
pip install nba_api

Collecting nba_api
  Downloading nba_api-1.1.11.tar.gz (125 kB)
Requirement already satisfied: requests in c:\users\vytis\anaconda3\lib\site-packages (from nba_api) (2.26.0)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\users\vytis\anaconda3\lib\site-packages (from requests->nba_api) (1.26.7)
Requirement already satisfied: charset-normalizer~=2.0.0 in c:\users\vytis\anaconda3\lib\site-packages (from requests->nba_api) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in c:\users\vytis\anaconda3\lib\site-packages (from requests->nba_api) (3.2)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\vytis\anaconda3\lib\site-packages (from requests->nba_api) (2021.10.8)
Building wheels for collected packages: nba-api
  Building wheel for nba-api (setup.py): started
  Building wheel for nba-api (setup.py): finished with status 'done'
  Created wheel for nba-api: filename=nba_api-1.1.11-py3-none-any.whl size=251504 sha256=bdad5891ebe83844172f6272cd65203f4d87590ca345f6aee30557126dcc3b19
  Stored in directory: c:\users\vytis\appdata\local\pip\cache\wheels\66\c2\3b\c87a243f9e5d2449e7f2c7bd65de4a6b5ce9a24b33978398a7
Successfully built nba-api
Installing collected packages: nba-api
Successfully installed nba-api-1.1.11
Note: you may need to restart the kernel to use updated packages.


import numpy as np
from nba_api.stats.endpoints import shotchartdetail
from nba_api.stats.endpoints import commonallplayers
from nba_api.stats.endpoints import playercareerstats
from nba_api.stats.endpoints import teamyearbyyearstats
from nba_api.stats.static import teams
from nba_api.stats.endpoints import playerawards
import time
import json
import requests
import pandas as pd
import matplotlib as mpl
import matplotlib.pyplot as plt
import pickle
import seaborn as sns
from tqdm import tqdm


players_stats = pd.read_pickle('C:\\Users\\Vytis\\player_stats.pkl')


players_stats


players_stats.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17974 entries, 0 to 17973
Data columns (total 27 columns):
 #   Column             Non-Null Count  Dtype  
---  ------             --------------  -----  
 0   PLAYER_ID          17974 non-null  object 
 1   SEASON_ID          17974 non-null  object 
 2   LEAGUE_ID          17974 non-null  object 
 3   TEAM_ID            17974 non-null  object 
 4   TEAM_ABBREVIATION  17974 non-null  object 
 5   PLAYER_AGE         17974 non-null  float64
 6   GP                 17974 non-null  object 
 7   GS                 17974 non-null  object 
 8   MIN                17974 non-null  float64
 9   FGM                17974 non-null  object 
 10  FGA                17974 non-null  object 
 11  FG_PCT             17974 non-null  float64
 12  FG3M               17966 non-null  object 
 13  FG3A               17966 non-null  object 
 14  FG3_PCT            17966 non-null  float64
 15  FTM                17974 non-null  object 
 16  FTA                17974 non-null  object 
 17  FT_PCT             17974 non-null  float64
 18  OREB               17974 non-null  object 
 19  DREB               17974 non-null  object 
 20  REB                17974 non-null  object 
 21  AST                17974 non-null  object 
 22  STL                17974 non-null  object 
 23  BLK                17974 non-null  object 
 24  TOV                17973 non-null  object 
 25  PF                 17974 non-null  object 
 26  PTS                17974 non-null  object 
dtypes: float64(5), object(22)
memory usage: 3.7+ MB


# Man svarbu FG3M, FG3A, FG3_PCT


players_stats['FG3M']

0        24
1        31
2        70
3        42
4        83
         ..
17969     0
17970     0
17971     0
17972     1
17973     0
Name: FG3M, Length: 17974, dtype: object


players_stats['FG3A']

0        100
1         94
2        197
3        133
4        215
        ... 
17969      0
17970      0
17971      2
17972      4
17973      0
Name: FG3A, Length: 17974, dtype: object


players_stats['FG3_PCT']

0        0.240
1        0.330
2        0.355
3        0.316
4        0.386
         ...  
17969    0.000
17970    0.000
17971    0.000
17972    0.250
17973    0.000
Name: FG3_PCT, Length: 17974, dtype: float64


players_stats['SEASON_ID'] = players_stats['SEASON_ID'].map(lambda x: int(x.split("-",1)[0]))


players_stats['SEASON_ID']

0        1990
1        1991
2        1992
3        1993
4        1994
         ... 
17969    2018
17970    2018
17971    2019
17972    2020
17973    2021
Name: SEASON_ID, Length: 17974, dtype: int64


# Prisijungsiu krepšininkų vardus, kad nebūtų tik ID


all_players = commonallplayers.CommonAllPlayers().get_data_frames()[0]
all_players


all_players.rename(columns={'PERSON_ID':'PLAYER_ID'}, inplace=True)


def get_player_id(first, last):
  for player in players:
    if player['firstName'] == first and player['lastName'] == last:
      return player['playerId']
  return -1


get_player_id('Jonas', 'Valanciunas')

202685


all_players[all_players['PLAYER_ID'] == get_player_id('Jonas', 'Valanciunas')]


players_stats[players_stats['PLAYER_ID'] == 2557]


all_players.columns

Index(['PLAYER_ID', 'DISPLAY_LAST_COMMA_FIRST', 'DISPLAY_FIRST_LAST',
       'ROSTERSTATUS', 'FROM_YEAR', 'TO_YEAR', 'PLAYERCODE', 'PLAYER_SLUG',
       'TEAM_ID', 'TEAM_CITY', 'TEAM_NAME', 'TEAM_ABBREVIATION', 'TEAM_CODE',
       'TEAM_SLUG', 'GAMES_PLAYED_FLAG', 'OTHERLEAGUE_EXPERIENCE_CH'],
      dtype='object')


all_players_columns = ['DISPLAY_LAST_COMMA_FIRST',
       'ROSTERSTATUS', 'FROM_YEAR', 'TO_YEAR', 'PLAYERCODE', 'PLAYER_SLUG',
       'TEAM_ID', 'TEAM_ABBREVIATION', 'TEAM_CODE',
       'TEAM_SLUG', 'GAMES_PLAYED_FLAG', 'OTHERLEAGUE_EXPERIENCE_CH']


all_players_to_merge = all_players.drop(all_players_columns, axis=1)


all_players_to_merge


players_stats = pd.merge(players_stats, all_players_to_merge, on=['PLAYER_ID'])
players_stats


players_stats['SEASON_ID'] = players_stats['SEASON_ID'].map(lambda x: int(x.split("-",1)[0]))

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_21052/3326492598.py in <module>
----> 1 players_stats['SEASON_ID'] = players_stats['SEASON_ID'].map(lambda x: int(x.split("-",1)[0]))
      2 players_stats

~\anaconda3\lib\site-packages\pandas\core\series.py in map(self, arg, na_action)
   4159         dtype: object
   4160         """
-> 4161         new_values = super()._map_values(arg, na_action=na_action)
   4162         return self._constructor(new_values, index=self.index).__finalize__(
   4163             self, method="map"

~\anaconda3\lib\site-packages\pandas\core\base.py in _map_values(self, mapper, na_action)
    868 
    869         # mapper is a function
--> 870         new_values = map_f(values, mapper)
    871 
    872         return new_values

~\anaconda3\lib\site-packages\pandas\_libs\lib.pyx in pandas._libs.lib.map_infer()

~\AppData\Local\Temp/ipykernel_21052/3326492598.py in <lambda>(x)
----> 1 players_stats['SEASON_ID'] = players_stats['SEASON_ID'].map(lambda x: int(x.split("-",1)[0]))
      2 players_stats

AttributeError: 'int' object has no attribute 'split'


players_stats


#sezonai po 2012-2021


players_stats_last10 = player_stats[player_stats['SEASON_ID'] > 2011]


#sezonai 2002 - 2011


players_stats_before10 = player_stats[(player_stats['SEASON_ID'] > 2001) & (player_stats['SEASON_ID'] < 2012)]


players_stats_before10


players_stats_last10['FG3A'].sum()

683631


players_stats_before10['FG3A'].sum()

447277


#Jei užtektų trumpo atsakymo, tai taip


players_stats_last10['FG3A'].sum() > players_stats_before10['FG3A'].sum()

True


#Tačiau svarbu pamatyti tendencijas, kaip keitėsi tritaškių kiekis per tuos metus


# Toliau tvarkausi duomenų paketą...


players_stats.drop(['LEAGUE_ID', 'DISPLAY_FIRST_LAST_x', 'TEAM_CITY_x', 'TEAM_NAME_x'], axis=1, inplace=True)


players_stats.rename(columns={'DISPLAY_FIRST_LAST_y':'PLAYER_NAME'}, inplace=True)


players_stats = players_stats.set_index(['PLAYER_ID'])


# Pastebėjau, kad yra TOT reikšmės, kurios iškreipia duomenis - TOTAL per season turbūt rodo.


players_stats = players_stats[players_stats.TEAM_ABBREVIATION != 'TOT']


players_stats[players_stats['TEAM_ABBREVIATION'] == 'TOT']


players_stats = players_stats.dropna()


#pasidariau susiaurinta variantą, nuo 2000 m.


players_stats1 = players_stats[players_stats['SEASON_ID'] > 2000]


data1 = pd.DataFrame(players_stats1.groupby('SEASON_ID')['FG3A'].sum())
data1


data1.plot.bar()
plt.title('Išmesti tritaškiai')

Text(0.5, 1.0, 'Išmesti tritaškiai')


# Matome yra didelių nuokrypių, nes dar paskutinis sezonas nepasibaiges, o ir ne visus sezonus vienodas varzybu skaicius buvo


players_stats1 = players_stats1[(players_stats1['FG3A'] > 100)] #kad nebūtų labai kraštutinių duomenų


players_stats1.groupby('PLAYER_NAME')['FG3_PCT'].mean().sort_values(ascending=False).head(10)

PLAYER_NAME
Fred Hoiberg          0.462500
Hubert Davis          0.452000
Pau Gasol             0.448000
Jason Kapono          0.441000
Seth Curry            0.439167
Wesley Person         0.438500
Joe Harris            0.435400
Steve Novak           0.434750
Michael Porter Jr.    0.433500
Stephen Curry         0.433333
Name: FG3_PCT, dtype: float64


data2= pd.DataFrame(players_stats1.groupby('PLAYER_NAME')['FG3_PCT'].mean().sort_values(ascending=False).head(10))
data2


data2.plot.bar()
plt.title('Tritaškių pataikymas')

Text(0.5, 1.0, 'Tritaškių pataikymas')


players_stats2 = players_stats[(players_stats['FG3A'] > 100)]


data3 = pd.DataFrame(players_stats2.groupby('PLAYER_NAME')['FG3M'].sum().sort_values(ascending=False).head(10))
data3


data3.plot.bar()
plt.title('Įmesti tritaškiai')

Text(0.5, 1.0, 'Įmesti tritaškiai')


players_stats2['FG3M'].max()

402


players_stats2[players_stats2['FG3M'] == 402] #taip pat ir per vieną sezoną daugiausiai įmetė 3pt


# Curry dar praeitų metų gale pralenkė R. Allen ir tapo tritaškių rekordininku


data4 = pd.DataFrame(players_stats2.groupby('PLAYER_NAME')['FG3A'].sum().sort_values(ascending=False).head(10))
data4


data4.plot.bar()
plt.title('Išmesti tritaškiai')

Text(0.5, 1.0, 'Išmesti tritaškiai')


players_stats


players_stats_test = players_stats[(players_stats['FG3A'] > 100)]


players_stats_test.plot.scatter(x='PLAYER_AGE', y='FG3_PCT')

*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2D array with a single row if you intend to specify the same RGB or RGBA value for all points.

<AxesSubplot:xlabel='PLAYER_AGE', ylabel='FG3_PCT'>


X = players_stats_test[['PLAYER_AGE']]
X


y = players_stats_test['FG3_PCT']


model = LinearRegression()
model

LinearRegression()


model.fit(X, y)

LinearRegression()


prediction = model.predict(X)
prediction

array([0.35420242, 0.35598891, 0.35777539, ..., 0.35241594, 0.35420242,
       0.35420242])


players_stats_test['prediction'] = prediction
players_stats_test[['PLAYER_AGE', 'FG3_PCT', 'prediction']]

C:\Users\Vytis\AppData\Local\Temp/ipykernel_21052/1002346078.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  players_stats_test['prediction'] = prediction


plt.scatter(X, y)
plt.plot(X, prediction, 'r')

[<matplotlib.lines.Line2D at 0x204e9458640>]


r_squared = model.score(X, y)
r_squared

0.026042196049113397


#labai prastai?


steph = players_stats[players_stats['PLAYER_NAME'] == 'Stephen Curry'][['PLAYER_AGE', 'FG_PCT']]


steph


steph.plot.scatter(x='PLAYER_AGE', y='FG_PCT')

*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2D array with a single row if you intend to specify the same RGB or RGBA value for all points.

<AxesSubplot:xlabel='PLAYER_AGE', ylabel='FG_PCT'>


from sklearn.linear_model import LinearRegression


X = steph[['PLAYER_AGE']]
X


y = steph['FG_PCT']
y

PLAYER_ID
201939    0.462
201939    0.480
201939    0.490
201939    0.451
201939    0.471
201939    0.487
201939    0.504
201939    0.468
201939    0.495
201939    0.472
201939    0.402
201939    0.482
201939    0.434
Name: FG_PCT, dtype: float64


model = LinearRegression()
model

LinearRegression()


model.fit(X, y)

LinearRegression()


model.predict([[50]])

C:\Users\Vytis\anaconda3\lib\site-packages\sklearn\base.py:450: UserWarning: X does not have valid feature names, but LinearRegression was fitted with feature names
  warnings.warn(

array([0.41961746])


prediction = model.predict(X)
prediction

array([0.48234653, 0.48010621, 0.47786589, 0.47562556, 0.47338524,
       0.47114491, 0.46890459, 0.46666427, 0.46442394, 0.46218362,
       0.45994329, 0.45770297, 0.45770297])


steph['prediction'] = prediction
steph


plt.scatter(X, y)
plt.plot(X, prediction, 'r')

[<matplotlib.lines.Line2D at 0x204e80accd0>]


r_squared = model.score(X, y)
r_squared

0.09472000198151331


#labai netikslu, tai reiškia nepriklauso


steph.plot.scatter('PLAYER_AGE', 'FG_PCT')

*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2D array with a single row if you intend to specify the same RGB or RGBA value for all points.

<AxesSubplot:xlabel='PLAYER_AGE', ylabel='FG_PCT'>


players_stats1['FG3A_per_game'] = players_stats1['FG3A']/players_stats1['GP']

C:\Users\Vytis\AppData\Local\Temp/ipykernel_21052/205737034.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  players_stats1['FG3A_per_game'] = players_stats1['FG3A']/players_stats1['GP']


players_stats['FG3A_per_game']

PLAYER_ID
203518     3.632353
203518     2.946667
203518     4.096774
1630173    0.016393
1630173    1.208333
             ...   
1627826         0.0
1627826         0.0
1627826    0.027778
1627826    0.055556
1627826         0.0
Name: FG3A_per_game, Length: 6888, dtype: object


players_stats.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 17974 entries, 51 to 1627826
Data columns (total 28 columns):
 #   Column             Non-Null Count  Dtype  
---  ------             --------------  -----  
 0   SEASON_ID          17974 non-null  int64  
 1   TEAM_ID            17974 non-null  object 
 2   TEAM_ABBREVIATION  17974 non-null  object 
 3   PLAYER_AGE         17974 non-null  float64
 4   GP                 17974 non-null  object 
 5   GS                 17974 non-null  object 
 6   MIN                17974 non-null  float64
 7   FGM                17974 non-null  object 
 8   FGA                17974 non-null  object 
 9   FG_PCT             17974 non-null  float64
 10  FG3M               17966 non-null  object 
 11  FG3A               17966 non-null  object 
 12  FG3_PCT            17966 non-null  float64
 13  FTM                17974 non-null  object 
 14  FTA                17974 non-null  object 
 15  FT_PCT             17974 non-null  float64
 16  OREB               17974 non-null  object 
 17  DREB               17974 non-null  object 
 18  REB                17974 non-null  object 
 19  AST                17974 non-null  object 
 20  STL                17974 non-null  object 
 21  BLK                17974 non-null  object 
 22  TOV                17973 non-null  object 
 23  PF                 17974 non-null  object 
 24  PTS                17974 non-null  object 
 25  PLAYER_NAME        17974 non-null  object 
 26  TEAM_CITY_y        17974 non-null  object 
 27  TEAM_NAME_y        17974 non-null  object 
dtypes: float64(5), int64(1), object(22)
memory usage: 4.5+ MB


players_stats['FG3A'].dropna().astype(int)

PLAYER_ID
51         100
51          94
51         197
51         133
51         215
          ... 
1627826      0
1627826      0
1627826      2
1627826      4
1627826      0
Name: FG3A, Length: 17966, dtype: int32


players_stats['GP'].astype(int)

PLAYER_ID
51         67
51         81
51         81
51         80
51         73
           ..
1627826    26
1627826    59
1627826    72
1627826    72
1627826    35
Name: GP, Length: 17974, dtype: int32


players_stats['FG3A_per_game'] = players_stats['FG3A']/players_stats['GP']


players_stats['FG3A_per_game']

PLAYER_ID
51         1.492537
51         1.160494
51         2.432099
51           1.6625
51         2.945205
             ...   
1627826         0.0
1627826         0.0
1627826    0.027778
1627826    0.055556
1627826         0.0
Name: FG3A_per_game, Length: 17974, dtype: object


players_stats


players_stats['SEASON_ID'].min()

1976


players_stats.to_csv('C:\\Users\\Vytis\\Desktop\\Perlis UŽT\\Duomenu analitika\\Python\\Baigiamasis darbas\\players_stats.csv')


from nba_api.stats.endpoints import shotchartdetail


# Yra sukurti pagalbiniai failai, kurie padeda surasti komandos ir žaidėjo ID


teams = json.loads(requests.get('https://raw.githubusercontent.com/bttmly/nba/master/data/teams.json').text)


players = json.loads(requests.get('https://raw.githubusercontent.com/bttmly/nba/master/data/players.json').text)


# Susikuriame funkcijas kaip gauti ID pagal komandos/žaidėjo pavadinimą


# Komandos:
def get_team_id(team):
  for team in teams:
    if team['teamName'] == team:
      return team['teamId']
  return -1


# Žaidėjo:
def get_player_id(first, last):
  for player in players:
    if player['firstName'] == first and player['lastName'] == last:
      return player['playerId']
  return -1


#surandame JV ID


get_player_id('Jonas', 'Valanciunas')

202685


#surandame Sabo ID


get_player_id('Domantas', 'Sabonis')

1627734


shot_json = shotchartdetail.ShotChartDetail(
            team_id = get_team_id('New Orleans Pelicans'),
            player_id = get_player_id('Jonas', 'Valanciunas'),
            context_measure_simple = 'PTS',
            season_nullable = '2020-21',
            season_type_all_star = 'Regular Season')

---------------------------------------------------------------------------
JSONDecodeError                           Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_21052/1692337837.py in <module>
----> 1 shot_json = shotchartdetail.ShotChartDetail(
      2             team_id = get_team_id('New Orleans Pelicans'),
      3             player_id = get_player_id('Jonas', 'Valanciunas'),
      4             context_measure_simple = 'PTS',
      5             season_nullable = '2020-21',

~\anaconda3\lib\site-packages\nba_api\stats\endpoints\shotchartdetail.py in __init__(self, team_id, player_id, context_measure_simple, last_n_games, league_id, month, opponent_team_id, period, season_type_all_star, ahead_behind_nullable, clutch_time_nullable, context_filter_nullable, date_from_nullable, date_to_nullable, end_period_nullable, end_range_nullable, game_id_nullable, game_segment_nullable, location_nullable, outcome_nullable, player_position_nullable, point_diff_nullable, position_nullable, range_type_nullable, rookie_year_nullable, season_nullable, season_segment_nullable, start_period_nullable, start_range_nullable, vs_conference_nullable, vs_division_nullable, proxy, headers, timeout, get_request)
     88         }
     89         if get_request:
---> 90             self.get_request()
     91 
     92     def get_request(self):

~\anaconda3\lib\site-packages\nba_api\stats\endpoints\shotchartdetail.py in get_request(self)
     98             timeout=self.timeout,
     99         )
--> 100         self.load_response()
    101 
    102     def load_response(self):

~\anaconda3\lib\site-packages\nba_api\stats\endpoints\shotchartdetail.py in load_response(self)
    101 
    102     def load_response(self):
--> 103         data_sets = self.nba_response.get_data_sets()
    104         self.data_sets = [Endpoint.DataSet(data=data_set) for data_set_name, data_set in data_sets.items()]
    105         self.league_averages = Endpoint.DataSet(data=data_sets['LeagueAverages'])

~\anaconda3\lib\site-packages\nba_api\stats\library\http.py in get_data_sets(self)
     81 
     82     def get_data_sets(self):
---> 83         raw_dict = self.get_dict()
     84         if 'resultSets' in raw_dict:
     85             results = raw_dict['resultSets']

~\anaconda3\lib\site-packages\nba_api\library\http.py in get_dict(self)
     39 
     40     def get_dict(self):
---> 41         return json.loads(self._response)
     42 
     43     def get_json(self):

~\anaconda3\lib\json\__init__.py in loads(s, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw)
    344             parse_int is None and parse_float is None and
    345             parse_constant is None and object_pairs_hook is None and not kw):
--> 346         return _default_decoder.decode(s)
    347     if cls is None:
    348         cls = JSONDecoder

~\anaconda3\lib\json\decoder.py in decode(self, s, _w)
    335 
    336         """
--> 337         obj, end = self.raw_decode(s, idx=_w(s, 0).end())
    338         end = _w(s, end).end()
    339         if end != len(s):

~\anaconda3\lib\json\decoder.py in raw_decode(self, s, idx)
    353             obj, end = self.scan_once(s, idx)
    354         except StopIteration as err:
--> 355             raise JSONDecodeError("Expecting value", s, err.value) from None
    356         return obj, end

JSONDecodeError: Expecting value: line 1 column 1 (char 0)


from nba_api.stats.endpoints import commonallplayers


#Ištraukiame visus egzistuojančius NBA žaidėjus ir susikuriame data frame


all_players = commonallplayers.CommonAllPlayers().get_data_frames()[0]
all_players


all_players.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4726 entries, 0 to 4725
Data columns (total 16 columns):
 #   Column                     Non-Null Count  Dtype 
---  ------                     --------------  ----- 
 0   PERSON_ID                  4726 non-null   int64 
 1   DISPLAY_LAST_COMMA_FIRST   4726 non-null   object
 2   DISPLAY_FIRST_LAST         4726 non-null   object
 3   ROSTERSTATUS               4726 non-null   int64 
 4   FROM_YEAR                  4726 non-null   object
 5   TO_YEAR                    4726 non-null   object
 6   PLAYERCODE                 4725 non-null   object
 7   PLAYER_SLUG                4726 non-null   object
 8   TEAM_ID                    4726 non-null   int64 
 9   TEAM_CITY                  4726 non-null   object
 10  TEAM_NAME                  4726 non-null   object
 11  TEAM_ABBREVIATION          4726 non-null   object
 12  TEAM_CODE                  4726 non-null   object
 13  TEAM_SLUG                  575 non-null    object
 14  GAMES_PLAYED_FLAG          4726 non-null   object
 15  OTHERLEAGUE_EXPERIENCE_CH  4726 non-null   object
dtypes: int64(3), object(13)
memory usage: 590.9+ KB


all_players['FROM_YEAR'].min()

'1946'


all_players['FROM_YEAR'] = all_players['FROM_YEAR'].astype('int64')


all_players.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 4726 entries, 0 to 4725
Data columns (total 16 columns):
 #   Column                     Non-Null Count  Dtype 
---  ------                     --------------  ----- 
 0   PERSON_ID                  4726 non-null   int64 
 1   DISPLAY_LAST_COMMA_FIRST   4726 non-null   object
 2   DISPLAY_FIRST_LAST         4726 non-null   object
 3   ROSTERSTATUS               4726 non-null   int64 
 4   FROM_YEAR                  4726 non-null   int64 
 5   TO_YEAR                    4726 non-null   object
 6   PLAYERCODE                 4725 non-null   object
 7   PLAYER_SLUG                4726 non-null   object
 8   TEAM_ID                    4726 non-null   int64 
 9   TEAM_CITY                  4726 non-null   object
 10  TEAM_NAME                  4726 non-null   object
 11  TEAM_ABBREVIATION          4726 non-null   object
 12  TEAM_CODE                  4726 non-null   object
 13  TEAM_SLUG                  575 non-null    object
 14  GAMES_PLAYED_FLAG          4726 non-null   object
 15  OTHERLEAGUE_EXPERIENCE_CH  4726 non-null   object
dtypes: int64(4), object(12)
memory usage: 590.9+ KB


all_players['FROM_YEAR'].min()

1946


all_players['TO_YEAR'] = all_players['TO_YEAR'].astype('int64')
all_players = all_players[all_players['TO_YEAR']>=1995]
all_players

C:\Users\Vytis\AppData\Local\Temp/ipykernel_20512/1869395000.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  all_players['TO_YEAR'] = all_players['TO_YEAR'].astype('int64')


# Noted That, one may got block by website if too many queries at the same time so an one-second pause is added 
# to each query to avoid that. In addition, an error log is also added since queries may still fail due to unknown issues, 
# so it would be better to record missing queries so it can be re-queried later.


pip install tqdm #rodo scriptinimo procesą terminalo apačioje

Requirement already satisfied: tqdm in c:\users\vytis\anaconda3\lib\site-packages (4.62.3)
Requirement already satisfied: colorama in c:\users\vytis\anaconda3\lib\site-packages (from tqdm) (0.4.4)
Note: you may need to restart the kernel to use updated packages.


from nba_api.stats.endpoints import playercareerstats
import time

# extract all IDs
players_ID = all_players['PERSON_ID']

# initial an empty dataframe
players_stats = pd.DataFrame()
# save fail queries
error_log = []
# extracting stats
from tqdm import tqdm #prisidėt reikėjo, kitaip nesuprato kas yra tqdm
for ID in tqdm(players_ID):
    try:
        time.sleep(1) # avoid too many queries submitted at the same time
        career = playercareerstats.PlayerCareerStats(player_id=ID)
        player_career = career.get_data_frames()[0]
        players_stats = pd.concat([players_stats,player_career],axis=0,ignore_index=True)
    except:
        error_log.append(ID)
        
# re-query missing IDs. Kaip supratau, kartais ne iškarto visus duomenis teisingai nuscriptina API, tai reikalingas papildomas kodas
for ID in tqdm(error_log):
    try:
        time.sleep(1)
        career = playercareerstats.PlayerCareerStats(player_id=ID)
        player_career = career.get_data_frames()[0]
        players_stats = pd.concat([players_stats,player_career],axis=0,ignore_index=True)
    except:
        error_log.append(ID)
 
# save result
with open("player_stats.pkl",'wb') as f:
    pickle.dump(players_stats,f)

100%|████████████████████████████████████████████████████████████████████████████| 2538/2538 [1:50:29<00:00,  2.61s/it]
0it [00:00, ?it/s]

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_20512/2667977556.py in <module>
     32 # save result
     33 with open("player_stats.pkl",'wb') as f:
---> 34     pickle.dump(players_stats,f)

NameError: name 'pickle' is not defined


import pickle


with open("player_stats.pkl",'wb') as f:
    pickle.dump(players_stats,f)


from nba_api.stats.endpoints import teamyearbyyearstats
from nba_api.stats.static import teams
import time
import pickle

nba_teams = teams.get_teams()

teams_stats = pd.DataFrame()
error_teams =[]

from tqdm import tqdm
for i in tqdm(nba_teams):
    try:
        time.sleep(1)
        team = teamyearbyyearstats.TeamYearByYearStats(team_id=i['id'])
        team_data = team.get_data_frames()[0]
        teams_stats = pd.concat([teams_stats,team_data],axis=0,ignore_index=True)
    except:
        error_teams.append(i)
        
with open("teams_stats.pkl","wb") as f:
    pickle.dump(teams_stats,f)

100%|██████████████████████████████████████████████████████████████████████████████████| 30/30 [00:51<00:00,  1.70s/it]


from nba_api.stats.endpoints import playerawards

players_awards = pd.DataFrame()
error_awards = []
from tqdm import tqdm
for ID in tqdm(players_ID):
    try:
        time.sleep(1)
        award = playerawards.PlayerAwards(player_id=ID)
        award.get_data_frames()[0]
        players_awards = pd.concat([players_awards,award.get_data_frames()[0]],axis=0,ignore_index=True)
    except:
        error_awards.append(ID)
        
for ID in tqdm(error_awards):
    time.sleep(1)
    award = playerawards.PlayerAwards(player_id=ID)
    award.get_data_frames()[0]
    players_awards = pd.concat([players_awards,award.get_data_frames()[0]],axis=0,ignore_index=True)

with open("players_awards.pkl","wb") as f:
    pickle.dump(players_awards,f)

100%|██████████████████████████████████████████████████████████████████████████████| 2538/2538 [57:54<00:00,  1.37s/it]
0it [00:00, ?it/s]


players_stats = pd.read_pickle('C:\\Users\\Vytis\\player_stats.pkl')


players_stats.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 17974 entries, 0 to 17973
Data columns (total 27 columns):
 #   Column             Non-Null Count  Dtype  
---  ------             --------------  -----  
 0   PLAYER_ID          17974 non-null  object 
 1   SEASON_ID          17974 non-null  object 
 2   LEAGUE_ID          17974 non-null  object 
 3   TEAM_ID            17974 non-null  object 
 4   TEAM_ABBREVIATION  17974 non-null  object 
 5   PLAYER_AGE         17974 non-null  float64
 6   GP                 17974 non-null  object 
 7   GS                 17974 non-null  object 
 8   MIN                17974 non-null  float64
 9   FGM                17974 non-null  object 
 10  FGA                17974 non-null  object 
 11  FG_PCT             17974 non-null  float64
 12  FG3M               17966 non-null  object 
 13  FG3A               17966 non-null  object 
 14  FG3_PCT            17966 non-null  float64
 15  FTM                17974 non-null  object 
 16  FTA                17974 non-null  object 
 17  FT_PCT             17974 non-null  float64
 18  OREB               17974 non-null  object 
 19  DREB               17974 non-null  object 
 20  REB                17974 non-null  object 
 21  AST                17974 non-null  object 
 22  STL                17974 non-null  object 
 23  BLK                17974 non-null  object 
 24  TOV                17973 non-null  object 
 25  PF                 17974 non-null  object 
 26  PTS                17974 non-null  object 
dtypes: float64(5), object(22)
memory usage: 3.7+ MB


teams_stats = pd.read_pickle('C:\\Users\\Vytis\\teams_stats.pkl')
teams_stats


players_awards = pd.read_pickle('C:\\Users\\Vytis\\players_awards.pkl')
players_awards


## Lentelių tvarkymas


import matplotlib.pyplot as plt
import seaborn as sns
from tqdm import tqdm
import pickle


teams_stats.rename(columns={'YEAR':'SEASON_ID'},inplace=True)
players_teams = players_stats.merge(teams_stats,how='inner',on=['TEAM_ID','SEASON_ID'],suffixes=('_player','_team'))
players_teams


# Papildomi argumentai - kiek kartų per sezoną laimėjo savaitės ar mėnesio prizus  
week = players_awards[players_awards['DESCRIPTION']=='NBA Player of the Week'].groupby(['PERSON_ID','SEASON'])
month = players_awards[players_awards['DESCRIPTION']=='NBA Player of the Month'].groupby(['PERSON_ID','SEASON'])
player_of_week = []
player_of_month = []
for i in tqdm(range(len(players_teams))):

    player_id = players_teams.loc[i,'PLAYER_ID']
    season_id = players_teams.loc[i,'SEASON_ID']
    try:
        player_of_month.append(month.get_group((player_id,season_id))['PERSON_ID'].count())
    except:
        player_of_month.append(0)
    
    try:
        player_of_week.append(week.get_group((player_id,season_id))['PERSON_ID'].count())
    except:
        player_of_week.append(0)
        

players_teams['player_of_week'] = player_of_week

players_teams['player_of_month'] = player_of_month

100%|█████████████████████████████████████████████████████████████████████████| 16400/16400 [00:01<00:00, 14564.11it/s]


# Svarbi detalė - ar laimėjo žaidėjas MVP ar ne


mvp = players_awards[players_awards['DESCRIPTION']=='NBA Most Valuable Player'].rename(columns={"PERSON_ID":"PLAYER_ID","SEASON":"SEASON_ID","TYPE":"MVP"})

df = players_teams.merge(mvp.loc[:,["PLAYER_ID","SEASON_ID","MVP"]],how="left",on=["PLAYER_ID","SEASON_ID"])

df["MVP"] = df.MVP.map({"Award":1,np.NaN:0}) #jei laimėjo tais metais MVP - 1, jei ne - 0

df


df.to_csv('nba_stats.csv')


with open("df.pkl",'wb') as f:
    pickle.dump(df,f)


df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 16400 entries, 0 to 16399
Data columns (total 59 columns):
 #   Column                 Non-Null Count  Dtype  
---  ------                 --------------  -----  
 0   PLAYER_ID              16400 non-null  object 
 1   SEASON_ID              16400 non-null  int64  
 2   PLAYER_AGE             16400 non-null  float64
 3   GP_player              16400 non-null  object 
 4   GS                     16400 non-null  object 
 5   MIN                    16400 non-null  float64
 6   FGM_player             16400 non-null  object 
 7   FGA_player             16400 non-null  object 
 8   FG_PCT_player          16400 non-null  float64
 9   FG3M_player            16393 non-null  object 
 10  FG3A_player            16393 non-null  object 
 11  FG3_PCT_player         16393 non-null  float64
 12  FTM_player             16400 non-null  object 
 13  FTA_player             16400 non-null  object 
 14  FT_PCT_player          16400 non-null  float64
 15  OREB_player            16400 non-null  object 
 16  DREB_player            16400 non-null  object 
 17  REB_player             16400 non-null  object 
 18  AST_player             16400 non-null  object 
 19  STL_player             16400 non-null  object 
 20  BLK_player             16400 non-null  object 
 21  TOV_player             16399 non-null  object 
 22  PF_player              16400 non-null  object 
 23  PTS_player             16400 non-null  object 
 24  TEAM_CITY              16400 non-null  object 
 25  TEAM_NAME              16400 non-null  object 
 26  GP_team                16400 non-null  int64  
 27  WINS                   16400 non-null  int64  
 28  LOSSES                 16400 non-null  int64  
 29  WIN_PCT                16400 non-null  float64
 30  CONF_RANK              16400 non-null  int64  
 31  DIV_RANK               16400 non-null  int64  
 32  PO_WINS                16400 non-null  int64  
 33  PO_LOSSES              16400 non-null  int64  
 34  CONF_COUNT             16400 non-null  float64
 35  DIV_COUNT              16400 non-null  int64  
 36  NBA_FINALS_APPEARANCE  16400 non-null  object 
 37  FGM_team               16400 non-null  int64  
 38  FGA_team               16400 non-null  int64  
 39  FG_PCT_team            16400 non-null  float64
 40  FG3M_team              16400 non-null  int64  
 41  FG3A_team              16400 non-null  int64  
 42  FG3_PCT_team           16400 non-null  float64
 43  FTM_team               16400 non-null  int64  
 44  FTA_team               16400 non-null  int64  
 45  FT_PCT_team            16400 non-null  float64
 46  OREB_team              16400 non-null  int64  
 47  DREB_team              16400 non-null  int64  
 48  REB_team               16400 non-null  int64  
 49  AST_team               16400 non-null  int64  
 50  PF_team                16400 non-null  int64  
 51  STL_team               16400 non-null  int64  
 52  TOV_team               16400 non-null  int64  
 53  BLK_team               16400 non-null  int64  
 54  PTS_team               16400 non-null  int64  
 55  PTS_RANK               16400 non-null  int64  
 56  player_of_week         16400 non-null  int64  
 57  player_of_month        16400 non-null  int64  
 58  MVP                    16400 non-null  int64  
dtypes: float64(10), int64(28), object(21)
memory usage: 7.5+ MB


df = pd.read_pickle('df.pkl')


# panaikinti pasikartojančius stulpelius
df.drop(columns=["LEAGUE_ID","TEAM_ID","TEAM_ABBREVIATION"],inplace=True)


# pakeisti sezono stulpelio tipa i int, 
df['SEASON_ID'] = df['SEASON_ID'].map(lambda x: int(x.split("-",1)[0]))


num_cols = ['PLAYER_AGE', 'GP', 'GS', 'MIN', 'FGM', 'FGA', 'FG_PCT', 'FG3M', 'FG3A',
       'FG3_PCT', 'FTM', 'FTA', 'FT_PCT', 'OREB', 'DREB', 'REB', 'AST', 'STL',
       'BLK', 'TOV', 'PF', 'PTS', 'WIN_PCT', 'CONF_RANK', 'DIV_RANK',
       'player_of_week', 'player_of_month']


pip install scikit-learn

Requirement already satisfied: scikit-learn in c:\users\vytis\anaconda3\lib\site-packages (1.0.2)
Requirement already satisfied: scipy>=1.1.0 in c:\users\vytis\anaconda3\lib\site-packages (from scikit-learn) (1.7.1)
Requirement already satisfied: joblib>=0.11 in c:\users\vytis\anaconda3\lib\site-packages (from scikit-learn) (1.1.0)
Requirement already satisfied: threadpoolctl>=2.0.0 in c:\users\vytis\anaconda3\lib\site-packages (from scikit-learn) (2.2.0)
Requirement already satisfied: numpy>=1.14.6 in c:\users\vytis\anaconda3\lib\site-packages (from scikit-learn) (1.20.3)
Note: you may need to restart the kernel to use updated packages.


# Pasidaryti data season by season principą. Nesigavo 
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
for year in  df["SEASON_ID"].unique().tolist():
    for col in num_cols:
        df.loc[df['SEASON_ID']==year, col] = scaler.fit_transform(df.loc[df['SEASON_ID']==year, col].to_numpy().reshape(-1,1))

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
~\anaconda3\lib\site-packages\pandas\core\indexes\base.py in get_loc(self, key, method, tolerance)
   3360             try:
-> 3361                 return self._engine.get_loc(casted_key)
   3362             except KeyError as err:

~\anaconda3\lib\site-packages\pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_loc()

~\anaconda3\lib\site-packages\pandas\_libs\index.pyx in pandas._libs.index.IndexEngine.get_loc()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item()

pandas\_libs\hashtable_class_helper.pxi in pandas._libs.hashtable.PyObjectHashTable.get_item()

KeyError: 'GP'

The above exception was the direct cause of the following exception:

KeyError                                  Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_21052/2661967775.py in <module>
      4 for year in  df["SEASON_ID"].unique().tolist():
      5     for col in num_cols:
----> 6         df.loc[df['SEASON_ID']==year, col] = scaler.fit_transform(df.loc[df['SEASON_ID']==year, col].to_numpy().reshape(-1,1))

~\anaconda3\lib\site-packages\pandas\core\indexing.py in __getitem__(self, key)
    923                 with suppress(KeyError, IndexError):
    924                     return self.obj._get_value(*key, takeable=self._takeable)
--> 925             return self._getitem_tuple(key)
    926         else:
    927             # we by definition only have the 0th axis

~\anaconda3\lib\site-packages\pandas\core\indexing.py in _getitem_tuple(self, tup)
   1098     def _getitem_tuple(self, tup: tuple):
   1099         with suppress(IndexingError):
-> 1100             return self._getitem_lowerdim(tup)
   1101 
   1102         # no multi-index, so validate all of the indexers

~\anaconda3\lib\site-packages\pandas\core\indexing.py in _getitem_lowerdim(self, tup)
    836                 # We don't need to check for tuples here because those are
    837                 #  caught by the _is_nested_tuple_indexer check above.
--> 838                 section = self._getitem_axis(key, axis=i)
    839 
    840                 # We should never have a scalar section here, because

~\anaconda3\lib\site-packages\pandas\core\indexing.py in _getitem_axis(self, key, axis)
   1162         # fall thru to straight lookup
   1163         self._validate_key(key, axis)
-> 1164         return self._get_label(key, axis=axis)
   1165 
   1166     def _get_slice_axis(self, slice_obj: slice, axis: int):

~\anaconda3\lib\site-packages\pandas\core\indexing.py in _get_label(self, label, axis)
   1111     def _get_label(self, label, axis: int):
   1112         # GH#5667 this will fail if the label is not present in the axis.
-> 1113         return self.obj.xs(label, axis=axis)
   1114 
   1115     def _handle_lowerdim_multi_index_axis0(self, tup: tuple):

~\anaconda3\lib\site-packages\pandas\core\generic.py in xs(self, key, axis, level, drop_level)
   3759         if axis == 1:
   3760             if drop_level:
-> 3761                 return self[key]
   3762             index = self.columns
   3763         else:

~\anaconda3\lib\site-packages\pandas\core\frame.py in __getitem__(self, key)
   3456             if self.columns.nlevels > 1:
   3457                 return self._getitem_multilevel(key)
-> 3458             indexer = self.columns.get_loc(key)
   3459             if is_integer(indexer):
   3460                 indexer = [indexer]

~\anaconda3\lib\site-packages\pandas\core\indexes\base.py in get_loc(self, key, method, tolerance)
   3361                 return self._engine.get_loc(casted_key)
   3362             except KeyError as err:
-> 3363                 raise KeyError(key) from err
   3364 
   3365         if is_scalar(key) and isna(key) and not self.hasnans:

KeyError: 'GP'


train = df.loc[(df['SEASON_ID']>=1995)&(df['SEASON_ID']<=2020),:]
test = df.loc[(df['SEASON_ID']==2021),:]


y = train['MVP']
X = train.iloc[:,:-1]

y_test = test['MVP']
X_test = test.iloc[:,:-1]


print(y.value_counts())
y.value_counts().plot(kind='bar')

0    13640
1       26
Name: MVP, dtype: int64

<AxesSubplot:>


corrmat = df.corr()

k = 10 # show top k most correlated features
cols = corrmat.nlargest(k, 'MVP')['MVP'].index
cm = np.corrcoef(df[cols].values.T)
sns.set(font_scale=1.0)
hm = sns.heatmap(cm, cbar=True, annot=True, square=True, fmt='.2f', annot_kws={'size': 8}, yticklabels=cols.values, xticklabels=cols.values)
plt.show()


# drop correlated feautures
X_corr = X.corr()
corr_names = set()
for i in range(len(X_corr .columns)):
    for j in range(i):
        if abs(X_corr.iloc[i, j]) > 0.8:
            col = X_corr.columns[i]
            corr_names.add(col)

X.drop(columns=corr_names,inplace=True)
X_test.drop(columns=corr_names,inplace=True)


features = ['PLAYER_AGE', 'GP_player', 'GS', 'MIN', 'FGM_player', 'FGA_player',
       'FG_PCT_player', 'FG3M_player', 'FG3A_player', 'FG3_PCT_player',
       'FTM_player', 'FTA_player', 'FT_PCT_player', 'OREB_player',
       'DREB_player', 'REB_player', 'AST_player', 'STL_player', 'BLK_player',
       'TOV_player', 'PF_player', 'PTS_player', 'WINS','player_of_week', 'player_of_month']


pip install statsmodels

Requirement already satisfied: statsmodels in c:\users\vytis\anaconda3\lib\site-packages (0.13.1)
Requirement already satisfied: pandas>=0.25 in c:\users\vytis\anaconda3\lib\site-packages (from statsmodels) (1.3.4)
Requirement already satisfied: numpy>=1.17 in c:\users\vytis\anaconda3\lib\site-packages (from statsmodels) (1.20.3)
Requirement already satisfied: patsy>=0.5.2 in c:\users\vytis\anaconda3\lib\site-packages (from statsmodels) (0.5.2)
Requirement already satisfied: scipy>=1.3 in c:\users\vytis\anaconda3\lib\site-packages (from statsmodels) (1.7.1)
Requirement already satisfied: pytz>=2017.3 in c:\users\vytis\anaconda3\lib\site-packages (from pandas>=0.25->statsmodels) (2021.3)
Requirement already satisfied: python-dateutil>=2.7.3 in c:\users\vytis\anaconda3\lib\site-packages (from pandas>=0.25->statsmodels) (2.8.2)
Requirement already satisfied: six in c:\users\vytis\anaconda3\lib\site-packages (from patsy>=0.5.2->statsmodels) (1.16.0)
Note: you may need to restart the kernel to use updated packages.

WARNING: Ignoring invalid distribution -tatsmodels (c:\users\vytis\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -tatsmodels (c:\users\vytis\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -tatsmodels (c:\users\vytis\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -tatsmodels (c:\users\vytis\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -tatsmodels (c:\users\vytis\anaconda3\lib\site-packages)
WARNING: Ignoring invalid distribution -tatsmodels (c:\users\vytis\anaconda3\lib\site-packages)


import statsmodels.api as sm
X_logit = X.drop(columns=['PLAYER_ID','SEASON_ID'])
ols = sm.Logit(y,X_logit.astype('float'))
result = ols.fit()
p_values = result.summary2().tables[1]['P>|z|']
p_values = pd.Series(p_values).sort_values(ascending = True)
p_values[p_values<=0.05]

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_21052/4172156579.py in <module>
      1 import statsmodels.api as sm
      2 X_logit = X.drop(columns=['PLAYER_ID','SEASON_ID'])
----> 3 ols = sm.Logit(y,X_logit.astype('float'))
      4 result = ols.fit()
      5 p_values = result.summary2().tables[1]['P>|z|']

~\anaconda3\lib\site-packages\pandas\core\generic.py in astype(self, dtype, copy, errors)
   5813         else:
   5814             # else, only a single dtype is given
-> 5815             new_data = self._mgr.astype(dtype=dtype, copy=copy, errors=errors)
   5816             return self._constructor(new_data).__finalize__(self, method="astype")
   5817 

~\anaconda3\lib\site-packages\pandas\core\internals\managers.py in astype(self, dtype, copy, errors)
    416 
    417     def astype(self: T, dtype, copy: bool = False, errors: str = "raise") -> T:
--> 418         return self.apply("astype", dtype=dtype, copy=copy, errors=errors)
    419 
    420     def convert(

~\anaconda3\lib\site-packages\pandas\core\internals\managers.py in apply(self, f, align_keys, ignore_failures, **kwargs)
    325                     applied = b.apply(f, **kwargs)
    326                 else:
--> 327                     applied = getattr(b, f)(**kwargs)
    328             except (TypeError, NotImplementedError):
    329                 if not ignore_failures:

~\anaconda3\lib\site-packages\pandas\core\internals\blocks.py in astype(self, dtype, copy, errors)
    589         values = self.values
    590 
--> 591         new_values = astype_array_safe(values, dtype, copy=copy, errors=errors)
    592 
    593         new_values = maybe_coerce_values(new_values)

~\anaconda3\lib\site-packages\pandas\core\dtypes\cast.py in astype_array_safe(values, dtype, copy, errors)
   1307 
   1308     try:
-> 1309         new_values = astype_array(values, dtype, copy=copy)
   1310     except (ValueError, TypeError):
   1311         # e.g. astype_nansafe can fail on object-dtype of strings

~\anaconda3\lib\site-packages\pandas\core\dtypes\cast.py in astype_array(values, dtype, copy)
   1255 
   1256     else:
-> 1257         values = astype_nansafe(values, dtype, copy=copy)
   1258 
   1259     # in pandas we don't store numpy str dtypes, so convert to object

~\anaconda3\lib\site-packages\pandas\core\dtypes\cast.py in astype_nansafe(arr, dtype, copy, skipna)
   1093     if arr.ndim > 1:
   1094         flat = arr.ravel()
-> 1095         result = astype_nansafe(flat, dtype, copy=copy, skipna=skipna)
   1096         # error: Item "ExtensionArray" of "Union[ExtensionArray, ndarray]" has no
   1097         # attribute "reshape"

~\anaconda3\lib\site-packages\pandas\core\dtypes\cast.py in astype_nansafe(arr, dtype, copy, skipna)
   1199     if copy or is_object_dtype(arr.dtype) or is_object_dtype(dtype):
   1200         # Explicit copy, or required since NumPy can't view from / to object.
-> 1201         return arr.astype(dtype, copy=True)
   1202 
   1203     return arr.astype(dtype, copy=copy)

ValueError: could not convert string to float: 'Denver'


X_logit.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 13666 entries, 52 to 16289
Data columns (total 43 columns):
 #   Column                 Non-Null Count  Dtype  
---  ------                 --------------  -----  
 0   PLAYER_AGE             13666 non-null  float64
 1   GP_player              13666 non-null  object 
 2   GS                     13666 non-null  object 
 3   MIN                    13666 non-null  float64
 4   FGM_player             13666 non-null  object 
 5   FGA_player             13666 non-null  object 
 6   FG_PCT_player          13666 non-null  float64
 7   FG3M_player            13666 non-null  object 
 8   FG3A_player            13666 non-null  object 
 9   FG3_PCT_player         13666 non-null  float64
 10  FTM_player             13666 non-null  object 
 11  FTA_player             13666 non-null  object 
 12  FT_PCT_player          13666 non-null  float64
 13  OREB_player            13666 non-null  object 
 14  DREB_player            13666 non-null  object 
 15  REB_player             13666 non-null  object 
 16  AST_player             13666 non-null  object 
 17  STL_player             13666 non-null  object 
 18  BLK_player             13666 non-null  object 
 19  TOV_player             13666 non-null  object 
 20  PF_player              13666 non-null  object 
 21  PTS_player             13666 non-null  object 
 22  TEAM_CITY              13666 non-null  object 
 23  TEAM_NAME              13666 non-null  object 
 24  GP_team                13666 non-null  int64  
 25  WINS                   13666 non-null  int64  
 26  PO_WINS                13666 non-null  int64  
 27  CONF_COUNT             13666 non-null  float64
 28  DIV_COUNT              13666 non-null  int64  
 29  NBA_FINALS_APPEARANCE  13666 non-null  object 
 30  FG_PCT_team            13666 non-null  float64
 31  FG3M_team              13666 non-null  int64  
 32  FG3_PCT_team           13666 non-null  float64
 33  FTM_team               13666 non-null  int64  
 34  FT_PCT_team            13666 non-null  float64
 35  OREB_team              13666 non-null  int64  
 36  PF_team                13666 non-null  int64  
 37  STL_team               13666 non-null  int64  
 38  TOV_team               13666 non-null  int64  
 39  BLK_team               13666 non-null  int64  
 40  PTS_RANK               13666 non-null  int64  
 41  player_of_week         13666 non-null  int64  
 42  player_of_month        13666 non-null  int64  
dtypes: float64(9), int64(14), object(20)
memory usage: 4.6+ MB


mkdir nba-sql


cd nba-sql

C:\Users\Vytis\nba-sql


python -m venv venv

  File "C:\Users\Vytis\AppData\Local\Temp/ipykernel_20512/717537615.py", line 1
    python -m venv venv
              ^
SyntaxError: invalid syntax


$ source venv/bin/activate

  File "C:\Users\Vytis\AppData\Local\Temp/ipykernel_20512/1330130224.py", line 1
    $ source venv/bin/activate
    ^
SyntaxError: invalid syntax

	PERSON_ID	DISPLAY_LAST_COMMA_FIRST	DISPLAY_FIRST_LAST	ROSTERSTATUS	FROM_YEAR	TO_YEAR	PLAYERCODE	PLAYER_SLUG	TEAM_ID	TEAM_CITY	TEAM_NAME	TEAM_ABBREVIATION	TEAM_CODE	TEAM_SLUG	GAMES_PLAYED_FLAG	OTHERLEAGUE_EXPERIENCE_CH
0	76001	Abdelnaby, Alaa	Alaa Abdelnaby	0	1990	1994	HISTADD_alaa_abdelnaby	alaa_abdelnaby	0					None	Y	00
1	76002	Abdul-Aziz, Zaid	Zaid Abdul-Aziz	0	1968	1977	HISTADD_zaid_abdul-aziz	zaid_abdul-aziz	0					None	Y	00
2	76003	Abdul-Jabbar, Kareem	Kareem Abdul-Jabbar	0	1969	1988	HISTADD_kareem_abdul-jabbar	kareem_abdul-jabbar	0					None	Y	00
3	51	Abdul-Rauf, Mahmoud	Mahmoud Abdul-Rauf	0	1990	2000	mahmoud_abdul-rauf	mahmoud_abdul-rauf	0					None	Y	00
4	1505	Abdul-Wahad, Tariq	Tariq Abdul-Wahad	0	1997	2003	tariq_abdul-wahad	tariq_abdul-wahad	0					None	Y	00
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
4722	1627790	Zizic, Ante	Ante Zizic	0	2017	2019	ante_zizic	ante_zizic	0					None	Y	01
4723	78647	Zoet, Jim	Jim Zoet	0	1982	1982	HISTADD_jim_zoet	jim_zoet	0					None	Y	00
4724	78648	Zopf, Bill	Bill Zopf	0	1970	1970	HISTADD_zip_zopf	bill_zopf	0					None	Y	00
4725	1627826	Zubac, Ivica	Ivica Zubac	1	2016	2021	ivica_zubac	ivica_zubac	1610612746	LA	Clippers	LAC	clippers	clippers	Y	01
4726	78650	Zunic, Matt	Matt Zunic	0	1948	1948	HISTADD_matt_zunic	matt_zunic	0					None	Y	00

	PLAYER_ID	DISPLAY_FIRST_LAST	TEAM_CITY	TEAM_NAME
0	76001	Alaa Abdelnaby
1	76002	Zaid Abdul-Aziz
2	76003	Kareem Abdul-Jabbar
3	51	Mahmoud Abdul-Rauf
4	1505	Tariq Abdul-Wahad
...	...	...	...	...
4722	1627790	Ante Zizic
4723	78647	Jim Zoet
4724	78648	Bill Zopf
4725	1627826	Ivica Zubac	LA	Clippers
4726	78650	Matt Zunic

	FG3A
SEASON_ID
2001	35074
2002	34912
2003	35492
2004	38748
2005	39313
2006	41672
2007	44544
2008	44583
2009	44622
2010	44313
2011	36395
2012	49067
2013	52974
2014	55137
2015	59241
2016	66421
2017	71339
2018	78742
2019	72252
2020	74822
2021	38286

	FG3M
PLAYER_NAME
Stephen Curry	2998
Ray Allen	2973
Reggie Miller	2560
James Harden	2495
Kyle Korver	2425
Jason Terry	2282
Vince Carter	2225
Jamal Crawford	2159
Damian Lillard	2143
Paul Pierce	2128

	FG3A
PLAYER_NAME
Ray Allen	7429
Stephen Curry	6936
James Harden	6879
Reggie Miller	6486
Jamal Crawford	6242
Jason Terry	6010
Vince Carter	5965
LeBron James	5944
Paul Pierce	5773
Damian Lillard	5752

NBA stats įsikelimas¶

Duomenų tvarkymas¶

Analizė¶

Pastarąjį dešimtmetį žaidėjai išmeta žymiai daugiau tritaškių, nei anksčiau - TAIP¶

Steph Curry yra taikliausias tritaškininkas¶

Steph Curry yra ne tik daugiausiai įmetęs tritaškiu, bet ir daugiausiai išmetęs¶

Ar nuo žaidėjo amžiaus priklauso tritaškių pataikymas?¶

Rezultatas¶

Shotchart braižymas¶

NBA MVP prediction¶

Kodel pickle o ne csv? https://towardsdatascience.com/stop-using-csvs-for-storage-pickle-is-an-80-times-faster-alternative-832041bbc199 ¶

Creating virtual environment¶

	PLAYER_ID	SEASON_ID	LEAGUE_ID	TEAM_ID	TEAM_ABBREVIATION	PLAYER_AGE	GP	GS	MIN	FGM	...	FT_PCT	OREB	DREB	REB	AST	STL	BLK	TOV	PF	PTS
0	51	1990-91	00	1610612743	DEN	22.0	67	19	1505.0	417	...	0.857	34	87	121	206	55	4	110	149	942
1	51	1991-92	00	1610612743	DEN	23.0	81	11	1538.0	356	...	0.870	22	92	114	192	44	4	117	130	837
2	51	1992-93	00	1610612743	DEN	24.0	81	81	2710.0	633	...	0.935	51	174	225	344	84	8	187	179	1553
3	51	1993-94	00	1610612743	DEN	25.0	80	78	2617.0	588	...	0.956	27	141	168	362	82	10	151	150	1437
4	51	1994-95	00	1610612743	DEN	26.0	73	43	2082.0	472	...	0.885	32	105	137	263	77	9	119	126	1165
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
17969	1627826	2018-19	00	1610612746	LAC	22.0	26	25	524.0	100	...	0.733	61	139	200	38	10	24	37	64	244
17970	1627826	2018-19	00	0	TOT	22.0	59	37	1039.0	212	...	0.802	115	247	362	63	14	51	70	137	525
17971	1627826	2019-20	00	1610612746	LAC	23.0	72	70	1326.0	236	...	0.747	197	346	543	82	16	66	61	168	596
17972	1627826	2020-21	00	1610612746	LAC	24.0	72	33	1609.0	257	...	0.789	189	330	519	90	24	62	81	187	650
17973	1627826	2021-22	00	1610612746	LAC	24.0	35	35	876.0	126	...	0.737	95	193	288	38	19	41	48	80	336

	PLAYER_ID	SEASON_ID	TEAM_ID	TEAM_ABBREVIATION	PLAYER_AGE	GP	GS	MIN	FGM	...	FT_PCT	OREB	DREB	REB	AST	STL	BLK	TOV	PF	PTS
13696	2557	2003-04	1610612760	SEA	23.0	69	6	1114.0	145	...	0.823	35	73	108	163	52	7	80	105	382
13697	2557	2004-05	1610612760	SEA	24.0	82	82	2572.0	299	...	0.883	55	149	204	483	94	23	149	185	824
13698	2557	2005-06	1610612760	SEA	25.0	79	77	2625.0	334	...	0.877	47	188	235	550	123	22	162	178	910
13699	2557	2006-07	1610612760	SEA	26.0	71	58	2091.0	301	...	0.805	28	136	164	368	83	19	156	181	779
13700	2557	2007-08	1610612760	SEA	27.0	61	5	1223.0	147	...	0.857	13	81	94	241	37	14	80	120	393
13701	2557	2008-09	1610612749	MIL	28.0	72	50	2033.0	250	...	0.869	33	185	218	365	92	16	128	200	688
13702	2557	2009-10	1610612749	MIL	29.0	82	0	1759.0	328	...	0.907	24	119	143	324	54	7	106	182	852
13703	2557	2010-11	1610612750	MIN	30.0	71	66	2159.0	319	...	0.883	37	162	199	384	89	10	158	146	840
13704	2557	2011-12	1610612750	MIN	31.0	53	53	1750.0	242	...	0.891	19	122	141	252	56	16	96	136	639
13705	2557	2012-13	1610612750	MIN	32.0	82	82	2474.0	367	...	0.848	41	165	206	311	82	15	130	187	939
13706	2557	2013-14	1610612749	MIL	33.0	36	12	763.0	84	...	0.684	15	47	62	122	23	2	46	55	206
13707	2557	2013-14	1610612766	CHA	33.0	25	2	378.0	42	...	0.571	8	27	35	54	9	6	20	37	100
13708	2557	2013-14	0	TOT	33.0	61	14	1141.0	126	...	0.654	23	74	97	176	32	8	66	92	306
13709	2557	2014-15	1610612753	ORL	34.0	47	0	683.0	75	...	0.857	8	60	68	96	20	4	38	67	188

	PLAYER_ID	SEASON_ID	LEAGUE_ID	TEAM_ID	TEAM_ABBREVIATION	PLAYER_AGE	GP	GS	MIN	FGM	...	FT_PCT	OREB	DREB	REB	AST	STL	BLK	TOV	PF	PTS
18	1505	2002	00	1610612742	DAL	28.0	14	0	204.0	27	...	0.500	14	26	40	21	6	3	7	26	57
25	949	2002	00	1610612737	ATL	26.0	81	81	3087.0	566	...	0.841	175	502	677	242	87	38	212	240	1608
26	949	2003	00	1610612737	ATL	27.0	53	53	1955.0	383	...	0.880	141	354	495	127	44	19	131	147	1065
27	949	2003	00	1610612757	POR	27.0	32	3	729.0	118	...	0.832	48	96	144	47	24	18	53	75	319
28	949	2003	00	0	TOT	27.0	85	56	2684.0	501	...	0.869	189	450	639	174	68	37	184	222	1384
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
17948	1917	2003	00	1610612746	LAC	26.0	2	0	9.0	0	...	1.000	0	4	4	0	0	1	0	0	4
17949	1917	2003	00	1610612748	MIA	26.0	14	0	106.0	17	...	0.833	4	10	14	2	3	4	6	13	43
17950	1917	2003	00	0	TOT	26.0	16	0	115.0	17	...	0.900	4	14	18	2	3	5	6	13	47
17951	1917	2004	00	1610612748	MIA	27.0	20	0	91.0	17	...	0.583	6	12	18	5	3	2	5	9	43
17959	2583	2005	00	1610612751	NJN	24.0	2	0	32.0	2	...	0.000	1	3	4	7	0	0	4	4	4

	FG3_PCT
PLAYER_NAME
Fred Hoiberg	0.462500
Hubert Davis	0.452000
Pau Gasol	0.448000
Jason Kapono	0.441000
Seth Curry	0.439167
Wesley Person	0.438500
Joe Harris	0.435400
Steve Novak	0.434750
Michael Porter Jr.	0.433500
Stephen Curry	0.433333

	SEASON_ID	TEAM_ID	TEAM_ABBREVIATION	PLAYER_AGE	GP	GS	MIN	FGM	FGA	FG_PCT	...	AST	STL	BLK	TOV	PF	PTS	PLAYER_NAME	TEAM_CITY_y	TEAM_NAME_y	FG3A_per_game
PLAYER_ID
201939	2015	1610612744	GSW	28.0	79	79	2700.0	805	1598	0.504	...	527	169	15	262	161	2375	Stephen Curry	Golden State	Warriors	11.21519

	PLAYER_AGE	FG3_PCT	prediction
PLAYER_ID
51	24.0	0.355	0.354202
51	25.0	0.316	0.355989
51	26.0	0.386	0.357775
51	27.0	0.392	0.359562
51	28.0	0.382	0.361348
...	...	...	...
1629027	21.0	0.361	0.348843
1629027	22.0	0.343	0.350629
1629027	23.0	0.371	0.352416
1917	24.0	0.414	0.354202
1627835	24.0	0.336	0.354202

	TEAM_ID	TEAM_CITY	TEAM_NAME	YEAR	GP	WINS	LOSSES	WIN_PCT	CONF_RANK	DIV_RANK	...	OREB	DREB	REB	AST	PF	STL	TOV	BLK	PTS	PTS_RANK
0	1610612737	Tri-Cities	Blackhawks	1949-50	64	29	35	0.453	0	3	...	0	0	0	1330	2057	0	0	0	5313	10
1	1610612737	Tri-Cities	Blackhawks	1950-51	68	25	43	0.368	0	5	...	0	0	0	1476	2092	0	0	0	5730	3
2	1610612737	Milwaukee	Hawks	1951-52	66	17	49	0.258	0	5	...	0	0	0	1229	1848	0	0	0	4833	10
3	1610612737	Milwaukee	Hawks	1952-53	71	27	44	0.380	0	5	...	0	0	0	1427	2120	0	0	0	5389	9
4	1610612737	Milwaukee	Hawks	1953-54	72	21	51	0.292	0	4	...	0	0	0	1298	1771	0	0	0	5038	9
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1562	1610612766	Charlotte	Hornets	2017-18	82	36	46	0.439	10	3	...	827	2901	3728	1770	1409	559	1041	373	8874	10
1563	1610612766	Charlotte	Hornets	2018-19	82	39	43	0.476	9	2	...	814	2778	3592	1905	1550	591	1001	405	9081	19
1564	1610612766	Charlotte	Hornets	2019-20	65	23	42	0.354	10	4	...	715	2066	2781	1549	1223	428	949	268	6687	30
1565	1610612766	Charlotte	Hornets	2020-21	72	33	39	0.458	10	4	...	762	2389	3151	1933	1298	565	1069	344	7881	23
1566	1610612766	Charlotte	Hornets	2021-22	37	19	18	0.514	7	2	...	388	1222	1610	992	708	328	484	200	4245	2

	PERSON_ID	FIRST_NAME	LAST_NAME	TEAM	DESCRIPTION	ALL_NBA_TEAM_NUMBER	SEASON	MONTH	WEEK	CONFERENCE	TYPE	SUBTYPE1	SUBTYPE2	SUBTYPE3
0	51	Chris	Jackson	Denver Nuggets	All-Rookie Team	2	1990-91	None	None	1610612743	Award	Kia Motors	KIART	None
1	51	Mahmoud	Abdul-Rauf	Denver Nuggets	NBA Most Improved Player	None	1992-93	None	None	None	Award	Kia Motors	KIMIP	None
2	949	Shareef	Abdur-Rahim	Vancouver Grizzlies	All-Rookie Team	1	1996-97	None	None	1610612763	Award	Kia Motors	KIART	None
3	949	Shareef	Abdur-Rahim	Atlanta Hawks	NBA Player of the Week	None	2001-02	None	2001-11-25T00:00:00	East	Award	Kia Motors	KIPWK	None
4	949	Shareef	Abdur-Rahim	Vancouver Grizzlies	NBA Rookie of the Month	None	1996-97	02/01/1997	None	None	Award	Kia Motors	KIRMO	None
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
3739	203092	Tyler	Zeller	Cleveland Cavaliers	All-Rookie Team	2	2012-13	None	None	1610612739	Award	Kia Motors	KIART	None
3740	1917	Wang	Zhizhi	China	Olympic Appearance	None	1996	None	None	None	Award	Olympic	Appearance	None
3741	1917	Wang	Zhizhi	China	Olympic Appearance	None	2000	None	None	None	Award	Olympic	Appearance	None
3742	1917	Wang	Zhizhi	China	Olympic Appearance	None	2008	None	None	None	Award	Olympic	Appearance	None
3743	1917	Wang	Zhizhi	China	Olympic Appearance	None	2012	None	None	None	Award	Olympic	Appearance	None

	PLAYER_ID	SEASON_ID	LEAGUE_ID	TEAM_ID	TEAM_ABBREVIATION	PLAYER_AGE	GP_player	GS	MIN	FGM_player	...	OREB_team	DREB_team	REB_team	AST_team	PF_team	STL_team	TOV_team	BLK_team	PTS_team	PTS_RANK
0	51	1990-91	00	1610612743	DEN	22.0	67	19	1505.0	417	...	1520	2530	4050	2005	2235	856	1332	406	9828	1
1	149	1990-91	00	1610612743	DEN	28.0	66	66	2346.0	560	...	1520	2530	4050	2005	2235	856	1332	406	9828	1
2	246	1990-91	00	1610612743	DEN	27.0	41	2	659.0	85	...	1520	2530	4050	2005	2235	856	1332	406	9828	1
3	76433	1990-91	00	1610612743	DEN	24.0	58	25	1121.0	118	...	1520	2530	4050	2005	2235	856	1332	406	9828	1
4	422	1990-91	00	1610612743	DEN	26.0	21	4	217.0	29	...	1520	2530	4050	2005	2235	856	1332	406	9828	1
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
16395	433	1983-84	00	1610612751	NJN	24.0	81	81	3003.0	495	...	1221	2313	3534	2148	2243	814	1608	499	9019	13
16396	433	1987-88	00	1610612751	NJN	28.0	70	70	2637.0	466	...	1075	2262	3337	1795	2042	727	1503	385	8235	22
16397	1006	1981-82	00	1610612754	IND	24.0	82	75	2277.0	407	...	1141	2372	3513	1897	2041	753	1393	494	8379	22
16398	1006	1982-83	00	1610612754	IND	25.0	78	74	2513.0	580	...	1299	2294	3593	2150	2086	755	1535	411	8911	12
16399	1006	1983-84	00	1610612754	IND	26.0	69	53	2279.0	411	...	1002	2398	3400	2169	2061	834	1525	398	8566	19

NBA stats įsikelimas¶

Duomenų tvarkymas¶

Analizė¶

Pastarąjį dešimtmetį žaidėjai išmeta žymiai daugiau tritaškių, nei anksčiau - TAIP¶

Steph Curry yra taikliausias tritaškininkas¶

Steph Curry yra ne tik daugiausiai įmetęs tritaškiu, bet ir daugiausiai išmetęs¶

Ar nuo žaidėjo amžiaus priklauso tritaškių pataikymas?¶

Rezultatas¶

Shotchart braižymas¶

NBA MVP prediction¶

Kodel pickle o ne csv? https://towardsdatascience.com/stop-using-csvs-for-storage-pickle-is-an-80-times-faster-alternative-832041bbc199¶

Creating virtual environment¶

Kodel pickle o ne csv? https://towardsdatascience.com/stop-using-csvs-for-storage-pickle-is-an-80-times-faster-alternative-832041bbc199 ¶