Could you please share the snippet of code where this problem arrise? My guess is that you have written array.zscore when it should be ..zscore(array) but I might be wrong.

Below is the code, thanks for the help!

class eps_growth(CustomFactor):  
    fq1_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 1) #upcoming quarter  
    fq2_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 2) #2 quarters into future  
    fq1_eps_mean = fq1_eps_cons.mean.latest  
    fq2_eps_mean = fq2_eps_cons.mean.latest  
    inputs = [fq1_eps_mean, fq2_eps_mean]  
    window_length = 1  
    window_safe=True  
    def compute(self, today, assets, out, fq1_eps_mean, fq2_eps_mean):  
        eps_estimated_growth_factor = (fq2_eps_mean - fq1_eps_mean) / fq1_eps_mean  
        x = eps_estimated_growth_factor  
        out[:] = x.zscore()  

Ahh well, that seems to be the problem!

lol sorry, what is the fix? I am dumb and still can't get it to work

Ahh excuse me, I suppose that slices Fq#_eps_cons are numpy arrays and therefore so is x. Numpy arrays does not have a function named zscore but a function that takes in a numpy array.

Try

out[:] = scipy.stats.zscore(x)  

I guess this is the zscore function you seek.

Thanks! I think this is a step closer to correct but I just received a different error - " NoSuchSymbols: Failed to find securities matching [] "

My code looks like this now

class eps_growth(CustomFactor):  
    fq1_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 1) #upcoming quarter  
    fq2_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 2) #2 quarters into future  
    fq1_eps_mean = fq1_eps_cons.mean.latest  
    fq2_eps_mean = fq2_eps_cons.mean.latest  
    inputs = [fq1_eps_mean, fq2_eps_mean]  
    window_length = 1  
    window_safe=True  
    def compute(self, today, assets, out, fq1_eps_mean, fq2_eps_mean):  
        eps_estimated_growth_factor = (fq2_eps_mean - fq1_eps_mean) / fq1_eps_mean  
        x = eps_estimated_growth_factor  
        y = scipy.stats.zscore(x)  
        out[:] = y  

Any help would be greatly appreciated!

There are always many ways to code. A lot is personal preference. I like sticking to the built in factors and methods myself. So, rather than compute zscore inside the custom factor, maybe just output the 'growth factor' and then use the built in zscore factor (see the docs https://www.quantopian.com/docs/api-reference/pipeline-api-reference#zipline.pipeline.Factor.zscore). Something like this

class EPS_Growth(CustomFactor):  
    fq1_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 1) #upcoming quarter  
    fq2_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 2) #2 quarters into future  
    fq1_eps_mean = fq1_eps_cons.mean.latest  
    fq2_eps_mean = fq2_eps_cons.mean.latest  
    inputs = [fq1_eps_mean, fq2_eps_mean]  
    window_length = 1  
    window_safe=True  

    def compute(self, today, assets, out, fq1_eps_mean, fq2_eps_mean):  
        eps_estimated_growth_factor = (fq2_eps_mean - fq1_eps_mean) / fq1_eps_mean  
        out[:] = eps_estimated_growth_factor 

eps_growth = EPS_Growth()  
# Good practice to mask for nan and infinite values  
# These values do not play well with zscore  
eps_growth_zscore = eps_growth.zscore(mask=eps_growth.isfinite())  

However, you don't even really need a custom factor. Just use the built in factor operations. Something like this

fq1_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 1) #upcoming quarter  
fq2_eps_cons = PeriodicConsensus.slice('EPS', 'qf', 2) #2 quarters into future  
eps_growth = (fq2_eps_cons.mean.latest - fq1_eps_cons.mean.latest) / fq1_eps_cons.mean.latest  
# Good practice to mask for nan and infinite values  
# These values do not play well with zscore  
eps_growth_z_score = eps_growth.zscore(mask=eps_growth.isfinite())

See the attached notebook. Both approaches are included showing they both return the same values.

@ Dan, Thank you for this awesome response!