asd

app.py

@@ -52,8 +52,9 @@
 # import dependencies
 import gradio as gr
 from openai import OpenAI
-import torch
-from transformers import AutoTokenizer, AutoModelForSequenceClassification, TFAutoModelForSequenceClassification
+import os
+import re
+from transformers import pipeline, DistilBertForSequenceClassification, DistilBertTokenizerFast
 
 # define the openai key
 api_key = "sk-proj-UCoZZMs4MyfyHwXdHjT8T3BlbkFJjYkSZyPfIPNqXfXwoekm"
@@ -64,24 +65,16 @@ client = OpenAI(api_key = api_key)
 # finetuned model instance
 finetuned_model = "ft:gpt-3.5-turbo-0125:personal::9qGC8cwZ"
 
-# Load model directly
+# Load the AI detection model
 model_name = "tommyliphys/ai-detector-distilbert"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = DistilBertForSequenceClassification.from_pretrained(model_name)
+tokenizer = DistilBertTokenizerFast.from_pretrained(model_name)
 
-# Load TensorFlow model
-tf_model = TFAutoModelForSequenceClassification.from_pretrained(model_name, from_tf=True)
-
-# Convert to PyTorch model
-model = AutoModelForSequenceClassification.from_pretrained(model_name, from_tf=True)
+pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)
 
 # Define the function to get predictions
 def get_prediction(text):
-    inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=512)
-    with torch.no_grad():
-        outputs = model(**inputs)
-    probabilities = torch.nn.functional.softmax(outputs.logits, dim=-1)
-    ai_probability = probabilities[0][1].item()  # Assuming 1 is the index for "AI"
-    return {"label": "AI" if ai_probability > 0.5 else "Human", "score": ai_probability}
+    return pipe(text)[0]
 
 # function to humanize the text
 def humanize_text(AI_text):

requirements.txt

@@ -1,3 +1,5 @@
 openai
 transformers
-torch
+torch
+tensorflow
+tf-keras