in manual file 100 hould be ok, still need to modify 10

data-raw/cifar.R

@@ -92,40 +92,43 @@ constructor_cifar100 = function(path) {
   # )
   tv_ds = torchvision::cifar100_dataset(root = path, download = TRUE)
   class_names = readLines(file.path(path, "cifar-100-binary", "fine_label_names.txt"))
-
-  # TODO: use $.getitem() instead of $x
-  # data.table(
-  #   class = factor(as.integer(tv_ds$y), labels = class_names),
-  #   image = self$x,
-  #   split = factor(rep(c("train", "test"), c(50000, 10000))),
-  #   ..row_id = seq_len(60000)
-  # )
+
+  tv_data = tv_ds$.getitem(1:60000)
+
+  tv_data$class_names = class_names
+
+  tv_data
 }
 
 data = constructor_cifar100(path)
 
 cifar100_ds_generator = torch::dataset(
-  initialize = function() {
-    self$.data = data
+  initialize = function(data) {
+    self$.img_arr = data$x
   },
   .getitem = function(idx) {
     force(idx)
 
-    x = torch_tensor(read_cifar_image(self$.data$file[idx], self$.data$idx_in_file[idx], type = 100))
+    x = torch_tensor(self$.img_arr[i, , , ])
 
     return(list(x = x))
   },
   .length = function() {
-    nrow(self$.data)
+    dim(self$.img_arr)[1]
   }
 )
 
-cifar100_ds = cifar100_ds_generator()
+cifar100_ds = cifar100_ds_generator(data)
 
 dd = as_data_descriptor(cifar100_ds, list(x = c(NA, 32, 32, 3)))
 lt = lazy_tensor(dd)
 
-dt = cbind(data, data.table(image = lt))
+dt = data.table(
+  class = factor(data$y, labels = data$class_names),
+  image = lt,
+  split = factor(rep(c("train", "test"), c(50000, 10000))),
+  ..row_id = seq_len(60000)
+)
 
 task = as_task_classif(dt, target = "class")