First, I want to say what a fantastic and valuable resource this is.
I was using your table of normalized vendor names and noticed some major ones are missing. For example, Pratt & Whitney and its variations. Also, Cogeco Cable, Enbridge, Air Canada, Bayer.
I can add a few ones that I found with a PR, but there might be many more missing.
I tried to replicate your code to normalize vendor names in Python (remove punctuation, make upper case and strip suffixes like INC and LTD) and ended up with 124,381 unique values, compared to 7,093 in the normalized table. I'm wondering if I did something wrong or what else can be done to have more complete dataset.
First, I want to say what a fantastic and valuable resource this is.
I was using your table of normalized vendor names and noticed some major ones are missing. For example, Pratt & Whitney and its variations. Also, Cogeco Cable, Enbridge, Air Canada, Bayer.
I can add a few ones that I found with a PR, but there might be many more missing.
I tried to replicate your code to normalize vendor names in Python (remove punctuation, make upper case and strip suffixes like INC and LTD) and ended up with 124,381 unique values, compared to 7,093 in the normalized table. I'm wondering if I did something wrong or what else can be done to have more complete dataset.