Hello,
I'm having big troubles running your project on my machine. The datasets as described in the paper aren't available for download. Instead, we get the unprocessed OpenData tables for Singapore, USA, Canada and UK (which after processing should be OpenData Large). Since you extract 64,698 tables but don't specify which ones exactly, I cannot reproduce the dataset myself.
WebTable Large also isn't available. You process 16,670,064 tables but don't provide the processed dataset.
The same issue exists for the non-large versions of WebTable and OpenData. You sample 17 % from WebTable Large and 10 % from OpenData Large, but don't provide the dataset for download.
Could you provide the datasets as shown in the paper? Your Python scripts as such aren't runnable either
Best,
Denis
Hello,
I'm having big troubles running your project on my machine. The datasets as described in the paper aren't available for download. Instead, we get the unprocessed OpenData tables for Singapore, USA, Canada and UK (which after processing should be OpenData Large). Since you extract 64,698 tables but don't specify which ones exactly, I cannot reproduce the dataset myself.
WebTable Large also isn't available. You process 16,670,064 tables but don't provide the processed dataset.
The same issue exists for the non-large versions of WebTable and OpenData. You sample 17 % from WebTable Large and 10 % from OpenData Large, but don't provide the dataset for download.
Could you provide the datasets as shown in the paper? Your Python scripts as such aren't runnable either
Best,
Denis