Python + .parquet + csv
Budget: ₹600 – ₹1,500 INR
I have some folders containing .parquet files.
These .parquet files contain important columns: 'frame', 'type' and 'x', 'y'
Where x and y are normalized spatial coordinates.
I have a csv file that contains important columns: 'path', 'sign'.
Here 'path' is the .parquet file name and 'sign' is the target class.
The task is using the given .parquet and csv file, prepare the dataset in the following YOLO format-
images
labels
Here images folder will have images
And labels folder will have .txt files related to images.
The .txt file names should be the same as the image file name.
And it's format is like the below-
<object-class> <x_center> <y_center> <width> <height>
Where object-class is an integer object number from 0 to (classes-1). Here classes are the 'sign' column in csv file.
The Python script must be fast and scalable since I have 21 folders and each folder has 4000+ .parquet files.
Ping me if you can do this or in case of any queries.
These .parquet files contain important columns: 'frame', 'type' and 'x', 'y'
Where x and y are normalized spatial coordinates.
I have a csv file that contains important columns: 'path', 'sign'.
Here 'path' is the .parquet file name and 'sign' is the target class.
The task is using the given .parquet and csv file, prepare the dataset in the following YOLO format-
images
labels
Here images folder will have images
And labels folder will have .txt files related to images.
The .txt file names should be the same as the image file name.
And it's format is like the below-
<object-class> <x_center> <y_center> <width> <height>
Where object-class is an integer object number from 0 to (classes-1). Here classes are the 'sign' column in csv file.
The Python script must be fast and scalable since I have 21 folders and each folder has 4000+ .parquet files.
Ping me if you can do this or in case of any queries.