Python LSTM Keras中的生成器函数,用于输出一个文件的小批量
我有一个发电机功能,工作正常。我有一个很大的.txt文件列表,其中每个文件都很长。现在的任务是编写一个生成器函数,该函数采用:Python LSTM Keras中的生成器函数,用于输出一个文件的小批量,python,r,tensorflow,keras,mini-batch,Python,R,Tensorflow,Keras,Mini Batch,我有一个发电机功能,工作正常。我有一个很大的.txt文件列表,其中每个文件都很长。现在的任务是编写一个生成器函数,该函数采用: 一批文件 然后从一个文件中取出一批大小为128的文件 我现在的代码: data_files_generator <- function(train_set) { files <- train_set next_file <- 0 function() { # move to the next file (note the &l
data_files_generator <- function(train_set) {
files <- train_set
next_file <- 0
function() {
# move to the next file (note the <<- assignment operator)
next_file <<- next_file + 1
# if we've exhausted all of the files then start again at the
# beginning of the list (keras generators need to yield
# data infinitely -- termination is controlled by the epochs
# and steps_per_epoch arguments to fit_generator())
if (next_file > length(files))
{next_file <<- 1}
# determine the file name
file <- files[[next_file]]
text <- read_lines(paste(data_dir, file, sep = "" )) %>%
str_to_lower() %>%
str_c(collapse = "\n") %>%
removeNumbers() %>%
tokenize_characters(strip_non_alphanum = FALSE, simplify = TRUE)
text <- text[text %in% chars]
dataset <- map(
seq(1, length(text) - maxlen - 1, by = 3),
~list(sentece = text[.x:(.x + maxlen - 1)], next_char = text[.x + maxlen])
)
dataset <- transpose(dataset)
# Vectorization
x <- array(0, dim = c(length(dataset$sentece), maxlen, length(chars)))
y <- array(0, dim = c(length(dataset$sentece), length(chars)))
for(i in 1:length(dataset$sentece)){
x[i,,] <- sapply(chars, function(x){
as.integer(x == dataset$sentece[[i]])
})
y[i,] <- as.integer(chars == dataset$next_char[[i]])
}
rounded_dim <- floor(dim(x)[1]/mini_batch_size)
match_size_to_batch <- 128 * rounded_dim
x <- x[1:match_size_to_batch, 1:maxlen, 1:length(chars)]
y <- y_val[1:match_size_to_batch, 1:length(chars)]
return(list(x, y))
}
}
希望我已经解释清楚了。我想我必须输入某种for循环来迭代样本长度,但我不知道如何将其包含到gen.函数中。根据错误,您试图输入一个shape
(112512,40,43)
的对象,但您的LSTM层需要一个shape(128,40,43)
的对象。似乎缺少一些代码,但在定义输入层时,是否正在修复批大小?我很幸运地将我的输入层定义为:
l_input = Input(shape = (None, num_features), name = 'input_layer')
我怀疑错误是由以下代码行引起的:
rounded_dim <- floor(dim(x)[1]/mini_batch_size)
match_size_to_batch <- 128 * rounded_dim
rounded\u dim根据错误,您试图输入一个形状为(112512,40,43)
的对象,但您的LSTM层需要一个形状为(128,40,43)
的对象。似乎缺少一些代码,但在定义输入层时,是否正在修复批大小?我很幸运地将我的输入层定义为:
l_input = Input(shape = (None, num_features), name = 'input_layer')
我怀疑错误是由以下代码行引起的:
rounded_dim <- floor(dim(x)[1]/mini_batch_size)
match_size_to_batch <- 128 * rounded_dim
rounded_dim我已经实现了一个for循环,现在返回大小为128的批:
更改代码:
data_files_generator <- function(train_set) {
files <- train_set
next_file <- 0
function() {
# move to the next file (note the <<- assignment operator)
next_file <<- next_file + 1
# if we've exhausted all of the files then start again at the
# beginning of the list (keras generators need to yield
# data infinitely -- termination is controlled by the epochs
# and steps_per_epoch arguments to fit_generator())
if (next_file > length(files))
{next_file <<- 1}
# determine the file name
file <- files[[next_file]]
text <- read_lines(paste(data_dir, file, sep = "" )) %>%
str_to_lower() %>%
str_c(collapse = "\n") %>%
removeNumbers() %>%
tokenize_characters(strip_non_alphanum = FALSE, simplify = TRUE)
text <- text[text %in% chars]
dataset <- map(
seq(1, length(text) - maxlen - 1, by = 3),
~list(sentece = text[.x:(.x + maxlen - 1)], next_char = text[.x + maxlen])
)
dataset <- transpose(dataset)
# Vectorization
x <- array(0, dim = c(length(dataset$sentece), maxlen, length(chars)))
y <- array(0, dim = c(length(dataset$sentece), length(chars)))
for(i in 1:length(dataset$sentece)){
x[i,,] <- sapply(chars, function(x){
as.integer(x == dataset$sentece[[i]])
})
y[i,] <- as.integer(chars == dataset$next_char[[i]])
}
rounded_dim <- floor(dim(x)[1]/mini_batch_size)
match_size_to_batch <- 128 * rounded_dim
x <- x[1:match_size_to_batch, 1:maxlen, 1:length(chars)]
y <- y_val[1:match_size_to_batch, 1:length(chars)]
#Edit:
span_start <-1
for (iter in 1:rounded_dim){
i <- iter * 128
span_end <- iter * 128
x <- x[span_start:span_end, 1:maxlen, 1:length(chars)]
y <- y[span_start:span_end, 1:length(chars)]
span_start <- i
return(list(x, y))
}
}
}
data\u files\u generator我实现了一个for循环,它现在返回大小为128的批:
更改代码:
data_files_generator <- function(train_set) {
files <- train_set
next_file <- 0
function() {
# move to the next file (note the <<- assignment operator)
next_file <<- next_file + 1
# if we've exhausted all of the files then start again at the
# beginning of the list (keras generators need to yield
# data infinitely -- termination is controlled by the epochs
# and steps_per_epoch arguments to fit_generator())
if (next_file > length(files))
{next_file <<- 1}
# determine the file name
file <- files[[next_file]]
text <- read_lines(paste(data_dir, file, sep = "" )) %>%
str_to_lower() %>%
str_c(collapse = "\n") %>%
removeNumbers() %>%
tokenize_characters(strip_non_alphanum = FALSE, simplify = TRUE)
text <- text[text %in% chars]
dataset <- map(
seq(1, length(text) - maxlen - 1, by = 3),
~list(sentece = text[.x:(.x + maxlen - 1)], next_char = text[.x + maxlen])
)
dataset <- transpose(dataset)
# Vectorization
x <- array(0, dim = c(length(dataset$sentece), maxlen, length(chars)))
y <- array(0, dim = c(length(dataset$sentece), length(chars)))
for(i in 1:length(dataset$sentece)){
x[i,,] <- sapply(chars, function(x){
as.integer(x == dataset$sentece[[i]])
})
y[i,] <- as.integer(chars == dataset$next_char[[i]])
}
rounded_dim <- floor(dim(x)[1]/mini_batch_size)
match_size_to_batch <- 128 * rounded_dim
x <- x[1:match_size_to_batch, 1:maxlen, 1:length(chars)]
y <- y_val[1:match_size_to_batch, 1:length(chars)]
#Edit:
span_start <-1
for (iter in 1:rounded_dim){
i <- iter * 128
span_end <- iter * 128
x <- x[span_start:span_end, 1:maxlen, 1:length(chars)]
y <- y[span_start:span_end, 1:length(chars)]
span_start <- i
return(list(x, y))
}
}
}
data\u files\u生成器是正确的,这是我的问题。我想以这种方式更改代码,以获得大小为128的批。我想我已经设法做到了,但我不确定是否会返回整个文本或只返回最后一批。我将在问题中编辑我的代码。没错,这是我的问题。我想以这种方式更改代码,以获得大小为128的批。我想我已经设法做到了,但我不确定是否会返回整个文本或只返回最后一批。我将编辑问题中的代码。您是否仍然收到错误?我想您还需要span_start no error:)为什么我需要+1?我想没关系。Thx无论如何,您的范围将如下所示:1:128,然后128:256,然后256:384,依此类推,因此一些样本将在单独的批次中出现两次。这可能没什么大不了的,但这是需要注意的。你仍然会犯错误吗?我想您还需要span_start no error:)为什么我需要+1?我想没关系。Thx无论如何,您的范围将如下所示:1:128,然后128:256,然后256:384,依此类推,因此一些样本将在单独的批次中出现两次。这可能没什么大不了的,但这是需要注意的。