[Thanks] build a repo based on DeepQA

[hailiang-wang]

After using DeepQA for a while, I looked into the codebase and it is a bit complicated, because it provides many features.

To understand the training logic, I almost go through all codes, the works are mainly done in chatbot.py. So, I think if we have a plain code for training and provide some other codes for serving is better to understand the workflow. So, I started from DeepQA project and commit to DeepQA2.

The model and textdata are mostly the same, the changes are about -

• Simplify the configuration with a config.ini
• Enable run captured model with django restapi framework
• Pre-process data, train model and serving model are split into three entries.

Well, the magic works have been done in DeepQA Project. Great thanks.

[hunkim]

Thanks for the information. I was wondering if DeepQA2 is just refactoring or do you add new algorithms to enhance the performance of DeepQA.

[Conchylicultor]

Glad you succeeded. I had a quick look at the code and it seems you didn't use C++ as you originally planed. Was it because of technical limitation of the C++ API or just because keeping the python code was simpler ?