1 files changed, 20 insertions, 8 deletions

diff --git a/data/csv.py b/data/csv.py
index d9f9db8..58c833e 100644
--- a/data/csv.py
+++ b/data/csv.py
@@ -1,6 +1,7 @@
 import math
 import torch
 import json
+import copy
 from pathlib import Path
 from typing import NamedTuple, Optional, Union, Callable
 
@@ -296,8 +297,25 @@ class VlpnDataset(IterableDataset):
 
         self.bucket_item_range = torch.arange(len(self.bucket_items))
 
+        self.cache = {}
         self.length_ = (self.bucket_assignments.bincount() / self.batch_size).ceil().long().sum().item()
 
+    def get_static_example(self, item: VlpnDataItem, item_index: int):
+        if item_index in self.cache:
+            return copy.copy(self.cache[item_index])
+
+        example = {}
+
+        example["prompt_ids"] = self.prompt_processor.get_input_ids(keywords_to_prompt(item.prompt))
+        example["nprompt_ids"] = self.prompt_processor.get_input_ids(item.nprompt)
+
+        if self.num_class_images != 0:
+            example["class_prompt_ids"] = self.prompt_processor.get_input_ids(item.cprompt)
+
+        self.cache[item_index] = example
+
+        return example
+
     def __len__(self):
         return self.length_
 
@@ -356,19 +374,13 @@ class VlpnDataset(IterableDataset):
                 item = self.items[item_index]
                 mask[self.bucket_item_range[bucket_mask][0]] = False
 
-                example = {}
-
-                example["prompt_ids"] = self.prompt_processor.get_input_ids(keywords_to_prompt(item.prompt))
-                example["nprompt_ids"] = self.prompt_processor.get_input_ids(item.nprompt)
-
-                example["instance_images"] = image_transforms(get_image(item.instance_image_path))
+                example = self.get_static_example(item, item_index)
                 example["instance_prompt_ids"] = self.prompt_processor.get_input_ids(
                     keywords_to_prompt(item.prompt, self.dropout, True)
                 )
-
+                example["instance_images"] = image_transforms(get_image(item.instance_image_path))
                 if self.num_class_images != 0:
                     example["class_images"] = image_transforms(get_image(item.class_image_path))
-                    example["class_prompt_ids"] = self.prompt_processor.get_input_ids(item.cprompt)
 
                 batch.append(example)