如何在R语言中从proxy包dist函数生成的crossmat类型矩阵提取0<值<5的元素并转换为指定格式DataFrame

Got it, let's work through this problem to get exactly the output you need. Here's a step-by-step solution tailored to your use case:

Step 1: Recap & Setup #

First, let's confirm we're working with the crossmat object generated by proxy::dist()—this is essentially a matrix where rows map to your df1 entries and columns map to df2 entries. We need to pull out values between 0 and 5, while retaining the original row/column indices from df1 and df2 (not the matrix's relative positions).

Step 2: Code Implementation #

Here's the full code to achieve your desired result:

# Load the proxy package (if not already loaded)
library(proxy)

# Your existing code to create the distance matrix
a <- df1
b <- df2
dm <- dist(a, b, method = "euclidean")

# 1. Find positions in the matrix where values are between 0 and 5
# arr.ind = TRUE returns matrix row/column positions as a data frame
filter_positions <- which(dm > 0 & dm < 5, arr.ind = TRUE)

# 2. Map matrix positions to original df1/df2 indices
# Matrix rows correspond to df1 rows—use df1's rownames for original indices
original_rows <- rownames(a)[filter_positions[, "row"]]
# Matrix columns correspond to df2 rows—use df2's rownames for original indices
original_cols <- rownames(b)[filter_positions[, "col"]]

# 3. Extract the filtered values
filtered_values <- dm[filter_positions]

# 4. Combine into the final DataFrame
result_df <- data.frame(
  row = as.integer(original_rows),
  col = as.integer(original_cols),
  value = filtered_values,
  stringsAsFactors = FALSE
)

# View the result
head(result_df)

Key Notes: #

• Original Indices: This code assumes your df1 and df2 have their original row indices stored as rownames() (like the 3672, 192 values in your example). If you store original indices in a separate column (e.g., df1$original_row), replace rownames(a) with a$original_row (and similarly for df2).
• Data Types: Using as.integer() converts the character rownames back to integer indices, matching your sample output format.
• Filter Logic: The which() function efficiently finds all positions where values meet your 0 < value < 5 condition.

Example Output #

Running this code will produce a DataFrame structured exactly like your sample:

row  col     value
1 3672  192 4.963545
2 3673  192 4.848763
3 3672  193 4.769458
4 3673  193 4.645692
5 3674  193 4.889749
6 3672  194 4.125620

内容的提问来源于stack exchange，提问作者Nancy