Create Player DF

My initial focus is on importing the text file and then storing the relevant data into a dataframe that can be used for additional pre-processing

data_string = readLines(file_name, warn = F)

cols = c('pair_num', 'player_name','player_points', 'player_state', 'pre_rating',
         'round1_opponent', 'round2_opponent', 'round3_opponent', 'round4_opponent',
         'round5_opponent', 'round6_opponent', 'round7_opponent', 'avg_opponent_rating')

player_matrix = matrix(nrow=64, ncol=length(cols))
colnames(player_matrix) = cols
player_df = as_tibble(player_matrix)


player_index = 1
sub_index = 2

while(player_index<=64) {
  row1 = data_string[2*player_index+sub_index+1]
  row2 = data_string[2*player_index+sub_index+2]
  
  row1_string_vect = str_split(row1,'[|]')
  row2_string_vect = str_split(row2,'[|]')
  
  
  
  player_df$pair_num[player_index] = row1_string_vect[[1]][1]
  player_df$player_name[player_index] = row1_string_vect[[1]][2]
  player_df$player_points[player_index] = row1_string_vect[[1]][3]
  player_df$player_state[player_index] = row2_string_vect[[1]][1]
  player_df$pre_rating[player_index] = row2_string_vect[[1]][2]
  player_df$round1_opponent[player_index] = row1_string_vect[[1]][4]
  player_df$round2_opponent[player_index] = row1_string_vect[[1]][5]
  player_df$round3_opponent[player_index] = row1_string_vect[[1]][6]
  player_df$round4_opponent[player_index] = row1_string_vect[[1]][7]
  player_df$round5_opponent[player_index] = row1_string_vect[[1]][8]
  player_df$round6_opponent[player_index] = row1_string_vect[[1]][9]
  player_df$round7_opponent[player_index] = row1_string_vect[[1]][10]
  
  #print(pre_rating)
  player_index=player_index+1
  sub_index = sub_index+1
}

Clean Up Dataframe

In this data cleaning chunk I want to complete the following pre-processing steps for the following columns:

  1. pair_num - remove the extra white space; no need to convert it into an integer since there will be no mathematical computation performed on it
  2. player_name - remove the front and ending white space characters
  3. player_points - remove the front and ending white space characters
  4. player_state - remove the front and ending white space characters
  5. pre_rating - remove the front and ending white space characters; extract just the pre-rating portion of the character string
  6. roundx_opponent - extract only the numeric portion of the field (for each opponent in rounds 1 through 7)
#Clean up the white space in the first four columns
player_df = player_df %>% mutate(pair_num = str_squish(pair_num),
                     player_name = str_squish(player_name),
                     player_points = str_squish(player_points),
                     player_state = str_squish(player_state),
                     pre_rating = str_squish(pre_rating))

#Capture the player's pre-rating
player_df = player_df %>% mutate(pre_rating = str_extract(str_extract_all(player_df$pre_rating,'R:\\s+[\\d]+'),'[\\d]+'))

#Capture the player's opponent number for each round

player_df = player_df %>% mutate(round1_opponent = str_extract(round1_opponent, '[\\d]+'),
                     round2_opponent = str_extract(round2_opponent, '[\\d]+'),
                     round3_opponent = str_extract(round3_opponent, '[\\d]+'),
                     round4_opponent = str_extract(round4_opponent, '[\\d]+'),
                     round5_opponent = str_extract(round5_opponent, '[\\d]+'),
                     round6_opponent = str_extract(round6_opponent, '[\\d]+'),
                     round7_opponent = str_extract(round7_opponent, '[\\d]+'))

Calculate the Average Pre-Chess Rating of Opponents

This block of code loops through the player dataframe and calculates the average pre-chess rating of the opponents for each player

#Create a function that takes a row and calculates the average pre-match ranking for each opponent
calc_average = function(df_row) {

  
  round1_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round1_opponent))$pre_rating
  round2_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round2_opponent))$pre_rating
  round3_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round3_opponent))$pre_rating
  round4_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round4_opponent))$pre_rating
  round5_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round5_opponent))$pre_rating
  round6_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round6_opponent))$pre_rating
  round7_opponent_score = (player_df %>% filter(player_df$pair_num == df_row$round7_opponent))$pre_rating
  
  pre_rating_scores = c(round1_opponent_score,round2_opponent_score,round3_opponent_score,
                        round4_opponent_score,round5_opponent_score,round6_opponent_score,
                        round7_opponent_score)
  
  average_opponent_pre_rating = mean(as.numeric(pre_rating_scores))
  
  
  
  return(average_opponent_pre_rating)
  
}

for(i in 1:nrow(player_df)) {
  avg_score = calc_average(player_df[i,])
  player_df[i,]$avg_opponent_rating = round(avg_score,0)
}

Simplify the Dataframe and export the output

This block of code creates a simplified dataframe with only the requested fields and then exports it to a .csv file.

simple_df = player_df %>% select(c('player_name','player_state','player_points','pre_rating', 'avg_opponent_rating'))

knitr::kable(head(simple_df))
player_name player_state player_points pre_rating avg_opponent_rating
GARY HUA ON 6.0 1794 1605
DAKSHESH DARURI MI 6.0 1553 1469
ADITYA BAJAJ MI 6.0 1384 1564
PATRICK H SCHILLING MI 5.5 1716 1574
HANSHI ZUO MI 5.5 1655 1501
HANSEN SONG OH 5.0 1686 1519 
write_csv(simple_df, 'KMartin_project1_output.csv')