英文:
Find max time difference within each year in R
问题
这是您要翻译的内容：
"I have a function that calculates the average, min and max values for each year in my dataframe, then merges them to output the alltime average, min and max values. Each year needs to be calculated separately first because my dates only refer to the months of April through August. If I didn't group it by year, there would be calculations between August of one year and April of the next year. I want to avoid this.
Example dataframe:
date NDVI cloud_cover field_id23/04/2017 0.6494 12 KM6023/04/2017 0.5683 0 KM105/05/2017 0.3467 0 KM6031/07/2017 0.6743 05 KM6031/07/2017 NA 97 KM131/07/2017 0.3456 07 LM2701/04/2018 NA 100 KM6003/06/2018 0.6743 11 KM6003/06/2018 0.2346 12 KM104/05/2019 NA 99 KM6005/05/2019 0.5432 20 KM60
NDVI and cloud_cover shouldn't influence calculations. Although field_ids most times provide the same dates, this also shouldn't influence them.
This is the current code:
calculate_time_diff <- function(df) {# Convert "date" column to datetimedf$date <- as.POSIXct(df$date)# Group the data by yeardf_calc <- split(df, format(df$date, "%Y"))# Calculate time differences between consecutive observations for each yeartime_diffs <- lapply(df_calc, function(group) {# Sort dataframe based on "date"group <- group[order(group$date), ]# Filter out duplicate datesgroup <- group[!duplicated(group$date), ]# Calculate time differences between consecutive observationsdiff(group$date)})# Combine time differences from all years into a single vectorall_time_diffs <- unlist(time_diffs)# Compute average time differenceavg_time_diff <- mean(all_time_diffs)# Calculate smallest and biggest time differencessmallest_time_diff <- min(all_time_diffs)biggest_time_diff <- max(all_time_diffs)return(list(avg_time_diff = avg_time_diff,smallest_time_diff = smallest_time_diff,biggest_time_diff = biggest_time_diff))}
The output is giving me "240" as max time difference, which I know to be unrealistic. My dataframe refers to the revisit dates of three satellites and none of them should be more than at the very most a month apart.
I thought it could have something to do with the way years are being extracted, but this user seems to have successfully used format() just as I did. lapply() should iterate through each split year group in the same way as group_by(). So what could be the problem in my script?"
英文:
I have a function that calculates the average, min and max values for each year in my dataframe, then merges them to output the alltime average, min and max values. Each year needs to be calculated separately first because my dates only refer to the months of April through August. If I didn't group it by year, there would be calculations between August of one year and April of the next year. I want to avoid this.
Example dataframe:
date NDVI cloud_cover field_id23/04/2017 0.6494 12 KM6023/04/2017 0.5683 0 KM105/05/2017 0.3467 0 KM6031/07/2017 0.6743 05 KM6031/07/2017 NA 97 KM131/07/2017 0.3456 07 LM2701/04/2018 NA 100 KM6003/06/2018 0.6743 11 KM6003/06/2018 0.2346 12 KM104/05/2019 NA 99 KM6005/05/2019 0.5432 20 KM60
NDVI and cloud_cover shouldn't influence calculations. Although field_ids most times provide the same dates, this also shouldn't influence them.
This is the current code:
calculate_time_diff <- function(df) {# Convert "date" column to datetimedf$date <- as.POSIXct(df$date)# Group the data by yeardf_calc <- split(df, format(df$date, "%Y"))# Calculate time differences between consecutive observations for each yeartime_diffs <- lapply(df_calc, function(group) {# Sort dataframe based on "date"group <- group[order(group$date), ]# Filter out duplicate datesgroup <- group[!duplicated(group$date), ]# Calculate time differences between consecutive observationsdiff(group$date)})# Combine time differences from all years into a single vectorall_time_diffs <- unlist(time_diffs)# Compute average time differenceavg_time_diff <- mean(all_time_diffs)# Calculate smallest and biggest time differencessmallest_time_diff <- min(all_time_diffs)biggest_time_diff <- max(all_time_diffs)return(list(avg_time_diff = avg_time_diff,smallest_time_diff = smallest_time_diff,biggest_time_diff = biggest_time_diff))}
The output is giving me "240" as max time difference, which I know to be unrealistic. My dataframe refers to the revisit dates of three satellites and none of them should be more than at the very most a month apart.
I thought it could have something to do with the way years are being extracted, but this user seems to have successfully used format() just as I did. lapply() should iterate through each split year group in the same way as group_by(). So what could be the problem in my script?
答案1
得分: 0
Using dplyr:
data %>%distinct(date) %>%arrange(date) %>%group_by(format(date, "%Y")) %>%reframe(dateDiff = diff(date)) %>%with(list(avg_time_diff = mean(dateDiff),smallest_time_diff = min(dateDiff),biggest_time_diff = max(dateDiff)))
Result:
$avg_time_diffTime difference of 30.02198 days$smallest_time_diffTime difference of 12 days$biggest_time_diffTime difference of 51 days
Dummy data:
data <- data.frame(date = seq(as.Date("2017-01-01"), by = "month", length.out = 100) + sample(0:20, 100, TRUE))
英文:
Using dplyr:
data %>%distinct(date) %>% #remove duplicatesarrange(date) %>% #order by dategroup_by(format(date, "%Y")) %>% #group by yearreframe(dateDiff = diff(date)) %>% #apply 'diff' to every groupwith(list(avg_time_diff = mean(dateDiff),smallest_time_diff = min(dateDiff),biggest_time_diff = max(dateDiff))) #create your metrics
Result:
$avg_time_diffTime difference of 30.02198 days$smallest_time_diffTime difference of 12 days$biggest_time_diffTime difference of 51 days
Dummy data:
data <- data.frame(date = seq(as.Date("2017-01-01"), by = "month", length.out = 100) + sample(0:20, 100, TRUE))
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论