NameError: name 'rmse' is not defined

fastai
#1

I’ve imported all libraries as shown in many examples:
from fastai.vision import *
from fastai.tabular import *
from fastai import *

When I try to execute
learn = tabular_learner(data, layers=[200,50], ps=[0.001,0.01], emb_drop=0.04,
y_range=y_range, metrics=rmse)

I get:
NameError: name ‘rmse’ is not defined

I have fastai version 1.0.34

What is the problem here? how can I specify the metric to be root mean square error?

1 Like
#2

It’s probably because you have an old version, you should upgrade your library.

#4

I’m having this problem too and I’m using the latest version of fastai (2.7.7). Most of the predefined metrics do not work for me. E.g., if I use mean_squared_error as

learn = tabular_learner(dls, y_range=(1,100), 
    metrics=mean_squared_error)

I get:

NameError: name ‘mean_squared_error’ is not defined

The same thing happens if I use mean_absolute_error or most of the other predefined metrics. But if I use exp_rmspe. It works fine.
Am I doing something wrong?

#5

There is no metric in fastai called mean_squared_error.

#6

Thank you for your response. I’m confused then, because I see it on this metrics page.

#7

That’s the fastai1 docs. The current version (v2) is at docs.fast.ai

1 Like