Datasets:
Change parquet dataset
Browse files- .gitattributes +1 -0
- README.md +12 -3
- convert.py +1 -1
- train.parquet → handwriting/train-00000-of-00001.parquet +0 -0
.gitattributes
CHANGED
@@ -3607,3 +3607,4 @@ processed_images/r32-0862.png filter=lfs diff=lfs merge=lfs -text
|
|
3607 |
processed_images/r31-0919.png filter=lfs diff=lfs merge=lfs -text
|
3608 |
processed_images/r31-1876.png filter=lfs diff=lfs merge=lfs -text
|
3609 |
processed_images/r32-0784.png filter=lfs diff=lfs merge=lfs -text
|
|
|
|
3607 |
processed_images/r31-0919.png filter=lfs diff=lfs merge=lfs -text
|
3608 |
processed_images/r31-1876.png filter=lfs diff=lfs merge=lfs -text
|
3609 |
processed_images/r32-0784.png filter=lfs diff=lfs merge=lfs -text
|
3610 |
+
handwriting/train-00000-of-00001.parquet filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
@@ -12,6 +12,15 @@ pretty_name: Thai Handwriting Dataset
|
|
12 |
size_categories:
|
13 |
- 10K<n<100K
|
14 |
maintainer: Kobkrit Viriyayudhakorn ([email protected])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
---
|
16 |
|
17 |
# Thai Handwriting Dataset
|
@@ -26,9 +35,9 @@ The dataset consists of handwritten Thai text images along with their ground tru
|
|
26 |
|
27 |
The dataset is provided in a parquet file format with the following columns:
|
28 |
|
29 |
-
- `
|
30 |
-
- `text`: Ground truth transcription of the handwritten text
|
31 |
-
- `label_file`: Source label file name
|
32 |
|
33 |
## Original Dataset
|
34 |
|
|
|
12 |
size_categories:
|
13 |
- 10K<n<100K
|
14 |
maintainer: Kobkrit Viriyayudhakorn ([email protected])
|
15 |
+
dataset_info:
|
16 |
+
- config_name: handwriting
|
17 |
+
features:
|
18 |
+
- name: text
|
19 |
+
dtype: string
|
20 |
+
- name: image
|
21 |
+
dtype: image
|
22 |
+
- name: label_file
|
23 |
+
dtype: string
|
24 |
---
|
25 |
|
26 |
# Thai Handwriting Dataset
|
|
|
35 |
|
36 |
The dataset is provided in a parquet file format with the following columns:
|
37 |
|
38 |
+
- `image`: Image data (Image type)
|
39 |
+
- `text`: Ground truth transcription of the handwritten text (String)
|
40 |
+
- `label_file`: Source label file name (String)
|
41 |
|
42 |
## Original Dataset
|
43 |
|
convert.py
CHANGED
@@ -61,7 +61,7 @@ def create_dataset():
|
|
61 |
print("Converting to dataframe...")
|
62 |
df = pd.DataFrame(data)
|
63 |
print("Saving to parquet file...")
|
64 |
-
df.to_parquet("
|
65 |
print("Dataset creation complete!")
|
66 |
|
67 |
if __name__ == "__main__":
|
|
|
61 |
print("Converting to dataframe...")
|
62 |
df = pd.DataFrame(data)
|
63 |
print("Saving to parquet file...")
|
64 |
+
df.to_parquet("train.parquet", index=False)
|
65 |
print("Dataset creation complete!")
|
66 |
|
67 |
if __name__ == "__main__":
|
train.parquet → handwriting/train-00000-of-00001.parquet
RENAMED
File without changes
|