Error using PassiveAgressiveClassifier on Text Data

Hello, I’m trying to use the PassiveAgressive classifier on my confidential text dataset. I’m getting the following error on Google Colab Environment=> TypeError: unsupported operand type(s) for -: ‘str’ and ‘int’

This is the code I’m using, X is the text dataframe and y is a Series describing the labels (i.e., category)

Python 3.9.15: Windows:

Steps/code to reproduce

model = linear_model.PAClassifier(C=0.1, mode=1)
pipe_pac = Pipeline(('vectorizer',BagOfWords(lowercase=True,ngram_range=(1, 3))),('pac',model))
X= df['Text']
y= df['CATEGORY']

X_train, X_test, y_train, y_test = train_test_split(X, y,test_size=0.3, random_state = 0)
newdf = pd.concat([X_train, pd.DataFrame(y_train)], axis=1)

data = list(zip(*map(newdf.get, newdf)))

for text,label in data:
    pipe_pac = pipe_pac.learn_one(text,label)

[smastelini]

How about using something like this?

[MaxHalford]

I was just wondering if there is a way to use river to classify a 2-level hierarchy, that is passing the labels as a 2-d array (e.g., product, sub-product).

Sadly not yet no 😃

It is on the roadmap though to frame multi-class classification as a tree of binary classifiers. But it will take a lot of work to get that done.