Question

在预测时，我得到的矢量是val_steps长度的两倍......预测被复制：1到val_steps的元素与从valsteps + 1到结尾的元素相同。我使用以下参数：

lookback <- 1 #look at X days past 
step <- 1 # sample every X days
delay <- 1 # Forecast horizon
batch_size <- 1
myfc <- predict_generator(model, val_gengp,val_steps, verbose=T)

你知道为什么预测会重复吗？可能我不太了解我正在使用的发电机：

generatorgp <- function(data, lookback, delay, min_index, max_index,
                      shuffle = FALSE, batch_size = batch_size, step = step) { #6
  if (is.null(max_index)) max_index <- nrow(data) - delay - 1
  i <- min_index + lookback
  function() {
    if (shuffle) {
      rows <- sample(c((min_index+lookback):max_index), size = batch_size)
    } else {
      if (i + batch_size >= max_index)
        i <<- min_index + lookback  # i <<- 1 + lookback
      rows <- c(i:min(i+batch_size, max_index))  # rows <- c(i:min(i+batch_size, maxtrain))
      i <<- i + length(rows)
    }
    samples <- array(0, dim = c(length(rows),
                                lookback / step,
                                dim(data)[[-1]])) #View(samples)
    targets <- array(0, dim = c(length(rows))) #targets
    for (j in 1:length(rows)) {
      indices <- seq(rows[[j]] - lookback, rows[[j]],
                     length.out = dim(samples)[[2]])
      samples[j,,] <- data[indices,]
      targets[[j]] <- data[rows[[j]] + delay,myy] #
    }
    list(samples)#, targets)
    #samples
  }
}
val_gengp = generatorgp(
  data,
  lookback = lookback,
  delay = delay,
  min_index = minval,
  max_index = maxval,
  step = step,
  batch_size = batch_size
)

还想知道实际预测的第一个元素的索引是什么？第一个目标prediction be: data[(min_validation_set+lookback+1)] ？非常感谢

使用R Keras的重复预测predict_generator（）

0 个答案: