Datasets:
Tasks:
Image Classification
Sub-tasks:
multi-class-image-classification
Languages:
English
Size:
100K<n<1M
ArXiv:
License:
iterate trought image names and store it in a list for each dataset
Browse files- NIH-Chest-X-ray-dataset.py +34 -3
NIH-Chest-X-ray-dataset.py
CHANGED
@@ -91,11 +91,42 @@ class XChest(datasets.GeneratorBasedBuilder):
|
|
91 |
|
92 |
|
93 |
def _split_generators(self, dl_manager):
|
94 |
-
|
95 |
train_val_list = get(_URLS['train_val_list']).iter_lines()
|
96 |
train_val_list = set([x.decode('UTF8') for x in train_val_list])
|
97 |
-
print(train_val_list)
|
98 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
99 |
|
100 |
def _generate_examples(self, files):
|
101 |
pass
|
|
|
91 |
|
92 |
|
93 |
def _split_generators(self, dl_manager):
|
94 |
+
# Get the image names that belong to the train-val dataset
|
95 |
train_val_list = get(_URLS['train_val_list']).iter_lines()
|
96 |
train_val_list = set([x.decode('UTF8') for x in train_val_list])
|
97 |
+
print(train_val_list))
|
98 |
+
|
99 |
+
# Create list for store the name of the images for each dataset
|
100 |
+
train_files = []
|
101 |
+
test_files = []
|
102 |
+
|
103 |
+
# Download batches
|
104 |
+
data_files = dl_manager.download_and_extract(_URLS['image_urls'])
|
105 |
+
|
106 |
+
# Iterate trought image folder and check if they belong to
|
107 |
+
# the trainset or testset
|
108 |
+
for batch in data_files:
|
109 |
+
for img in batch:
|
110 |
+
if img in train_val_list:
|
111 |
+
train_files.append(img)
|
112 |
+
else:
|
113 |
+
test_files.append(img)
|
114 |
+
|
115 |
+
print(train_files)
|
116 |
+
print(test_files)
|
117 |
+
|
118 |
+
return [
|
119 |
+
datatsets.SplitGenerator(
|
120 |
+
name=datasets.Split.TRAIN,
|
121 |
+
gen_kwargs={
|
122 |
+
'files': dl_manager.iter_files()
|
123 |
+
}
|
124 |
+
|
125 |
+
),
|
126 |
+
datasets.SplitGenerator(
|
127 |
+
|
128 |
+
)
|
129 |
+
]
|
130 |
|
131 |
def _generate_examples(self, files):
|
132 |
pass
|