refactor: add init file for easier usage of transformers (#45)

docs/README.md

@@ -74,7 +74,7 @@ Let's assume you want to use some method from [NumPy's mathematical functions, t
 use the [`MathExpressionTransformer`](https://chrislemke.github.io/sk-transformers/number_transformer-reference/#sk-transformers.transformer.number_transformer.MathExpressionTransformer).
 ```python
 import pandas as pd
-from sk_transformers.number_transformer import MathExpressionTransformer
+from sk_transformers import MathExpressionTransformer
 
 X = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
 transformer = MathExpressionTransformer([("foo", "np.sum", "bar", {"axis": 0})])
@@ -91,8 +91,8 @@ In the next example, we additionally add the [`MapTransformer`](https://chrislem
 Together with [scikit-learn's pipelines](https://scikit-learn.org/stable/modules/compose.html#combining-estimators) it would look like this:
 ```python
 import pandas as pd
-from sk_transformers.number_transformer import MathExpressionTransformer
-from sk_transformers.generic_transformer import MapTransformer
+from sk_transformers import MathExpressionTransformer
+from sk_transformers import MapTransformer
 from sklearn.pipeline import Pipeline
 
 X = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})

examples/playground.ipynb

@@ -56,7 +56,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.datetime_transformer import DurationCalculatorTransformer\n",
+    "from sk_transformers import DurationCalculatorTransformer\n",
     "\n",
     "X = pd.DataFrame(\n",
     "    {\n",
@@ -85,7 +85,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.datetime_transformer import TimestampTransformer\n",
+    "from sk_transformers import TimestampTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"1960-01-01\", \"1970-01-01\", \"1990-01-01\"]})\n",
     "transformer = TimestampTransformer([\"foo\"])\n",
@@ -123,7 +123,7 @@
     "import numpy as np\n",
     "import pandas as pd\n",
     "from pytorch_widedeep.datasets import load_adult\n",
-    "from sk_transformers.deep_transformer import ToVecTransformer\n",
+    "from sk_transformers import ToVecTransformer\n",
     "\n",
     "df = load_adult(as_frame=True)\n",
     "df[\"target\"] = (df[\"income\"].apply(lambda x: \">50K\" in x)).astype(int)\n",
@@ -168,7 +168,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.encoder_transformer import MeanEncoderTransformer\n",
+    "from sk_transformers import MeanEncoderTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"a\", \"b\", \"a\", \"c\", \"b\", \"a\", \"c\", \"a\", \"b\", \"c\"]})\n",
     "y = pd.Series([1, 0, 1, 0, 1, 0, 1, 0, 1, 0])\n",
@@ -204,7 +204,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import AggregateTransformer\n",
+    "from sk_transformers import AggregateTransformer\n",
     "\n",
     "X = pd.DataFrame(\n",
     "    {\n",
@@ -234,7 +234,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import ColumnDropperTransformer\n",
+    "from sk_transformers import ColumnDropperTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, 2, 3], \"bar\": [4, 5, 6]})\n",
     "transformer = ColumnDropperTransformer([\"foo\"])\n",
@@ -259,7 +259,7 @@
    "source": [
     "import numpy as np\n",
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import DtypeTransformer\n",
+    "from sk_transformers import DtypeTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, 2, 3], \"bar\": [\"a\", \"a\", \"b\"]})\n",
     "transformer = DtypeTransformer([(\"foo\", np.float32), (\"bar\", \"category\")])\n",
@@ -285,7 +285,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import FunctionsTransformer\n",
+    "from sk_transformers import FunctionsTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, 2, 3], \"bar\": [4, 5, 6]})\n",
     "transformer = FunctionsTransformer([(\"foo\", np.log1p, None), (\"bar\", np.sqrt, None)])\n",
@@ -309,7 +309,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import MapTransformer\n",
+    "from sk_transformers import MapTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, 2, 3], \"bar\": [4, 5, 6]})\n",
     "transformer = MapTransformer([(\"foo\", lambda x: x + 1)])\n",
@@ -335,7 +335,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import LeftJoinTransformer\n",
+    "from sk_transformers import LeftJoinTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"A\", \"B\", \"C\", \"A\", \"C\"]})\n",
     "lookup_df = pd.Series([1, 2, 3], index=[\"A\", \"B\", \"C\"], name=\"values\")\n",
@@ -359,7 +359,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from sk_transformers.generic_transformer import NaNTransformer\n",
+    "from sk_transformers import NaNTransformer\n",
     "import pandas as pd\n",
     "import numpy as np\n",
     "\n",
@@ -389,7 +389,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import QueryTransformer\n",
+    "from sk_transformers import QueryTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, 8, 3, 6, 5, 4, 7, 2]})\n",
     "transformer = QueryTransformer([\"foo > 4\"])\n",
@@ -418,7 +418,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from sk_transformers.generic_transformer import ValueIndicatorTransformer\n",
+    "from sk_transformers import ValueIndicatorTransformer\n",
     "import pandas as pd\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, -999, 3], \"bar\": [\"a\", \"-999\", \"c\"]})\n",
@@ -446,7 +446,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.generic_transformer import ValueReplacerTransformer\n",
+    "from sk_transformers import ValueReplacerTransformer\n",
     "\n",
     "X = pd.DataFrame(\n",
     "    {\"foo\": [\"0000-01-01\", \"2022/01/08\", \"bar\", \"1982-12-7\", \"28-09-2022\"]}\n",
@@ -492,7 +492,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.number_transformer import MathExpressionTransformer\n",
+    "from sk_transformers import MathExpressionTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [1, 2, 3], \"bar\": [4, 5, 6]})\n",
     "transformer = MathExpressionTransformer([(\"foo\", \"np.sum\", \"bar\", {\"axis\": 0})])\n",
@@ -524,7 +524,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.string_transformer import EmailTransformer\n",
+    "from sk_transformers import EmailTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"[email protected]\"]})\n",
     "transformer = EmailTransformer([\"foo\"])\n",
@@ -550,7 +550,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.string_transformer import IPAddressEncoderTransformer\n",
+    "from sk_transformers import IPAddressEncoderTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"192.168.1.1\", \"2001:0db8:3c4d:0015:0000:0000:1a2f:1a2b\"]})\n",
     "transformer = IPAddressEncoderTransformer([\"foo\"])\n",
@@ -574,7 +574,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.string_transformer import PhoneTransformer\n",
+    "from sk_transformers import PhoneTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"+49123456789\", \"0044987654321\", \"3167891234\"]})\n",
     "transformer = PhoneTransformer([\"foo\"])\n",
@@ -598,7 +598,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.string_transformer import StringSimilarityTransformer\n",
+    "from sk_transformers import StringSimilarityTransformer\n",
     "\n",
     "X = pd.DataFrame(\n",
     "    {\n",
@@ -630,7 +630,7 @@
    "outputs": [],
    "source": [
     "import pandas as pd\n",
-    "from sk_transformers.string_transformer import StringSlicerTransformer\n",
+    "from sk_transformers import StringSlicerTransformer\n",
     "\n",
     "X = pd.DataFrame({\"foo\": [\"abc\", \"def\", \"ghi\"], \"bar\": [\"jkl\", \"mno\", \"pqr\"]})\n",
     "transformer = StringSlicerTransformer([(\"foo\", (0, 3, 2)), (\"bar\", (2,))])\n",
@@ -662,7 +662,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.8"
+   "version": "3.10.8 | packaged by conda-forge | (main, Nov 22 2022, 08:25:29) [Clang 14.0.6 ]"
   },
   "vscode": {
    "interpreter": {

src/sk_transformers/__init__.py

@@ -0,0 +1,26 @@
+from sk_transformers.datetime_transformer import (
+    DurationCalculatorTransformer,
+    TimestampTransformer,
+)
+from sk_transformers.deep_transformer import ToVecTransformer
+from sk_transformers.encoder_transformer import MeanEncoderTransformer
+from sk_transformers.generic_transformer import (
+    AggregateTransformer,
+    ColumnDropperTransformer,
+    DtypeTransformer,
+    FunctionsTransformer,
+    LeftJoinTransformer,
+    MapTransformer,
+    NaNTransformer,
+    QueryTransformer,
+    ValueIndicatorTransformer,
+    ValueReplacerTransformer,
+)
+from sk_transformers.number_transformer import MathExpressionTransformer
+from sk_transformers.string_transformer import (
+    EmailTransformer,
+    IPAddressEncoderTransformer,
+    PhoneTransformer,
+    StringSimilarityTransformer,
+    StringSlicerTransformer,
+)

src/sk_transformers/datetime_transformer.py

@@ -13,7 +13,7 @@ class DurationCalculatorTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.datetime_transformer import DurationCalculatorTransformer
+    from sk_transformers import DurationCalculatorTransformer
 
     X = pd.DataFrame(
         {
@@ -79,7 +79,7 @@ class TimestampTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.datetime_transformer import TimestampTransformer
+    from sk_transformers import TimestampTransformer
 
     X = pd.DataFrame({"foo": ["1960-01-01", "1970-01-01", "1990-01-01"]})
     transformer = TimestampTransformer(["foo"])

src/sk_transformers/deep_transformer.py

@@ -27,7 +27,7 @@ class ToVecTransformer(BaseEstimator, TransformerMixin):
     import numpy as np
     import pandas as pd
     from pytorch_widedeep.datasets import load_adult
-    from sk_transformers.deep_transformer import ToVecTransformer
+    from sk_transformers import ToVecTransformer
 
     df = load_adult(as_frame=True)
     df["target"] = (df["income"].apply(lambda x: ">50K" in x)).astype(int)

src/sk_transformers/encoder_transformer.py

@@ -12,7 +12,7 @@ class MeanEncoderTransformer(BaseEstimator, TransformerMixin):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.encoder_transformer import MeanEncoderTransformer
+    from sk_transformers import MeanEncoderTransformer
 
     X = pd.DataFrame({"foo": ["a", "b", "a", "c", "b", "a", "c", "a", "b", "c"]})
     y = pd.Series([1, 0, 1, 0, 1, 0, 1, 0, 1, 0])

src/sk_transformers/generic_transformer.py

@@ -16,7 +16,7 @@ class DtypeTransformer(BaseTransformer):
     ```python
     import numpy as np
     import pandas as pd
-    from sk_transformers.generic_transformer import DtypeTransformer
+    from sk_transformers import DtypeTransformer
 
     X = pd.DataFrame({"foo": [1, 2, 3], "bar": ["a", "a", "b"]})
     transformer = DtypeTransformer([("foo", np.float32), ("bar", "category")])
@@ -69,7 +69,7 @@ class AggregateTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.generic_transformer import AggregateTransformer
+    from sk_transformers import AggregateTransformer
 
     X = pd.DataFrame(
         {
@@ -161,7 +161,7 @@ class FunctionsTransformer(BaseTransformer):
     ```python
     import numpy as np
     import pandas as pd
-    from sk_transformers.generic_transformer import FunctionsTransformer
+    from sk_transformers import FunctionsTransformer
 
     X = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
     transformer = FunctionsTransformer([("foo", np.log1p, None), ("bar", np.sqrt, None)])
@@ -216,7 +216,7 @@ class MapTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.generic_transformer import MapTransformer
+    from sk_transformers import MapTransformer
 
     X = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
     transformer = MapTransformer([("foo", lambda x: x + 1)])
@@ -263,7 +263,7 @@ class ColumnDropperTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.generic_transformer import ColumnDropperTransformer
+    from sk_transformers import ColumnDropperTransformer
 
     X = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
     transformer = ColumnDropperTransformer(["foo"])
@@ -304,7 +304,7 @@ class NaNTransformer(BaseTransformer):
 
     Example:
     ```python
-    from sk_transformers.generic_transformer import NaNTransformer
+    from sk_transformers import NaNTransformer
     import pandas as pd
     import numpy as np
 
@@ -357,7 +357,7 @@ class ValueIndicatorTransformer(BaseTransformer):
 
     Example:
     ```python
-    from sk_transformers.generic_transformer import ValueIndicatorTransformer
+    from sk_transformers import ValueIndicatorTransformer
     import pandas as pd
 
     X = pd.DataFrame({"foo": [1, -999, 3], "bar": ["a", "-999", "c"]})
@@ -410,14 +410,12 @@ class QueryTransformer(BaseTransformer):
     """Applies a list of queries to a dataframe. If it operates on a dataset
     used for supervised learning this transformer should be applied on the
     dataframe containing `X` and `y`. So removing of columns by queries also
-    removes the corresponding `y` value. Read more about queries [here](https:/
-
-    /pandas.pydata.org/docs/reference/api/pandas.DataFrame.query.html).
+    removes the corresponding `y` value.
 
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.generic_transformer import QueryTransformer
+    from sk_transformers import QueryTransformer
 
     X = pd.DataFrame({"foo": [1, 8, 3, 6, 5, 4, 7, 2]})
     transformer = QueryTransformer(["foo > 4"])
@@ -468,7 +466,7 @@ class ValueReplacerTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.generic_transformer import ValueReplacerTransformer
+    from sk_transformers import ValueReplacerTransformer
 
     X = pd.DataFrame(
         {"foo": ["0000-01-01", "2022/01/08", "bar", "1982-12-7", "28-09-2022"]}
@@ -554,7 +552,7 @@ class LeftJoinTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.generic_transformer import LeftJoinTransformer
+    from sk_transformers import LeftJoinTransformer
 
     X = pd.DataFrame({"foo": ["A", "B", "C", "A", "C"]})
     lookup_df = pd.Series([1, 2, 3], index=["A", "B", "C"], name="values")

src/sk_transformers/number_transformer.py

@@ -20,7 +20,7 @@ class MathExpressionTransformer(BaseTransformer):
     Example:
     ```python
     import pandas as pd
-    from sk_transformers.number_transformer import MathExpressionTransformer
+    from sk_transformers import MathExpressionTransformer
 
     X = pd.DataFrame({"foo": [1, 2, 3], "bar": [4, 5, 6]})
     transformer = MathExpressionTransformer([("foo", "np.sum", "bar", {"axis": 0})])