基于字典映射列值并按行求和的R可扩展实现方案问询

Problem Overview #

You need to map dataframe columns using named dictionaries, compute row-wise sums for each dictionary's mapped values, and optionally combine these sums into a total score. The goal is a scalable approach that avoids inefficient loops.

Solution Code #

We'll use vectorized operations (via dplyr) for efficiency and scalability. This approach works for any number of columns or dictionaries.

Step 1: Define the Scalable Function #

library(dplyr)

compute_score <- function(data, dict, score_name = "score", keep_mapped = FALSE) {
  # Generate names for mapped columns
  mapped_col_names <- paste0(names(dict), "_mapped")
  
  data %>%
    # Create mapped columns: multiply each column by its corresponding dict value
    mutate(across(all_of(names(dict)), ~ .x * dict[[cur_column()]], .names = "{.col}_mapped")) %>%
    # Calculate row-wise sum of mapped columns to get the score
    mutate(!!score_name := rowSums(select(., all_of(mapped_col_names)))) %>%
    # Remove intermediate mapped columns unless specified to keep them
    { if (!keep_mapped) select(., -all_of(mapped_col_names)) else . }
}

Step 2: Apply to Single Dictionary #

# Sample data
df = data.frame(
  sex = c(0,0,0,1,0,1,0,1,1,1,0),
  icu = c(1,1,0,1,0,1,0,1,1,1,1),
  niv = c(0,1,0,1,0,1,0,0,0,1,0),
  mv = c(1,0,0,1,1,1,0,0,0,1,0),
  o2 = c(1,0,0,1,0,1,0,0,0,1,0)
)

# Mapping dictionaries
dict1 <- list(sex = 0, icu = 2, niv = 1, mv = 3, o2 = 2)
dict2 <- list(sex = 3, icu = 4, niv = 2, mv = 6, o2 = 1)

# Compute score for dict1
df <- compute_score(df, dict1, score_name = "score_dict1")

Step 3: Apply to Multiple Dictionaries #

# Compute score for dict2
df <- compute_score(df, dict2, score_name = "score_dict2")

# Calculate total score by summing all individual scores
df <- df %>% mutate(total_score = score_dict1 + score_dict2)

Step 4: Batch Processing for N Dictionaries #

If you have multiple dictionaries stored in a list, you can loop through them efficiently:

# Store dictionaries in a named list
dict_list <- list(dict1 = dict1, dict2 = dict2)

# Initialize result dataframe
result_df <- df

# Apply function to each dictionary
for (dict_name in names(dict_list)) {
  result_df <- compute_score(result_df, dict_list[[dict_name]], score_name = paste0("score_", dict_name))
}

# Compute total score across all dictionary scores
result_df <- result_df %>% mutate(total_score = rowSums(select(., starts_with("score_"))))

Key Advantages Over Loop-Based Approach #

• Vectorized Operations: Avoids slow row-wise loops, making it significantly faster for large datasets.
• Scalability: Works seamlessly with any number of columns (matching dict keys) or dictionaries.
• Flexibility: Option to keep intermediate mapped columns or discard them.
• Readability: Uses modern dplyr syntax for clear, maintainable code.

Base R Alternative (No dplyr Required) #

If you prefer base R, use this function:

compute_score_base <- function(data, dict, score_name = "score", keep_mapped = FALSE) {
  # Create mapped columns
  mapped_cols <- lapply(names(dict), function(col) data[[col]] * dict[[col]])
  names(mapped_cols) <- paste0(names(dict), "_mapped")
  
  # Combine mapped columns with original data
  data <- cbind(data, mapped_cols)
  
  # Calculate row-wise sum for score
  data[[score_name]] <- rowSums(data[, names(mapped_cols)])
  
  # Remove mapped columns if requested
  if (!keep_mapped) {
    data <- data[, !names(data) %in% names(mapped_cols)]
  }
  
  return(data)
}

Content of the question originates from Stack Exchange, question author Henrique