SaulLu commited on
Commit
4e7b9d3
·
1 Parent(s): 7d1652c

remove duplicates

Browse files
Files changed (1) hide show
  1. TextCaps.py +5 -11
TextCaps.py CHANGED
@@ -105,11 +105,8 @@ class TextCapsDataset(datasets.GeneratorBasedBuilder):
105
  "set_name": datasets.Value("string"),
106
  "image_name": datasets.Value("string"),
107
  "image_path": datasets.Value("string"),
108
- "caption_id": datasets.Value("int32"),
109
- "caption_str": datasets.Value("string"),
110
- "caption_tokens": [datasets.Value("string")],
111
  "reference_strs": [datasets.Value("string")],
112
- "reference_tokens": [datasets.Value("string")],
113
  }
114
  )
115
  return datasets.DatasetInfo(
@@ -157,7 +154,7 @@ class TextCapsDataset(datasets.GeneratorBasedBuilder):
157
  def _generate_examples(
158
  self, captions_path, ocr_tokens_path, images_path, split_name
159
  ):
160
-
161
  captions = json.load(open(captions_path, "r"))["data"]
162
  ocr_tokens = json.load(open(ocr_tokens_path, "r"))["data"]
163
 
@@ -166,6 +163,9 @@ class TextCapsDataset(datasets.GeneratorBasedBuilder):
166
  ocr_tokens_per_image_id[ocr_item["image_id"]] = ocr_item
167
 
168
  for caption_item in captions:
 
 
 
169
  ocr_item = ocr_tokens_per_image_id[caption_item["image_id"]]
170
  record = {
171
  "ocr_tokens": ocr_item["ocr_tokens"],
@@ -182,15 +182,9 @@ class TextCapsDataset(datasets.GeneratorBasedBuilder):
182
  "image" : str(images_path / f'{caption_item["image_name"]}.jpg')
183
  }
184
  if not split_name == "test":
185
- record["caption_id"] = caption_item["caption_id"]
186
- record["caption_str"] = caption_item["caption_str"]
187
- record["caption_tokens"] = caption_item["caption_tokens"]
188
  record["reference_strs"] = caption_item["reference_strs"]
189
  record["reference_tokens"] = caption_item["reference_tokens"]
190
  else:
191
- record["caption_id"] = None
192
- record["caption_str"] = None
193
- record["caption_tokens"] = None
194
  record["reference_strs"] = None
195
  record["reference_tokens"] = None
196
  yield caption_item["image_id"], record
 
105
  "set_name": datasets.Value("string"),
106
  "image_name": datasets.Value("string"),
107
  "image_path": datasets.Value("string"),
 
 
 
108
  "reference_strs": [datasets.Value("string")],
109
+ "reference_tokens": [[datasets.Value("string")]],
110
  }
111
  )
112
  return datasets.DatasetInfo(
 
154
  def _generate_examples(
155
  self, captions_path, ocr_tokens_path, images_path, split_name
156
  ):
157
+ seen_image_ids = set()
158
  captions = json.load(open(captions_path, "r"))["data"]
159
  ocr_tokens = json.load(open(ocr_tokens_path, "r"))["data"]
160
 
 
163
  ocr_tokens_per_image_id[ocr_item["image_id"]] = ocr_item
164
 
165
  for caption_item in captions:
166
+ if caption_item["image_id"] in seen_image_ids:
167
+ continue
168
+ seen_image_ids.add(caption_item["image_id"])
169
  ocr_item = ocr_tokens_per_image_id[caption_item["image_id"]]
170
  record = {
171
  "ocr_tokens": ocr_item["ocr_tokens"],
 
182
  "image" : str(images_path / f'{caption_item["image_name"]}.jpg')
183
  }
184
  if not split_name == "test":
 
 
 
185
  record["reference_strs"] = caption_item["reference_strs"]
186
  record["reference_tokens"] = caption_item["reference_tokens"]
187
  else:
 
 
 
188
  record["reference_strs"] = None
189
  record["reference_tokens"] = None
190
  yield caption_item["image_id"], record