{ "cells": [ { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "!pip -q install -U transformers smirk datasets\n" ], "id": "install" }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "from transformers import AutoModel\nfrom smirk import SmirkTokenizerFast\n\n# Load the model\nmodel = AutoModel.from_pretrained(\n 'mist-models/mist-1.8B-g53ib4q2-tmQM',\n trust_remote_code=True\n)\n\n# Make predictions on tmQM test-set molecules\nsmiles_batch = [\n \"Cl[Pt@SP2]12N3[CH][CH][CH][CH][C]3[N][C]3[C]4[N][C]([C]5[CH][CH][CH][CH][CH]5)[C]([C]5[CH][CH][CH][CH][CH]5)[N][C]4[C]([N][C]4[CH][CH][CH][CH]N41)N32\", # FUSTEU\n \"CC12O[Zn]134(O2)(N1[CH][CH][CH][C]2[CH][CH][C]5[CH][CH][CH]N3[C]5[C]21)N1[CH][CH][CH][C]2[CH][CH][C]3[CH][CH][CH]N4[C]3[C]21\", # YEHLIG\n \"CS(C)([O])[Ru@OH21]([OH2])([OH2])([OH2])(N1[C]([O])[C]2[CH][CH][CH][CH][C]2S1([O])[O])N1[C]([O])[C]2[CH][CH][CH][CH][C]2S1([O])[O]\", # ETUDAB\n]\nresults = model.predict(smiles_batch)\nresults\n" ], "id": "inference" } ], "metadata": { "colab": { "provenance": [] }, "kernelspec": { "name": "python3", "display_name": "Python 3" }, "language_info": { "name": "python" } }, "nbformat": 4, "nbformat_minor": 5 }