openai-clip-puipui.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! git clone https://github.com/openai/CLIP.git"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! pip install ftfy regex tqdm"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.chdir('CLIP')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import clip\n",
    "from PIL import Image\n",
    "from IPython.display import Image as display_image\n",
    "\n",
    "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
    "model, preprocess = clip.load(\"ViT-B/32\", device=device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def predict(image_file, class_text):\n",
    "    image = preprocess(Image.open(image_file)).unsqueeze(0).to(device)\n",
    "    text = clip.tokenize(class_text).to(device)\n",
    "\n",
    "    with torch.no_grad():\n",
    "        image_features = model.encode_image(image)\n",
    "        text_features = model.encode_text(text)\n",
    "\n",
    "        logits_per_image, logits_per_text = model(image, text)\n",
    "        probs = logits_per_image.softmax(dim=-1).cpu().numpy()\n",
    "\n",
    "    pil_img = display_image(filename=image_file)\n",
    "    display(pil_img)\n",
    "    \n",
    "    for i in range(len(class_text)):\n",
    "        print(f\"{class_text[i]}: {probs[0][i]}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_file = '../puipui-1.png'\n",
    "class_text = [\"traffic jam\", \"traffic smooth\", \"emergency\"]\n",
    "predict(image_file, class_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_file = '../puipui-2.png'\n",
    "class_text = [\"traffic jam\", \"traffic smooth\", \"emergency\"]\n",
    "predict(image_file, class_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_file = '../puipui-3.png'\n",
    "class_text = [\"disappointment\", \"hesitation\", \"expectation\"]\n",
    "predict(image_file, class_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image_file = '../puipui-4.png'\n",
    "class_text = [\"spring\", \"summer\", \"fall\", \"winter\"]\n",
    "predict(image_file, class_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}