Hi All, I need your help to optimise a method that builds semantic vectors from text files. The code was originally written in Java, and I have converted it to Python. Given my poor knowledge of Python, no surprise that the Python code is extremely slower than the original Java code.
Here is my question, any volunteer to go through the code and improve its performance by tweaking my loops and dictionary lookups (at the first instance), and then helping me to implement Jenkins Hashing?
I hope to convince everybody in the team that the solution this code offers is one of the best choices for building vectors of any type for PeARS.
Looking forward to hearing from you.
PS. I will email the code