diff --git a/02_pandas.ipynb b/02_pandas.ipynb
index 1ec7bc8eb5aa4b8c09db6607286138a625d1c8c7..5fd57fb668d407df91f84176a01b9c3e94ef2cf6 100644
--- a/02_pandas.ipynb
+++ b/02_pandas.ipynb
@@ -522,7 +522,7 @@
"source": [
"* `Series` represents a single column/vector/data series.\n",
"* `DataFrame` consists of multiple `Series` (is a dictionary/list of row/column `Series`).\n",
- "* `GroupBy` is created by grouping `DataFrame` rows by values of one of the column `Series`."
+ "* `DataFrameGroupBy` is created by grouping `DataFrame` rows by values of one of the column `Series`."
]
},
{
@@ -582,13 +582,6 @@
"execution_count": 13,
"metadata": {},
"outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "<pandas.core.groupby.generic.DataFrameGroupBy object at 0x110573e80>\n"
- ]
- },
{
"data": {
"text/html": [
@@ -720,8 +713,6 @@
],
"source": [
"df_bygender = df[[\"Gender\", \"Weight\", \"Height\"]].groupby(\"Gender\")\n",
- "# it's lazy!\n",
- "print(df_bygender)\n",
"df_bygender.describe()"
]
},
@@ -736,13 +727,47 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "See [**Data Wrangling with pandas Cheat Sheet**](https://pandas.pydata.org/Pandas_Cheat_Sheet.pdf) for a great visual overview of ways of how tidy data frames can be transformed and summarized."
+ "See [**Data Wrangling with pandas Cheat Sheet**](https://pandas.pydata.org/Pandas_Cheat_Sheet.pdf) for a great visual overview of ways of how tidy data frames can be transformed and summarized.\n",
+ "\n",
+ "Notably, transfromations are lazy - they won't execute until actually needed."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Grouping"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "<pandas.core.groupby.generic.DataFrameGroupBy object at 0x11436fa20>\n"
+ ]
+ }
+ ],
+ "source": [
+ "df_bygender = df.groupby(\"Gender\")\n",
+ "print(df_bygender)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Compare mean brain weight with mean weight per gender:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
"outputs": [
{
"name": "stdout",
@@ -760,7 +785,6 @@
"source": [
"print(df.Weight.mean())\n",
"print()\n",
- "df_bygender = df.groupby(\"Gender\")\n",
"print(df_bygender.Weight.mean())"
]
},
@@ -768,7 +792,55 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "## Plotting directly"
+ "Grouping is simply a collection of splitted data frames:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Female\n",
+ " FSIQ VIQ PIQ Weight Height MRI_Count\n",
+ "count 20.000000 20.000000 20.000000 20.000000 20.000000 20.00000\n",
+ "mean 111.900000 109.450000 110.450000 137.200000 65.765000 862654.60000\n",
+ "std 23.686327 21.670924 21.946046 16.953807 2.288248 55893.55578\n",
+ "min 77.000000 71.000000 72.000000 106.000000 62.000000 790619.00000\n",
+ "25% 90.250000 90.000000 93.000000 125.750000 64.500000 828062.00000\n",
+ "50% 115.500000 116.000000 115.000000 138.500000 66.000000 855365.00000\n",
+ "75% 133.000000 129.000000 128.750000 146.250000 66.875000 882668.50000\n",
+ "max 140.000000 136.000000 147.000000 175.000000 70.500000 991305.00000\n",
+ "\n",
+ "Male\n",
+ " FSIQ VIQ PIQ Weight Height MRI_Count\n",
+ "count 20.000000 20.000000 20.000000 18.000000 19.000000 2.000000e+01\n",
+ "mean 115.000000 115.250000 111.600000 166.444444 71.431579 9.548554e+05\n",
+ "std 24.986312 25.640993 23.540335 20.047656 3.283131 5.591135e+04\n",
+ "min 80.000000 77.000000 74.000000 132.000000 66.300000 8.799870e+05\n",
+ "25% 89.750000 95.250000 86.000000 148.750000 68.900000 9.195292e+05\n",
+ "50% 118.000000 110.500000 117.000000 172.000000 70.500000 9.472415e+05\n",
+ "75% 139.250000 145.000000 128.000000 180.750000 73.750000 9.734960e+05\n",
+ "max 144.000000 150.000000 150.000000 192.000000 77.000000 1.079549e+06\n",
+ "\n"
+ ]
+ }
+ ],
+ "source": [
+ "for gender, gender_df in df_bygender:\n",
+ " print(gender)\n",
+ " print(gender_df.describe())\n",
+ " print()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Plotting"
]
},
{
@@ -780,16 +852,16 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 17,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
- "<matplotlib.axes._subplots.AxesSubplot at 0x1105732e8>"
+ "<matplotlib.axes._subplots.AxesSubplot at 0x11446bd30>"
]
},
- "execution_count": 15,
+ "execution_count": 17,
"metadata": {},
"output_type": "execute_result"
},
@@ -823,7 +895,7 @@
},
{
"cell_type": "code",
- "execution_count": 16,
+ "execution_count": 18,
"metadata": {},
"outputs": [
{
@@ -887,12 +959,12 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- ".. or directly for the `GroupBy` objects:"
+ ".. or directly on the `DataFrameGroupBy` objects:"
]
},
{
"cell_type": "code",
- "execution_count": 17,
+ "execution_count": 19,
"metadata": {},
"outputs": [
{
@@ -903,7 +975,7 @@
"dtype: object"
]
},
- "execution_count": 17,
+ "execution_count": 19,
"metadata": {},
"output_type": "execute_result"
},