# !pip install pandas lxml beautifulsoup4 html5lib matplotlib -U
Requirement already up-to-date: pandas in /opt/venv/lib/python3.7/site-packages (1.0.5)
Requirement already up-to-date: lxml in /opt/venv/lib/python3.7/site-packages (4.5.2)
Requirement already up-to-date: beautifulsoup4 in /opt/venv/lib/python3.7/site-packages (4.9.1)
Requirement already up-to-date: html5lib in /opt/venv/lib/python3.7/site-packages (1.1)
Requirement already up-to-date: matplotlib in /opt/venv/lib/python3.7/site-packages (3.3.0)
Requirement already satisfied, skipping upgrade: numpy>=1.13.3 in /opt/venv/lib/python3.7/site-packages (from pandas) (1.19.0)
Requirement already satisfied, skipping upgrade: python-dateutil>=2.6.1 in /opt/venv/lib/python3.7/site-packages (from pandas) (2.8.1)
Requirement already satisfied, skipping upgrade: pytz>=2017.2 in /opt/venv/lib/python3.7/site-packages (from pandas) (2020.1)
Requirement already satisfied, skipping upgrade: soupsieve>1.2 in /opt/venv/lib/python3.7/site-packages (from beautifulsoup4) (2.0.1)
Requirement already satisfied, skipping upgrade: webencodings in /opt/venv/lib/python3.7/site-packages (from html5lib) (0.5.1)
Requirement already satisfied, skipping upgrade: six>=1.9 in /opt/venv/lib/python3.7/site-packages (from html5lib) (1.15.0)
Requirement already satisfied, skipping upgrade: pillow>=6.2.0 in /opt/venv/lib/python3.7/site-packages (from matplotlib) (7.2.0)
Requirement already satisfied, skipping upgrade: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.3 in /opt/venv/lib/python3.7/site-packages (from matplotlib) (2.4.7)
Requirement already satisfied, skipping upgrade: kiwisolver>=1.0.1 in /opt/venv/lib/python3.7/site-packages (from matplotlib) (1.2.0)
Requirement already satisfied, skipping upgrade: cycler>=0.10 in /opt/venv/lib/python3.7/site-packages (from matplotlib) (0.10.0)
import sys
import pandas as pd
print(f"Python version {sys.version}")
print(f"pandas version: {pd.__version__}")
Python version 3.7.3 (default, Jun 11 2019, 01:11:15)
[GCC 6.3.0 20170516]
pandas version: 1.0.5
list_of_dfs = pd.read_html('https://www.ussoccer.com/uswnt-stats')
type(list_of_dfs)
len(list_of_dfs)
list_of_dfs[0]
| 0object Name3.7% Dahlkemper, Abby3.7% 25 others92.6% | 1object D29.6% 4 others59.3% Missing11.1% | 2object 825.9% 6 others70.4% Missing3.7% | 3object 518.5% 8 others77.8% Missing3.7% | 4object 9011.1% 21 others85.2% Missing3.7% | 5object 051.9% 214.8% 6 others33.3% | 6object 048.1% 6 others48.1% Missing3.7% | 7object 077.8% 4 others18.5% Missing3.7% | 8object 07.4% R3.7% Missing88.9% | 9object 1687.4% 20 others81.5% Missing11.1% | 10object 018.5% 14 others59.3% Missing22.2% | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
0 | Name | Pos. | GP | GS | Min | G | A | Y | R | Caps | Goals | |
1 | Dahlkemper, Abby | D | 8 | 7 | 658 | 0 | 0 | 0 | nan | 61 | 0 | |
2 | Davidson, Tierna | D | 1 | 1 | 45 | 0 | 0 | 0 | nan | 26 | 1 | |
3 | Dunn, Crystal | D | 8 | 8 | 627 | 0 | 0 | 0 | nan | 104 | 24 | |
4 | Ertz, Julie | M | 7 | 7 | 604 | 1 | 1 | 0 | nan | 102 | 20 | |
| Expand rows 5 - 21 | ||||||||||||
22 | Sullivan, Andi | M | 1 | 1 | 90 | 0 | 0 | 0 | nan | 16 | 0 | |
23 | Williams, Lynn | F | 7 | 4 | 396 | 3 | 5 | 0 | nan | 28 | 9 | |
24 | Own Goal | nan | nan | nan | nan | 0 | nan | nan | nan | nan | nan | |
25 | TOTAL | nan | 8 | 8 | 720 | 31 | 24 | 2 | 0 | nan | nan | |
26 | Opponents | nan | 8 | 8 | 720 | 1 | 0 | 4 | 0 | nan | nan | |
| 27 rows × 11 columns | ||||||||||||
list_of_dfs[1]
| 0object | 1object | 2object | 3object | 4object | 5object | 6object | 7object | 8object | 9object | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
0 | Player Name | GP | GS | Min. | GF | GA | SO* | GAA | W-L-T | Caps/SO | |
1 | Franch, Adrianna | 1 | 1 | 90 | 3 | 1 | 0 | 1 | 1-0-0 | 4/1 | |
2 | Harris, Ashlyn | 1 | 1 | 90 | 8 | 0 | 1 | 0 | 1-0-0 | 25/9 | |
3 | Naeher, Alyssa | 6 | 6 | 540 | 20 | 0 | 6 | 0 | 6-0-0 | 63/36 | |
4 | TOTAL | 8 | 8 | 720 | 31 | 1 | 6 | 0.13 | 8-0-0 | nan | |
5 | Opponents | 8 | 8 | 720 | 1 | 31 | 0 | 3.88 | 0-8-0 | nan | |
| 6 rows × 10 columns | |||||||||||
list_of_dfs_nicer = pd.read_html('https://www.ussoccer.com/uswnt-stats', header=0, index_col=0)
list_of_dfs_nicer[0].head(2)
| Pos.object | GPfloat64 | GSfloat64 | Minfloat64 | Gint64 | Afloat64 | Yfloat64 | Rfloat64 | Capsfloat64 | Goalsfloat64 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
Dahlkemper, Abby | D | 8 | 7 | 658 | 0 | 0 | 0 | nan | 61 | 0 | |
Davidson, Tierna | D | 1 | 1 | 45 | 0 | 0 | 0 | nan | 26 | 1 | |
| 2 rows × 10 columns | |||||||||||
runners_df = list_of_dfs_nicer[0]
goalies_df = list_of_dfs_nicer[1]
runners_df = runners_df.iloc[:-3]
runners_df.tail(2)
| Pos.object | GPfloat64 | GSfloat64 | Minfloat64 | Gint64 | Afloat64 | Yfloat64 | Rfloat64 | Capsfloat64 | Goalsfloat64 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
Sullivan, Andi | M | 1 | 1 | 90 | 0 | 0 | 0 | nan | 16 | 0 | |
Williams, Lynn | F | 7 | 4 | 396 | 3 | 5 | 0 | nan | 28 | 9 | |
| 2 rows × 10 columns | |||||||||||
goalies_df = goalies_df.iloc[:-2]
goalies_df
| GPint64 | GSint64 | Min.int64 | GFint64 | GAint64 | SO*int64 | GAAfloat64 | W-L-Tobject | Caps/SOobject | ||
|---|---|---|---|---|---|---|---|---|---|---|
Franch, Adrianna | 1 | 1 | 90 | 3 | 1 | 0 | 1 | 1-0-0 | 4/1 | |
Harris, Ashlyn | 1 | 1 | 90 | 8 | 0 | 1 | 0 | 1-0-0 | 25/9 | |
Naeher, Alyssa | 6 | 6 | 540 | 20 | 0 | 6 | 0 | 6-0-0 | 63/36 | |
| 3 rows × 9 columns | ||||||||||
goal_scorers_df = runners_df[runners_df['G']>0]
goal_scorers_df['G'].sort_values().plot(kind='barh', title='2020 USWNT Goals');
weightlifting_df_list = pd.read_html('https://en.wikipedia.org/wiki/List_of_weight_training_exercises', index_col=0)
len(weightlifting_df_list)
weightlifting_df_list[0]
| Calvesobject Some14.8% Yes7.4% Missing77.8% | Quad-ricepsobject Yes18.5% Missing81.5% | Ham-stringsobject Some11.1% Yes11.1% Missing77.8% | Gluteusobject Yes18.5% Missing81.5% | Hipsotherobject Yes18.5% Missing81.5% | Lowerbackobject Yes7.4% Some3.7% Missing88.9% | Latsobject Yes11.1% Missing88.9% | Trapeziusobject Some22.2% Yes7.4% Missing70.4% | Abdominalsobject Some14.8% Yes11.1% Missing74.1% | Pectoralsobject Yes11.1% Some7.4% Missing81.5% | Deltoidsobject Some22.2% Yes14.8% Missing63% | Tricepsobject Yes14.8% Some3.7% Missing81.5% | Bicepsobject Some14.8% Yes3.7% Missing81.5% | Forearmsobject Some33.3% Missing66.7% | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Squat | Some | Yes | Some | Yes | Yes | Some | nan | nan | Yes | nan | nan | nan | nan | nan | |
Leg press | Some | Yes | Some | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Lunge | nan | Yes | Yes | Yes | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Deadlift | Some | Yes | Yes | Yes | Yes | Yes | nan | Some | Some | nan | nan | nan | nan | Some | |
Leg extension | nan | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
| Expand rows 5 - 21 | |||||||||||||||
Biceps curl | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | Yes | Some | |
Crunch | nan | nan | nan | nan | nan | nan | nan | nan | Yes | nan | nan | nan | nan | nan | |
Russian twist | nan | nan | nan | nan | nan | nan | nan | nan | Yes | nan | nan | nan | nan | nan | |
Leg raise | nan | nan | nan | nan | Yes | nan | nan | nan | Some | nan | nan | nan | nan | nan | |
Back extension | nan | nan | Some | Yes | nan | Yes | nan | nan | nan | nan | nan | nan | nan | nan | |
| 27 rows × 14 columns | |||||||||||||||
exercises_df = weightlifting_df_list[0]
hammies = exercises_df[(exercises_df['Ham-strings']=='Yes') | (exercises_df['Ham-strings']=='Some')]
hammies
| Calvesobject | Quad-ricepsobject | Ham-stringsobject | Gluteusobject | Hipsotherobject | Lowerbackobject | Latsobject | Trapeziusobject | Abdominalsobject | Pectoralsobject | Deltoidsobject | Tricepsobject | Bicepsobject | Forearmsobject | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Squat | Some | Yes | Some | Yes | Yes | Some | nan | nan | Yes | nan | nan | nan | nan | nan | |
Leg press | Some | Yes | Some | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Lunge | nan | Yes | Yes | Yes | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Deadlift | Some | Yes | Yes | Yes | Yes | Yes | nan | Some | Some | nan | nan | nan | nan | Some | |
Leg curl | Some | nan | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Back extension | nan | nan | Some | Yes | nan | Yes | nan | nan | nan | nan | nan | nan | nan | nan | |
| 6 rows × 14 columns | |||||||||||||||
hammies.sort_values(by='Ham-strings', ascending=False)
| Calvesobject | Quad-ricepsobject | Ham-stringsobject | Gluteusobject | Hipsotherobject | Lowerbackobject | Latsobject | Trapeziusobject | Abdominalsobject | Pectoralsobject | Deltoidsobject | Tricepsobject | Bicepsobject | Forearmsobject | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Lunge | nan | Yes | Yes | Yes | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Deadlift | Some | Yes | Yes | Yes | Yes | Yes | nan | Some | Some | nan | nan | nan | nan | Some | |
Leg curl | Some | nan | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Squat | Some | Yes | Some | Yes | Yes | Some | nan | nan | Yes | nan | nan | nan | nan | nan | |
Leg press | Some | Yes | Some | Yes | nan | nan | nan | nan | nan | nan | nan | nan | nan | nan | |
Back extension | nan | nan | Some | Yes | nan | Yes | nan | nan | nan | nan | nan | nan | nan | nan | |
| 6 rows × 14 columns | |||||||||||||||