Kyryl Truskovskyi
Kyryl Truskovskyi
Hi @GeniusLight Thanks for the question! Yes, there might be a huge difference, it depends on - model I use mobile-net-v2 (https://arxiv.org/abs/1801.04381) in https://github.com/titu1994/neural-image-assessment implementation author use mobile-net-v1 (https://arxiv.org/abs/1704.04861) and...
@GeniusLight about comparing results with the paper (https://arxiv.org/pdf/1709.05424.pdf) Here the table with author results I got on test dataset EMD = 0.080 But we have different train/validation/test split
@GeniusLight Now I understand what you mean. So I need some time to figure out and fix this problem. As soon I will find the problem I will posts results
thanks for getting back to me @phughk! let me know if I can help with reproducing or fixing it, have full code to perform test end2end
Hey team, I ran some additional tests to determine how many vectors I can add & build index on top. Every attempt to build an index with more than approximately...