handle both ar and eng

Dockerfile

@@ -1,5 +1,5 @@
-# Use Python 3.10 as the base image to support scipy>=1.14.1
-FROM python:3.10
+# Use Python 3.9 as the base image
+FROM python:3.9
 
 # Set working directory in the container
 WORKDIR /app
@@ -22,7 +22,7 @@ COPY --chown=myuser:myuser requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Clear cache and pre-download models
-RUN rm -rf /app/.cache/huggingface/* && python -c "from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM; pipeline('text-classification', model='Hello-SimpleAI/chatgpt-detector-roberta'); pipeline('text-classification', model='openai-community/roberta-large-openai-detector'); pipeline('text-classification', model='sabaridsnfuji/arabic-ai-text-detector'); AutoTokenizer.from_pretrained('gpt2'); AutoModelForCausalLM.from_pretrained('gpt2'); AutoTokenizer.from_pretrained('aubmindlab/araGPT2'); AutoModelForCausalLM.from_pretrained('aubmindlab/araGPT2')"
+RUN rm -rf /app/.cache/huggingface/* && python -c "from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM; pipeline('text-classification', model='Hello-SimpleAI/chatgpt-detector-roberta'); pipeline('text-classification', model='openai-community/roberta-large-openai-detector'); pipeline('text-classification', model='sabaridsnfuji/arabic-ai-text-detector'); AutoTokenizer.from_pretrained('gpt2'); AutoModelForCausalLM.from_pretrained('gpt2'); AutoTokenizer.from_pretrained('aubmindlab/aragpt2-base'); AutoModelForCausalLM.from_pretrained('aubmindlab/aragpt2-base')"
 
 # Copy the application code
 COPY --chown=myuser:myuser . .

app.py

@@ -2,7 +2,7 @@ from fastapi import FastAPI, HTTPException
 from pydantic import BaseModel, validator
 import re
 import torch
-from transformers import AutoTokenizer, AutoModelForSequenceClassification, AutoModelForCausalLM, pipeline
+from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
 from collections import Counter
 import logging
 import numpy as np
@@ -29,8 +29,8 @@ ppl_english = {
     "model": AutoModelForCausalLM.from_pretrained("gpt2").to(device)
 }
 ppl_arabic = {
-    "tokenizer": AutoTokenizer.from_pretrained("aubmindlab/araGPT2"),
-    "model": AutoModelForCausalLM.from_pretrained("aubmindlab/araGPT2").to(device)
+    "tokenizer": AutoTokenizer.from_pretrained("aubmindlab/aragpt2-base"),
+    "model": AutoModelForCausalLM.from_pretrained("aubmindlab/aragpt2-base").to(device)
 }
 
 def detect_language(text: str) -> str: