The error 'TypeError: zipline.pipeline.pipeline.validate_column() expected a value of type zipline.pipeline.term.Term for argument 'term' but got pandas.core.series.Series instead. is saying that a pipeline column is being defined using a pandas series and not with what is expected which is either a factor, filter, or classifier. Pipeline columns can only be defined as one of those three types. The offending line is this

# Apple_df isn't explicitly defned but python interprets it as a series  
Apple_df['daygain'] = Apple_df['close_price'] - Apple_df['open_price']  
daygain_factor = Apple_df['daygain']  

# this series is then assigned to a column which pipeline doesn't like  
pipe = Pipeline(columns={'daygain_factor':daygain_factor,  

So how to create a "days gain" which is a factor (and therefor pipeline will accept as a column)? One could do it a couple of ways. 1) use built in operators and methods or 2) create a custom factor. Here's how one could do it using some built in operators

from quantopian.pipeline.data.builtin import USEquityPricing


# create factors directly from datasets using the latest method  
open_price = USEquityPricing.open.latest  
close_price = USEquityPricing.close.latest  

# create a 'gain' factor by combining existing factors using built-in operators  
# could have also done (close_price / open_price)-1  
 day_gain = (close_price - open_price) / open_price

Here's a custom factor which does the same thing

# create a custom factor to calc the days gain  
class DayGain(CustomFactor):  
        '''  
        Returns the daily open to close gain  
        '''  
        inputs = [USEquityPricing.open, USEquityPricing.close]  
        window_length = 1  
        def compute(self, today, assets, out, open_price, close_price):  
            out[:] =  (close_price / open_price) - 1.0  


day_gain_custom_factor = DayGain()

Attached is a notebook showing both of these approaches in action and how they return the same value. Hope that helps.

Good luck!