classifier_maker_checker

classifier_maker_checker.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from transformers import CLIPProcessor, CLIPModel\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def maker(image):\n",
    "    model = CLIPModel.from_pretrained(\"openai/clip-vit-base-patch32\")\n",
    "    processor = CLIPProcessor.from_pretrained(\"openai/clip-vit-base-patch32\")\n",
    "    descriptions = [\"A driver's license\", \"A passport\", \"A student ID\"]\n",
    "\n",
    "    inputs = processor(text=descriptions, images=image, return_tensors=\"pt\", padding=True)\n",
    "    outputs = model(**inputs)\n",
    "    logits_per_image = outputs.logits_per_image  \n",
    "    probs = logits_per_image.softmax(dim=1) \n",
    "    predicted_class = descriptions[probs.argmax()]\n",
    "    confidence = probs.max().item()\n",
    "\n",
    "    if confidence > .5:\n",
    "        return predicted_class\n",
    "    else:\n",
    "        return None\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os  \n",
    "import io\n",
    "import json\n",
    "import base64\n",
    "from openai import AzureOpenAI\n",
    "\n",
    "def checker(image, classification):\n",
    "    endpoint = \"https://aiftrained1464353287.openai.azure.com/\"\n",
    "    deployment = \"gpt-4o\"  \n",
    "    subscription_key = \"55gTRdNCbkCBWbFwE75y2U3HAjzUxCiydhDovi5Ilqd5SMMsaup4JQQJ99BAACYeBjFXJ3w3AAAAACOGyNll\"\n",
    "\n",
    "    client = AzureOpenAI(  azure_endpoint=endpoint, api_key=subscription_key, api_version=\"2024-05-01-preview\",)\n",
    "        \n",
    "    buffered = io.BytesIO()\n",
    "    image.save(buffered, format=\"PNG\")\n",
    "    encoded_image =  base64.b64encode(buffered.getvalue()).decode('ascii')        \n",
    "\n",
    "    chat_prompt = [\n",
    "        {\n",
    "            \"role\": \"system\",\n",
    "            \"content\": [\n",
    "                {\n",
    "                    \"type\": \"text\",\n",
    "                    \"text\": \"You are an AI assistant that helps classify fictional legal documents.\"\n",
    "                }\n",
    "            ]\n",
    "        },\n",
    "        {\n",
    "            \"role\": \"user\",\n",
    "            \"content\": [\n",
    "                {\n",
    "                    \"type\": \"text\",\n",
    "                    \"text\": \"Is this document \" + classification + \"? \"\n",
    "                },\n",
    "                {\n",
    "                    \"type\": \"image_url\",\n",
    "                    \"image_url\": {\n",
    "                        \"url\": f\"data:image/jpeg;base64,{encoded_image}\"\n",
    "                    }\n",
    "                },\n",
    "                {\n",
    "                    \"type\": \"text\",\n",
    "                    \"text\": \"\\n\"\n",
    "                }\n",
    "            ]\n",
    "        }\n",
    "    ] \n",
    "\n",
    "    messages = chat_prompt  \n",
    "    completion = client.chat.completions.create(model=deployment, messages=messages, max_tokens=800,  \n",
    "        temperature=0.7, top_p=0.95, frequency_penalty=0, presence_penalty=0, stop=None, stream=False)\n",
    "\n",
    "    json_string = completion.to_json() \n",
    "    data = json.loads(json_string)\n",
    "    return data['choices'][0]['message']['content']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "A driver's license\n",
      "Based on the layout and elements such as the use of the word \"license,\" this document resembles a fictional driver's license, although the text and format appear distorted or nonsensical.\n"
     ]
    }
   ],
   "source": [
    "input_folder = 'data_new'\n",
    "input_filename = '0.png'\n",
    "image_path = os.path.join(input_folder, input_filename)\n",
    "image = Image.open(image_path)\n",
    "\n",
    "classification = maker(image)\n",
    "print(classification)\n",
    "validation = checker(image, classification)\n",
    "print(validation)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}