C3 W1 A1 Ex 10: probability of spam given text

Hi everyone,

I am working on the last problem and I am running into some errors with my output. I’m getting fairly close where it is always one output that is off. I suspect it is the last line that is off. This is what I have:
prior_spam = class_freq[“spam”]/(class_freq[“spam”]+class_freq[“ham”])
prior_ham = class_freq[“ham”]/(class_freq[“spam”]+class_freq[“ham”])
prob_spam = (likelihood_word_given_spamprior_spam)/(likelihood_word_given_hamprior_ham+likelihood_word_given_spam*prior_spam)

Can someone tell me if this is right? I’m going off of the bayes’ theorem formula.
Screen Shot 2023-07-06 at 7.15.09 PM|690x204

Can someone respond to this?

I’m also stuck on this one.

Hi Stella, I personally did not need to create the prior_spam and prior_ham objects in order to pass this exercise. Perhaps a simpler approach where the class_freq variable is mentioned only once per class is better suited.

Would you mind elaborating further on what you need help with?

Hi Tom, thank you for getting back to me! I am not sure what you mean by mentioning class_freq once per class. Could you give me an example?

Hi Stella, what I meant was that I only use these variables class_freq['spam'] and class_freq['ham'] once. I hope it’ll help to start from blank for this exercise.

I see! I think I figured it out. Thank you!

1 Like