Datasets:
Tasks:
Image Classification
Modalities:
Image
Formats:
imagefolder
Languages:
English
Size:
1K - 10K
Tags:
art
License:
Commit
·
dcd93ac
1
Parent(s):
9ec3910
update
Browse files- dataset_ini.py +49 -0
- test_labels.csv +1 -0
- train_labels.csv +1 -0
dataset_ini.py
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import pandas as pd
|
2 |
+
from datasets import Dataset, DatasetDict, concatenate_datasets
|
3 |
+
import os
|
4 |
+
|
5 |
+
def load_data(csv_file, images_folder):
|
6 |
+
df = pd.read_csv(csv_file)
|
7 |
+
# Update the image_path column to include the full path
|
8 |
+
df['image_path'] = df['image_path'].apply(lambda x: os.path.join(images_folder, x))
|
9 |
+
dataset = Dataset.from_pandas(df)
|
10 |
+
return dataset
|
11 |
+
|
12 |
+
def load_dataset():
|
13 |
+
# Define directories
|
14 |
+
test_real = 'datasets/test_set/real/'
|
15 |
+
test_fake = 'datasets/test_set/fake/'
|
16 |
+
|
17 |
+
training_real = 'datasets/training_set/real/'
|
18 |
+
training_fake = 'datasets/training_set/fake/'
|
19 |
+
|
20 |
+
# Define CSV files
|
21 |
+
train_csv = 'train_labels.csv'
|
22 |
+
test_csv = 'test_labels.csv'
|
23 |
+
|
24 |
+
# Load datasets
|
25 |
+
train_dataset_real = load_data(train_csv, training_real)
|
26 |
+
|
27 |
+
train_dataset_fake = load_data(train_csv, training_fake)
|
28 |
+
|
29 |
+
# Combine datasets for training
|
30 |
+
train_dataset = concatenate_datasets([train_dataset_real, train_dataset_fake])
|
31 |
+
|
32 |
+
test_dataset_real = load_data(test_csv, test_real)
|
33 |
+
test_dataset_fake = load_data(test_csv, test_fake)
|
34 |
+
|
35 |
+
# Combine datasets for testing
|
36 |
+
test_dataset = concatenate_datasets([test_dataset_real, test_dataset_fake])
|
37 |
+
|
38 |
+
# Create DatasetDict
|
39 |
+
dataset_dict = DatasetDict({
|
40 |
+
'train': train_dataset,
|
41 |
+
'test': test_dataset
|
42 |
+
})
|
43 |
+
|
44 |
+
return dataset_dict
|
45 |
+
|
46 |
+
if __name__ == "__main__":
|
47 |
+
dataset = load_dataset()
|
48 |
+
print(dataset)
|
49 |
+
|
test_labels.csv
CHANGED
@@ -1,3 +1,4 @@
|
|
|
|
1 |
datasets/test_set/real/Gustave_Courbet_11.jpg,0
|
2 |
datasets/test_set/real/Gustave_Courbet_12.jpg,0
|
3 |
datasets/test_set/real/Gustave_Courbet_13.jpg,0
|
|
|
1 |
+
image_path,label
|
2 |
datasets/test_set/real/Gustave_Courbet_11.jpg,0
|
3 |
datasets/test_set/real/Gustave_Courbet_12.jpg,0
|
4 |
datasets/test_set/real/Gustave_Courbet_13.jpg,0
|
train_labels.csv
CHANGED
@@ -1,3 +1,4 @@
|
|
|
|
1 |
datasets/training_set/real/Alfred_Sisley_1.jpg,0
|
2 |
datasets/training_set/real/Alfred_Sisley_10.jpg,0
|
3 |
datasets/training_set/real/Alfred_Sisley_100.jpg,0
|
|
|
1 |
+
image_path,label
|
2 |
datasets/training_set/real/Alfred_Sisley_1.jpg,0
|
3 |
datasets/training_set/real/Alfred_Sisley_10.jpg,0
|
4 |
datasets/training_set/real/Alfred_Sisley_100.jpg,0
|