R-Python Reticulate : TypeError：'size‘必须是整数

Question

对于Reticulate来说，我正在尝试运行这个示例代码，但没有成功:(

library(reticulate)
use_virtualenv("~/virtualenv", required = TRUE)
np <- import("numpy", convert = FALSE)
plt <- import("matplotlib.pyplot")
img <- import("pyts")

# R variables
image_data <- runif(25)
reshape_image_data <- array_reshape(image_data, c(25,1))

# Reticulated Python variables
x_array <- np$array(reshape_image_data)

mtf <- img$image$MTF(c(5, 5))
#mtf <- img$image$MTF(as.integer(c(as.integer(5),as.integer(5))))

X_mtf <- mtf$fit_transform(x_array)

获取此错误：

Error in py_call_impl(callable, dots$args, dots$keywords) : 
  TypeError: 'size' must be an integer.

Detailed traceback: 
  File "~/virtualenv/lib/python3.6/site-packages/sklearn/base.py", line 517, in fit_transform
    return self.fit(X, **fit_params).transform(X)
  File "~/virtualenv/lib/python3.6/site-packages/pyts/image/image.py", line 268, in transform
    raise TypeError("'size' must be an integer.")

这来自pyts.image.image.py，其中的校验码如下所示：

if not isinstance(self.image_size, int):
            raise TypeError("'size' must be an integer.")

没有让它起作用。有人试过了吗？谢谢

Answer 1

终于明白了。注意array_reshape c(1,25)和MTF as.integer(5)：

library(reticulate)
use_virtualenv("~/virtualenv", required = TRUE)
np <- import("numpy", convert = FALSE)
plt <- import("matplotlib.pyplot")
img <- import("pyts")

# R variables
image_data <- runif(25)
reshape_image_data <- array_reshape(image_data, c(1,25))

# Reticulated Python variables
x_array <- np$array(reshape_image_data)

mtf <- img$image$MTF(as.integer(5))

X_mtf <- mtf$fit_transform(x_array)