Create a python dataframe using string of variable

Question

I have a function for running a logistic regression model. I would like to permanently save the dataframe generated within (cf) and amend it's name.

def model(ind, dep):
    global cf
    ind.fillna(0)
    #some modelling code

    #create confuson matix

    cf = pd.DataFrame(confusion_matrix(y_train, y_pred))
    cf.index = models
    cf.columns = models
    print(cf)
    cf.plot.barh()


    "cf_" + str(ind) = cf

    return "cf_" + str(ind)

model(X_tv, y_combo)
model(X_tv_chan, y_combo)

I get this error

File "", line 64 "cf_" + str(ind) = coeff ^ SyntaxError: can't assign to operator

is there something wrong with how I am trying to create the dataframe?

"cf_" + str(ind) = cf

jpp · Accepted Answer · 2018-11-09 11:13:59Z

1

Yes. This line doesn't create a new variable:

"cf_" + str(ind) = cf

Strings are immutable. You can't "assign a dataframe to a string" either, I'm not even sure what this is attempting to achieve. It's also good practice to avoid global variables.

Just return your dataframe and assign to an explicit variable name. If you are set on changing your variable name, use a dictionary and dict.pop:

def model(ind, dep):
    # ...
    cf = pd.DataFrame(confusion_matrix(y_train, y_pred))
    cf.index = models
    cf.columns = models
    # ...
    return cf

dfs = {}

dfs['ind'] = model(X_tv, y_combo)  # identify dataframe with key 'ind'
dfs['cf_ind'] = dfs.pop('ind')     # rename identifier to 'cf_ind'

answered Nov 9, 2018 at 11:13

jpp

166k37 gold badges301 silver badges362 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

James Adams Over a year ago

Thanks. I have a number of dataframes I want to pull out of the function and save. In the end I have used the global options and then returned then. Then in between each call of the function I have appended the new table.model (X_tv, y_combo, 'just_trans') masterCM=masterCM.append(cf) masterCOEFF=masterCOEFF.append(coeff) model(X_tv_cattot, y_combo, 'trans_cattotals') masterCM=masterCM.append(cf) masterCOEFF=masterCOEFF.append(coeff)

jpp Over a year ago

I want to pull out of the function and save. That's what this does: dfs['ind'] = model(X_tv, y_combo). Please don't post code in comments, edit your question to clarify your question.

Collectives™ on Stack Overflow

Create a python dataframe using string of variable

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related