How to handle imbalanced classes ?

[smi45]

Hi, I have a use case wherein there are two classes and are hugely imbalanced. How can i fix this issue.

I have used the source code.

Sentiment Analysis with BERT

[ashfarhangi]

You can use my code and replace the cat1 and cat2 for your imbalance classes
from sklearn.utils import resample
cat_1= df[df.school_handle == 'ourmedschool']
cat_2= df[df.school_handle == 'ucfcah']

cat_large = df[df.school_handle == 'UCFCECS']
cat_1 = resample(cat_1,
replace=True,
n_samples=len(cat_large),
random_state=RANDOM_SEED)
cat_2 = resample(cat_2,
replace=True,
n_samples=len(cat_large),
random_state=RANDOM_SEED)

df1 = pd.merge(cat_1,cat_large,how='outer')
df1 = pd.merge(df1,cat_2,how='outer')
df1
sns.countplot(df1.school_handle)
plt.xlabel('College')