{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e7caa8ad",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4, 320)\n",
      "tensor([[ 1.0000,  0.1580, -0.4305, -0.5529],\n",
      "        [ 0.1580,  1.0000,  0.6916,  0.6522],\n",
      "        [-0.4305,  0.6916,  1.0000,  0.9836],\n",
      "        [-0.5529,  0.6522,  0.9836,  1.0000]])\n"
     ]
    }
   ],
   "source": [
    "from sentence_transformers import SentenceTransformer\n",
    "\n",
    "# Download from the 🤗 Hub\n",
    "model = SentenceTransformer(\"gbyuvd/miniChembed-prototype\")\n",
    "# Run inference\n",
    "sentences = [\n",
    "    'O=C1/C=C\\\\C=C2/N1C[C@@H]3CNC[C@H]2C3', # Cytisine\n",
    "    \"n1c2cc3c(cc2ncc1)[C@@H]4CNC[C@H]3C4\", # Varenicline\n",
    "    \"c1ncccc1[C@@H]2CCCN2C\", # Nicotine\n",
    "    'Nc1nc2cncc-2co1', # CID: 162789184  \n",
    "]\n",
    "embeddings = model.encode(sentences)\n",
    "print(embeddings.shape)\n",
    "# (4, 768)\n",
    "\n",
    "# Get the similarity scores for the embeddings\n",
    "similarities = model.similarity(embeddings, embeddings)\n",
    "print(similarities)\n",
    "# tensor([[1.0000, 0.4397, 0.3172, 0.0382],\n",
    "#        [0.4397, 1.0000, 0.2543, 0.1725],\n",
    "#        [0.3172, 0.2543, 1.0000, 0.2371],\n",
    "#        [0.0382, 0.1725, 0.2371, 1.0000]])\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.13.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}