/v1 does not work out of the box about ml-powered-applications HOT 1 CLOSED

Hi @kedarmhaswade

I've updated the repository with a few changes that should make this easier.

Unfortunately, in order for the nltk tokenizer to work properly, you need to install a few nltk packages. The only way to install these packages that I know of requires a process that is slightly more involved than pip install.

In order to fix the particular error you are encountering, you can take the following steps:

Start by activating the venv. Then, in the same console, launch a python session (just typing python should work)
Then, run the following two lines:
import nltk
nltk.download('punkt')

After you do this, the error should be fixed. You may get errors for other required nltk packages, and should be able to fix those in the same manner. I'll update the docs to clarify this process.

from ml-powered-applications.