Scribd
Upload
12 views1 page

Model Utils

The document contains a Python script for predicting emotions using a BERT model. It loads a pre-trained model and tokenizer, normalizes input text, and predicts the emotion class based on the input. The script also includes functionality to load label classes from a JSON file.

Uploaded by

anubrata.oneplus2021
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
12 views1 page

Model Utils

The document contains a Python script for predicting emotions using a BERT model. It loads a pre-trained model and tokenizer, normalizes input text, and predicts the emotion class based on the input. The script also includes functionality to load label classes from a JSON file.

Uploaded by

anubrata.oneplus2021
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 1

# utils/model_utils.

py

import torch
import json
import os
from transformers import AutoTokenizer, AutoModelForSequenceClassification
from utils.normalizer import normalize_bengali_english # ✅ use shared normalizer

# Load model & tokenizer


model_dir = "bert-emotion-model"
tokenizer = AutoTokenizer.from_pretrained(model_dir)
model = AutoModelForSequenceClassification.from_pretrained(model_dir)
model.eval()

# Load label classes


with open(os.path.join(model_dir, "labels.json")) as f:
labels = json.load(f)
id2label = {i: label for i, label in enumerate(labels)}

# Emotion prediction using BERT


def predict_emotion(text):
text = normalize_bengali_english(text) # ✅ normalize slang input
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
with torch.no_grad():
logits = model(**inputs).logits
predicted_class_id = logits.argmax().item()
return id2label[predicted_class_id]

You might also like