本文共 18701 字，大约阅读时间需要 62 分钟。
All basketball teams have a camera system called installed in their arenas. These camera systems track players and the ball throughout a basketball game.
所有篮球队在赛场上安装了名为的摄像头系统。 这些摄像机系统在整个篮球比赛中跟踪球员和球。
The data produced by sportsvu camera systems used to be freely available on NBA.com, but was recently removed (I have no idea why). Luckily, the data for about 600 games are available on . In this post, I show how to create a video recreation of a given basketball play using the sportsvu data.
sportsvu摄像机系统产生的数据曾经可以在NBA.com上免费获得，但是最近被删除了（我不知道为什么）。 幸运的是，可以在上大约600场比赛的数据。 在本文中，我将展示如何使用sportsvu数据为给定的篮球比赛创建视频娱乐。
This code is also available as a jupyter notebook on my .
此代码也可以在我的上作为jupyter笔记本使用。
1122334455 |
The data is provided as a json. Here’s how to import the python json library and load the data. I’m a T-Wolves fan, so the game I chose is a wolves game.
数据作为json提供。 这是导入python json库和加载数据的方法。 我是T-狼队的球迷，所以我选择的游戏是狼队游戏。
112233 |
Let’s take a quick look at the data. It’s a dictionary with three keys: gamedate, gameid, and events. Gamedate and gameid are the date of this game and its specific id number, respectively. Events is the structure with data we’re interested in.
让我们快速看一下数据。 这是一本具有三个键的字典：gamedate，gameid和events。 Gamedate和gameid分别是该游戏的日期及其特定的ID号。 事件是我们感兴趣的数据结构。
11 |
[u'gamedate', u'gameid', u'events'][u'gamedate', u'gameid', u'events']
Lets take a look at the first event. The first event has an associated eventid number. We will use these later. There’s also data for each player on the visiting and home team. We will use these later too. Finally, and most importantly, there’s the “moments.” There are 25 moments for each second of the “event” (the data is sampled at 25hz).
让我们看一下第一个事件。 第一个事件具有关联的事件ID号。 我们稍后将使用它们。 还有来访和主队中每个球员的数据。 我们也将在以后使用它们。 最后，最重要的是，存在“时刻”。 “事件”的每一秒有25个瞬间（数据以25hz采样）。
data['events'][0].keys()data [ 'events' ][ 0 ] . keys () |
Here’s the first moment of the first event. The first number is the quarter. The second number is the time of the event in milliseconds. The third number is the number of seconds left in the quarter (the 1st quarter hasn’t started yet, so 12 * 60 = 720). The fourth number is the number of seconds left on the shot clock. I am not sure what fourth number (None) represents.
这是第一场比赛的第一刻。 第一个数字是四分之一。 第二个数字是事件的时间（以毫秒为单位）。 第三个数字是该季度剩余的秒数（第一个季度尚未开始，因此12 * 60 = 720）。 第四个数字是射击时钟上剩余的秒数。 我不确定第四个数字（无）代表什么。
The final matrix is 11×5 matrix. The first row describes the ball. The first two columns are the teamID and the playerID of the ball (-1 for both because the ball does not belong to a team and is not a player). The 3rd and 4th columns are xy coordinates of the ball. The final column is the height of the ball (z coordinate).
最终矩阵是11×5矩阵。 第一行描述了球。 前两列是球的teamID和玩家ID（对于两个都为-1，因为球不属于团队且不是玩家）。 第三和第四列是球的xy坐标。 最后一列是球的高度（z坐标）。
The next 10 rows describe the 10 players on the court. The first 5 players belong to the home team and the last 5 players belong to the visiting team. Each player has his teamID, playerID, xy&z coordinates (although I don’t think players’ z coordinates ever change).
接下来的10行描述了球场上的10位玩家。 前5名球员属于主队，后5名球员属于客队。 每个玩家都有自己的teamID，playerID，xy＆z坐标（尽管我认为玩家的z坐标不会改变）。
11 |
[1, 1452903036782, 720.0, 24.0, None, [[-1, -1, 44.16456, 26.34142, 5.74423], [1610612760, 201142, 45.46259, 32.01456, 0.0], [1610612760, 201566, 10.39347, 24.77219, 0.0], [1610612760, 201586, 25.86087, 25.55881, 0.0], [1610612760, 203460, 47.28525, 17.76225, 0.0], [1610612760, 203500, 43.68634, 26.63098, 0.0], [1610612750, 708, 55.6401, 25.55583, 0.0], [1610612750, 2419, 47.95942, 31.66328, 0.0], [1610612750, 201937, 67.28725, 25.10267, 0.0], [1610612750, 203952, 47.28525, 17.76225, 0.0], [1610612750, 1626157, 49.46814, 24.24193, 0.0]]][1, 1452903036782, 720.0, 24.0, None, [[-1, -1, 44.16456, 26.34142, 5.74423], [1610612760, 201142, 45.46259, 32.01456, 0.0], [1610612760, 201566, 10.39347, 24.77219, 0.0], [1610612760, 201586, 25.86087, 25.55881, 0.0], [1610612760, 203460, 47.28525, 17.76225, 0.0], [1610612760, 203500, 43.68634, 26.63098, 0.0], [1610612750, 708, 55.6401, 25.55583, 0.0], [1610612750, 2419, 47.95942, 31.66328, 0.0], [1610612750, 201937, 67.28725, 25.10267, 0.0], [1610612750, 203952, 47.28525, 17.76225, 0.0], [1610612750, 1626157, 49.46814, 24.24193, 0.0]]]
Alright, so we have the sportsvu data, but its not clear what each event is. Luckily, the NBA also provides play by play (pbp) data. I write a function for acquiring play by play game data. This function collects (and trims) the play by play data for a given sportsvu data set.
好了，所以我们有了sportsvu数据，但不清楚每个事件是什么。 幸运的是，NBA还提供逐场比赛（pbp）数据。 我编写了一个通过玩游戏数据获取玩游戏的功能。 此功能针对给定的sportsvu数据集按播放数据收集（和修剪）该播放。
def acquire_gameData(data): import requests header_data = { #I pulled this header from the py goldsberry library 'Accept-Encoding': 'gzip, deflate, sdch', 'Accept-Language': 'en-US,en;q=0.8', 'Upgrade-Insecure-Requests': '1', 'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; WOW64)' ' AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.82 ' 'Safari/537.36', 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9' ',image/webp,*/*;q=0.8', 'Cache-Control': 'max-age=0', 'Connection': 'keep-alive' } game_url = 'http://stats.nba.com/stats/playbyplayv2?EndPeriod=0&EndRange=0&GameID='+data['gameid']+ '&RangeType=0&StartPeriod=0&StartRange=0' #address for querying the data response = requests.get(game_url,headers = header_data) #go get the data headers = response.json()['resultSets'][0]['headers'] #get headers of data gameData = response.json()['resultSets'][0]['rowSet'] #get actual data from json object df = pd.DataFrame(gameData, columns=headers) #turn the data into a pandas dataframe df = df[[df.columns[1], df.columns[2],df.columns[7],df.columns[9],df.columns[18]]] #there's a ton of data here, so I trim it doown df['TEAM'] = df['PLAYER1_TEAM_ABBREVIATION'] df = df.drop('PLAYER1_TEAM_ABBREVIATION', 1) return dfdef acquire_gameData ( data ): import requests header_data = { #I pulled this header from the py goldsberry library 'Accept-Encoding' : 'gzip, deflate, sdch' , 'Accept-Language' : 'en-US,en;q=0.8' , 'Upgrade-Insecure-Requests' : '1' , 'User-Agent' : 'Mozilla/5.0 (Windows NT 10.0; WOW64)' ' AppleWebKit/537.36 (KHTML, like Gecko) Chrome/48.0.2564.82 ' 'Safari/537.36' , 'Accept' : 'text/html,application/xhtml+xml,application/xml;q=0.9' ',image/webp,*/*;q=0.8' , 'Cache-Control' : 'max-age=0' , 'Connection' : 'keep-alive' } game_url = 'http://stats.nba.com/stats/playbyplayv2?EndPeriod=0&EndRange=0&GameID=' + data [ 'gameid' ] + '&RangeType=0&StartPeriod=0&StartRange=0' #address for querying the data response = requests . get ( game_url , headers = header_data ) #go get the data headers = response . json ()[ 'resultSets' ][ 0 ][ 'headers' ] #get headers of data gameData = response . json ()[ 'resultSets' ][ 0 ][ 'rowSet' ] #get actual data from json object df = pd . DataFrame ( gameData , columns = headers ) #turn the data into a pandas dataframe df = df [[ df . columns [ 1 ], df . columns [ 2 ], df . columns [ 7 ], df . columns [ 9 ], df . columns [ 18 ]]] #there's a ton of data here, so I trim it doown df [ 'TEAM' ] = df [ 'PLAYER1_TEAM_ABBREVIATION' ] df = df . drop ( 'PLAYER1_TEAM_ABBREVIATION' , 1 ) return df |
Below I show what the play by play data looks like. There’s a column for event number (eventnum). These event numbers match up with the event numbers from the sportsvu data, so we will use this later for seeking out specific plays in the sportsvu data. There’s a column for the event type (eventmsgtype). This column has a number describing what occured in the play. I list these number codes in the comments below.
在下面，我显示了逐次播放数据的样子。 有一个事件号（eventnum）列。 这些事件编号与来自sportsvu数据的事件编号匹配，因此我们稍后将使用它来查找sportsvu数据中的特定比赛。 这里有一个事件类型列（eventmsgtype）。 此列中有一个数字，描述剧中发生的事情。 我在下面的评论中列出了这些数字代码。
There’s also short text descriptions of the plays in the home description and visitor description columns. Finally, I use the team column to represent the primary team involved in a play.
在首页说明和访客说明列中也有关于剧本的简短文字说明。 最后，我使用“团队”列来表示比赛中涉及的主要团队。
I stole the idea of using play by play data from .
我偷走了 （ 使用比赛数据的想法。
df = acquire_gameData(data)df.head()#EVENTMSGTYPE#1 - Make #2 - Miss #3 - Free Throw #4 - Rebound #5 - out of bounds / Turnover / Steal #6 - Personal Foul #7 - Violation #8 - Substitution #9 - Timeout #10 - Jumpball #12 - Start Q1? #13 - Start Q2?df = acquire_gameData ( data ) df . head () #EVENTMSGTYPE #1 - Make #2 - Miss #3 - Free Throw #4 - Rebound #5 - out of bounds / Turnover / Steal #6 - Personal Foul #7 - Violation #8 - Substitution #9 - Timeout #10 - Jumpball #12 - Start Q1? #13 - Start Q2? |
| EVENTNUM | 事件编号 | EVENTMSGTYPE | 偶数类型 | HOMEDESCRIPTION | 家庭说明 | VISITORDESCRIPTION | 访客说明 | TEAM | 球队 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 0 | 0 | 12 | 12 | None | 没有 | None | 没有 | None | 没有 |
| 1 | 1个 | 1 | 1个 | 10 | 10 | Jump Ball Adams vs. Towns: Tip to Ibaka | 亚当斯（Adams）vs. Towns（跳城镇）：伊巴卡（Ibaka）提示 | None | 没有 | OKC | OKC |
| 2 | 2 | 2 | 2 | 5 | 5 | Westbrook Out of Bounds Lost Ball Turnover (P1… | 威斯布鲁克超出禁区丢球失误（P1… | None | 没有 | OKC | OKC |
| 3 | 3 | 3 | 3 | 2 | 2 | None | 没有 | MISS Wiggins 16′ Jump Shot | 威金斯小姐16′跳投 | MIN | 最低 |
| 4 | 4 | 4 | 4 | 4 | 4 | Westbrook REBOUND (Off:0 Def:1) | 威斯布鲁克反弹（Off：0 Def：1） | None | 没有 | OKC | OKC |
When viewing the videos, its nice to know what players are on the court. I like to depict this by labeling each player with their number. Here I create a dictionary that contains each player’s id number (these are assigned by nba.com) as the key and their jersey number as the associated value.
观看视频时，很高兴知道场上有哪些球员。 我喜欢用每个球员的号码来标记他们的位置。 在这里，我创建了一个词典，其中包含每个玩家的ID号（这些密钥由nba.com分配）作为键，而其球衣号码作为关联值。
player_fields = data['events'][0]['home']['players'][0].keys()home_players = pd.DataFrame(data=[i for i in data['events'][0]['home']['players']], columns=player_fields)away_players = pd.DataFrame(data=[i for i in data['events'][0]['visitor']['players']], columns=player_fields)players = pd.merge(home_players, away_players, how='outer')jerseydict = dict(zip(players.playerid.values, players.jersey.values))player_fields = data [ 'events' ][ 0 ][ 'home' ][ 'players' ][ 0 ] . keys () home_players = pd . DataFrame ( data = [ i for i in data [ 'events' ][ 0 ][ 'home' ][ 'players' ]], columns = player_fields ) away_players = pd . DataFrame ( data = [ i for i in data [ 'events' ][ 0 ][ 'visitor' ][ 'players' ]], columns = player_fields ) players = pd . merge ( home_players , away_players , how = 'outer' ) jerseydict = dict ( zip ( players . playerid . values , players . jersey . values )) |
Alright, almost there! Below I write some functions for creating the actual video! First, there’s a short function for placing an image of the basketball court beneath our depiction of players moving around. This image is from gmf05’s github, but I will provide it on too.
好吧，快到了！ 在下面，我写了一些用于创建实际视频的功能！ 首先，有一个简短的功能将篮球场的图像放置在我们描绘的运动员走动下方。 该图像来自gmf05的github，但我也将在图像上提供它。
Much of this code is either straight from or slightly modified.
大部分代码要么直接来自要么进行了稍微的修改。
# Animation function / loopdef draw_court(axis): import matplotlib.image as mpimg img = mpimg.imread('./nba_court_T.png') #read image. I got this image from gmf05's github. plt.imshow(img,extent=axis, zorder=0) #show the image. def animate(n): #matplotlib's animation function loops through a function n times that draws a different frame on each iteration for i,ii in enumerate(player_xy[n]): #loop through all the players player_circ[i].center = (ii[1], ii[2]) #change each players xy position player_text[i].set_text(str(jerseydict[ii[0]])) #draw the text for each player. player_text[i].set_x(ii[1]) #set the text x position player_text[i].set_y(ii[2]) #set text y position ball_circ.center = (ball_xy[n,0],ball_xy[n,1]) #change ball xy position ball_circ.radius = 1.1 #i could change the size of the ball according to its height, but chose to keep this constant return tuple(player_text) + tuple(player_circ) + (ball_circ,)def init(): #this is what matplotlib's animation will create before drawing the first frame. for i in range(10): #set up players player_text[i].set_text('') ax.add_patch(player_circ[i]) ax.add_patch(ball_circ) #create ball ax.axis('off') #turn off axis dx = 5 plt.xlim([0-dx,100+dx]) #set axis plt.ylim([0-dx,50+dx]) return tuple(player_text) + tuple(player_circ) + (ball_circ,)# Animation function / loop def draw_court ( axis ): import matplotlib.image as mpimg img = mpimg . imread ( './nba_court_T.png' ) #read image. I got this image from gmf05's github. plt . imshow ( img , extent = axis , zorder = 0 ) #show the image. def animate ( n ): #matplotlib's animation function loops through a function n times that draws a different frame on each iteration for i , ii in enumerate ( player_xy [ n ]): #loop through all the players player_circ [ i ] . center = ( ii [ 1 ], ii [ 2 ]) #change each players xy position player_text [ i ] . set_text ( str ( jerseydict [ ii [ 0 ]])) #draw the text for each player. player_text [ i ] . set_x ( ii [ 1 ]) #set the text x position player_text [ i ] . set_y ( ii [ 2 ]) #set text y position ball_circ . center = ( ball_xy [ n , 0 ], ball_xy [ n , 1 ]) #change ball xy position ball_circ . radius = 1.1 #i could change the size of the ball according to its height, but chose to keep this constant return tuple ( player_text ) + tuple ( player_circ ) + ( ball_circ ,) def init (): #this is what matplotlib's animation will create before drawing the first frame. for i in range ( 10 ): #set up players player_text [ i ] . set_text ( '' ) ax . add_patch ( player_circ [ i ]) ax . add_patch ( ball_circ ) #create ball ax . axis ( 'off' ) #turn off axis dx = 5 plt . xlim ([ 0 - dx , 100 + dx ]) #set axis plt . ylim ([ 0 - dx , 50 + dx ]) return tuple ( player_text ) + tuple ( player_circ ) + ( ball_circ ,) |
The event that I want to depict is event 41. In this event, Karl Anthony Towns misses a shot, grabs his own rebounds, and puts it back in.
我要描述的事件是事件41。在此事件中，卡尔·安东尼·汤斯（Karl Anthony Towns）投篮未中，抢下自己的篮板，然后放回去。
df[37:38]df [ 37 : 38 ] |
| EVENTNUM | 事件编号 | EVENTMSGTYPE | 偶数类型 | HOMEDESCRIPTION | 家庭说明 | VISITORDESCRIPTION | 访客说明 | TEAM | 球队 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 37 | 37 | 41 | 41 | 1 | 1个 | None | 没有 | Towns 1′ Layup (2 PTS) | Towns 1′Layup（2分） | MIN | 最低 |
We need to find where event 41 is in the sportsvu data structure, so I created a function for finding the location of a particular event. I then create a matrix with position data for the ball and a matrix with position data for each player for event 41.
我们需要找到事件41在sportsvu数据结构中的位置，因此我创建了一个用于查找特定事件位置的函数。 然后，我为事件41创建一个包含球位置数据的矩阵和一个包含每个球员位置数据的矩阵。
#the order of events does not match up, so we have to use the eventIds. This loop finds the correct event for a given id#.search_id = 41def find_moment(search_id): for i,events in enumerate(data['events']): if events['eventId'] == str(search_id): finder = i break return finderevent_num = find_moment(search_id)ball_xy = np.array([x[5][0][2:5] for x in data['events'][event_num]['moments']]) #create matrix of ball dataplayer_xy = np.array([np.array(x[5][1:])[:,1:4] for x in data['events'][event_num]['moments']]) #create matrix of player data#the order of events does not match up, so we have to use the eventIds. This loop finds the correct event for a given id#. search_id = 41 def find_moment ( search_id ): for i , events in enumerate ( data [ 'events' ]): if events [ 'eventId' ] == str ( search_id ): finder = i break return finder event_num = find_moment ( search_id ) ball_xy = np . array ([ x [ 5 ][ 0 ][ 2 : 5 ] for x in data [ 'events' ][ event_num ][ 'moments' ]]) #create matrix of ball data player_xy = np . array ([ np . array ( x [ 5 ][ 1 :])[:, 1 : 4 ] for x in data [ 'events' ][ event_num ][ 'moments' ]]) #create matrix of player data |
Okay. We’re actually there! Now we get to create the video. We have to create figure and axes objects for the animation to draw on. Then I place a picture of the basketball court on this plot. Finally, I create the circle and text objects that will move around throughout the video (depicting the ball and players). The location of these objects are then updated in the animation loop.
好的。 我们实际上在那里！ 现在我们来创建视频。 我们必须创建图形对象和坐标轴对象才能绘制动画。 然后，我将篮球场的图片放在该图上。 最后，我创建了将在整个视频中移动的圆圈和文本对象（描述了球和球员）。 然后在动画循环中更新这些对象的位置。
import matplotlib.animation as animationfig = plt.figure(figsize=(15,7.5)) #create figure objectax = plt.gca() #create axis objectdraw_court([0,100,0,50]) #draw the courtplayer_text = range(10) #create player text vectorplayer_circ = range(10) #create player circle vectorball_circ = plt.Circle((0,0), 1.1, color=[1, 0.4, 0]) #create circle object for balfor i in range(10): #create circle object and text object for each player col=['w','k'] if i<5 else ['k','w'] #color scheme player_circ[i] = plt.Circle((0,0), 2.2, facecolor=col[0],edgecolor='k') #player circle player_text[i] = ax.text(0,0,'',color=col[1],ha='center',va='center') #player jersey # (text)ani = animation.FuncAnimation(fig, animate, frames=np.arange(0,np.size(ball_xy,0)), init_func=init, blit=True, interval=5, repeat=False, save_count=0) #function for making videoani.save('Event_%d.mp4' % (search_id),dpi=100,fps=25) #function for saving videoplt.close('all') #close the plotimport matplotlib.animation as animation fig = plt . figure ( figsize = ( 15 , 7.5 )) #create figure object ax = plt . gca () #create axis object draw_court ([ 0 , 100 , 0 , 50 ]) #draw the court player_text = range ( 10 ) #create player text vector player_circ = range ( 10 ) #create player circle vector ball_circ = plt . Circle (( 0 , 0 ), 1.1 , color = [ 1 , 0.4 , 0 ]) #create circle object for bal for i in range ( 10 ): #create circle object and text object for each player col = [ 'w' , 'k' ] if i < 5 else [ 'k' , 'w' ] #color scheme player_circ [ i ] = plt . Circle (( 0 , 0 ), 2.2 , facecolor = col [ 0 ], edgecolor = 'k' ) #player circle player_text [ i ] = ax . text ( 0 , 0 , '' , color = col [ 1 ], ha = 'center' , va = 'center' ) #player jersey # (text) ani = animation . FuncAnimation ( fig , animate , frames = np . arange ( 0 , np . size ( ball_xy , 0 )), init_func = init , blit = True , interval = 5 , repeat = False , save_count = 0 ) #function for making video ani . save ( 'Event_ %d .mp4' % ( search_id ), dpi = 100 , fps = 25 ) #function for saving video plt . close ( 'all' ) #close the plot |
翻译自:
转载地址：http://chqwd.baihongyu.com/