Tutorial for multimodal_transformers

data_loading.py

import pandas as pd
from multimodal_transformers.data import load_data
from transformers import AutoTokenizer

data_df = pd.read_csv('Womens Clothing E-Commerce Reviews.csv')
text_cols = ['Title', 'Review Text']
# The label col is expected to contain integers from 0 to N_classes - 1
label_col = 'Recommended IND' 
categorical_cols = ['Clothing ID', 'Division Name', 'Department Name', 'Class Name']
numerical_cols = ['Rating', 'Age', 'Positive Feedback Count']
label_list = ['Not Recommended', 'Recommended'] # what each label class represents

tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
# make sure NaN values for cat columns are filled before passing to load_data
for c in categorical_cols:  
    data_df.loc[:, c] = data_df.loc[:, c].astype(str).fillna("-9999999")

torch_dataset = load_data(
    data_df,
    text_cols,
    tokenzier,
    categorical_cols=categorical_cols,
    numerical_cols=numerical_cols,
    sep_text_token_str=tokenizer.sep_token
)