EduHinglish — MuRIL query-intent classifier (LoRA)

Classifies a student's Hinglish/English question into one of 15 intents (explain_concept, definition, ...). The prediction is passed to the EduHinglish Gemma generator as a hint.

from transformers import AutoModelForSequenceClassification, AutoTokenizer
from peft import PeftModel
base = AutoModelForSequenceClassification.from_pretrained("google/muril-base-cased", num_labels=15)
model = PeftModel.from_pretrained(base, "persuasivepost/eduhinglish-muril-intent")
tok = AutoTokenizer.from_pretrained("google/muril-base-cased")