n",
     "\n",
-    "dataset = load_dataset('census')"
+    "dataset = load_census()"
    ]
   },
   {
@@ -309,9 +309,7 @@
   {
    "cell_type": "code",
    "execution_count": 11,
-   "metadata": {
-    "scrolled": false
-   },
+   "metadata": {},
    "outputs": [
     {
      "data": {
@@ -536,9 +534,7 @@
   {
    "cell_type": "code",
    "execution_count": 13,
-   "metadata": {
-    "scrolled": false
-   },
+   "metadata": {},
    "outputs": [
     {
      "data": {
@@ -707,9 +703,7 @@
   {
    "cell_type": "code",
    "execution_count": 16,
-   "metadata": {
-    "scrolled": false
-   },
+   "metadata": {},
    "outputs": [
     {
      "data": {
@@ -772,7 +766,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
@@ -786,7 +780,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.9"
+   "version": "3.8.16"
   }
  },
  "nbformat": 4,