177 lines (176 with data), 4.5 kB

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.chdir('../')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from DeepPurpose import DTI as models \n",
    "from DeepPurpose import utils, dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are two ways to load the pretrained model, the first way is to load from local model directory. \n",
    "\n",
    "A model directory should consist of two files: \n",
    "\n",
    "config.pkl that describes the configuration of the model and \n",
    "model.pt, which is the model weights. \n",
    "\n",
    "If you use model.save(MODEL_DIR) to save the model, then, it should be good.\n",
    "\n",
    "The below code exemplifies suppose your model is in the 'path', then, you can load the model using path_dir parameter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Beginning Downloading Morgan_AAC_DAVIS Model...\n",
      "Downloading finished... Beginning to extract zip file...\n",
      "pretrained model Successfully Downloaded...\n"
     ]
    }
   ],
   "source": [
    "path = utils.download_pretrained_model('Morgan_AAC_DAVIS')\n",
    "net = models.model_pretrained(path_dir = path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input_dim_drug': 1024,\n",
       " 'input_dim_protein': 8420,\n",
       " 'hidden_dim_drug': 256,\n",
       " 'hidden_dim_protein': 256,\n",
       " 'cls_hidden_dims': [1024, 1024, 512],\n",
       " 'batch_size': 256,\n",
       " 'train_epoch': 100,\n",
       " 'test_every_X_epoch': 20,\n",
       " 'LR': 0.001,\n",
       " 'drug_encoding': 'Morgan',\n",
       " 'target_encoding': 'AAC',\n",
       " 'result_folder': './result/',\n",
       " 'binary': False,\n",
       " 'mlp_hidden_dims_drug': [1024, 256, 64],\n",
       " 'mlp_hidden_dims_target': [1024, 256, 64],\n",
       " 'num_workers': 0}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "net.config"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For models that provided by us, you can directly use the pre-designated model names. The full list is in the Github README https://github.com/kexinhuang12345/DeepPurpose/blob/master/README.md#pretrained-models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Beginning Downloading MPNN_CNN_DAVIS Model...\n",
      "Downloading finished... Beginning to extract zip file...\n",
      "pretrained model Successfully Downloaded...\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'input_dim_drug': 1024,\n",
       " 'input_dim_protein': 8420,\n",
       " 'hidden_dim_drug': 128,\n",
       " 'hidden_dim_protein': 256,\n",
       " 'cls_hidden_dims': [1024, 1024, 512],\n",
       " 'batch_size': 128,\n",
       " 'train_epoch': 100,\n",
       " 'LR': 0.001,\n",
       " 'drug_encoding': 'MPNN',\n",
       " 'target_encoding': 'CNN',\n",
       " 'result_folder': './result/',\n",
       " 'binary': False,\n",
       " 'mpnn_hidden_size': 128,\n",
       " 'mpnn_depth': 3,\n",
       " 'cnn_target_filters': [32, 64, 96],\n",
       " 'cnn_target_kernels': [4, 8, 12],\n",
       " 'num_workers': 0}"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "net = models.model_pretrained(model = 'MPNN_CNN_DAVIS')\n",
    "net.config"
   ]
  },
  {
   "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}