'h2o flow versus python shell
I have build a model in h2o flow. Once completed I save the file to my local machine. Next, I import some new data via Flow and python shell, I then load the model and run predict on the dataframe.
I should get the exact same results as the dataframe and model are both the same. Just different environments. However, I get a different predictions in the different environments. Anybody have a similar experience?
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|