$ pio app new ecommerce01
# if ecommerce01 already exists,clean data first
$ pio app data-delete ecommerce01
$ pio eventserver
Install Python SDK
$ pip install predictionio
# or
$ easy_install predictionio
Replace the value of access_key parameter by your Access Key and run
$ python data/loan.py --file data/loan_sample.csv --access_key $ACCESS_KEY
Find your access_key
$ pio app list
You should see something like this:
[INFO] [App$] Name | ID | Access Key | Allowed Event(s)
[INFO] [App$] ecommerce01 | 2 | Jhe05JGGEFzWHZpDDIkCs1lKyujGMXg4GXBa7i0QPMh2oqmW8G0RcbszZkMNHbjG | (all)
[INFO] [App$] tapster | 1 | Dw5lyfoJxdEz2N1lQ9xskPDn5G40xaVNyVLUwtza9IbsYp0shAc2CMHWWqh5yb3a | (all)
[INFO] [App$] Finished listing 2 app(s).
$ pio build --verbose
$ pio train
$ pio deploy
-
Create the basic engine template with the purpose of:
- Provide generic recommendation capability on top of machine learning algorithms
- Generalize scenarios where users buy products and write reviews