tutorials/villin_headpiece/SRVs_Cross_Validation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "807feeca",
   "metadata": {},
   "outputs": [],
   "source": [
    "MDtrajs_dir=\"./MDtrajs/\"\n",
    "Featurization_dir=\"./Featurization/\"\n",
    "FSelection_dir = \"./FSel/\"\n",
    "GMRQ_dir=\"./GMRQ/\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e7a3e40c",
   "metadata": {},
   "source": [
    "### GMRQ Cross Validation of SRVNet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "603b5078",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pickle\n",
    "import glob\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from matplotlib import pyplot as plt\n",
    "from sklearn.model_selection import KFold\n",
    "from msmbuilder.decomposition import tICA\n",
    "from msmbuilder.cluster import KMeans, KCenters, MiniBatchKMeans\n",
    "from msmbuilder.msm import MarkovStateModel\n",
    "import multiprocessing\n",
    "from joblib import Parallel, delayed\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "from tqdm import *\n",
    "\n",
    "import sys\n",
    "sys.path.insert(0, './scripts')\n",
    "from VAMPNet_SRVNet import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "bd2da99b",
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_pickles(file_dir):\n",
    "    pickle_files = glob.glob(file_dir)\n",
    "    data_list = []\n",
    "    for file in pickle_files:\n",
    "        with open(file, \"rb\") as f:\n",
    "            data = pickle.load(f)\n",
    "            data_list.append(data)\n",
    "    return data_list\n",
    "\n",
    "srvCV_dir = \"./GMRQ/srv/\"\n",
    "if not os.path.exists(srvCV_dir):\n",
    "        os.mkdir(srvCV_dir)\n",
    "\n",
    "for para in ['lagtime', 'n_cvs', 'n_clusters']:\n",
    "    para_dir = srvCV_dir + para + '/'\n",
    "    if not os.path.exists(para_dir):\n",
    "        os.mkdir(para_dir)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8f6a0358",
   "metadata": {},
   "outputs": [],
   "source": [
    "oasis_trajs = np.load(FSelection_dir + \"oasis_trajs.npy\", allow_pickle=True)\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    device = torch.device(\"cuda\")\n",
    "    torch.backends.cudnn.benchmark = True\n",
    "else:\n",
    "    device = torch.device(\"cpu\")\n",
    "\n",
    "def SRV_CV(trajs, lt=200, n_cvs=3, n_clusters=200, para='lagtime',\n",
    "           lr=1e-4, n_epochs=15, train_batchsize=50000,\n",
    "           runs=10, n_fold=5, device=None):\n",
    "    \n",
    "    output_files_dir = srvCV_dir + para + '/'\n",
    "    \n",
    "    results = []\n",
    "    for run in range(runs):\n",
    "        cv = KFold(n_splits=n_fold, shuffle=True)\n",
    "        for train_index, test_index in cv.split(trajs):\n",
    "            network_lobe = nn.Sequential(\n",
    "                nn.BatchNorm1d(400), \n",
    "                nn.Linear(400, 200), nn.ELU(), \n",
    "                nn.Linear(200, 100), nn.ELU(), \n",
    "                nn.Linear(100, 50), nn.ELU(), \n",
    "                nn.Linear(50, 20), nn.ELU(), \n",
    "                nn.Linear(20, 10), nn.ELU(), \n",
    "                nn.Linear(10, 5), nn.ELU(), \n",
    "                nn.Linear(5, n_cvs))\n",
    "            network_lobe = network_lobe.to(device=device)\n",
    "        \n",
    "            projector = deep_projector(network_type='SRVNet', lobe=network_lobe,\n",
    "                                       epsilon=1e-6, learning_rate=lr, device=device)\n",
    "        \n",
    "            \n",
    "            train_data = [trajs[i] for i in train_index]\n",
    "            test_data = [trajs[i] for i in test_index]\n",
    "            \n",
    "            past, future = TimeLaggedDataset(trajs=train_data, lagtime=lt, normalize=False)\n",
    "            train_loader =  split_train_validate_data(pastdata=past, futuredata=future, \n",
    "                                                      validation_ratio=0, train_batchsize=train_batchsize)\n",
    "            projector.fit(train_loader=train_loader, num_epochs=n_epochs)\n",
    "            \n",
    "            srv_traj_train = projector.transform(data=train_data)\n",
    "            srv_traj_test = projector.transform(data=test_data)\n",
    "            del projector\n",
    "            \n",
    "            # This tICA is not used for dimensionality reduction\n",
    "            # but for orthogonalization\n",
    "            tica = tICA(n_components=n_cvs, lag_time=lt, kinetic_mapping=False)\n",
    "            tica.fit(srv_traj_train)\n",
    "            ntrajs_train = tica.transform(srv_traj_train)\n",
    "            ntrajs_test = tica.transform(srv_traj_test)\n",
    "            del tica\n",
    "            \n",
    "            cluster = KCenters(n_clusters=n_clusters)\n",
    "            cluster.fit(ntrajs_train)\n",
    "            ctraj_train = cluster.predict(ntrajs_train)\n",
    "            ctraj_test = cluster.predict(ntrajs_test)\n",
    "            del cluster\n",
    "            \n",
    "            msm = MarkovStateModel(n_timescales=3, lag_time=500, reversible_type='transpose', \n",
    "                                   verbose=False, ergodic_cutoff='off')\n",
    "            msm.fit(ctraj_train)\n",
    "            train_score = msm.score(ctraj_train)\n",
    "            test_score = msm.score(ctraj_test)\n",
    "            \n",
    "            results.append({\n",
    "                'type': 'train', 'score': train_score, \n",
    "                'lagtime': lt, 'n_cvs': n_cvs, 'n_clusters': n_clusters})\n",
    "            results.append({\n",
    "                'type': 'test', 'score': test_score, \n",
    "                'lagtime': lt, 'n_cvs': n_cvs, 'n_clusters': n_clusters})\n",
    "            gc.collect()\n",
    "            torch.cuda.empty_cache()\n",
    "        print('Run {} for {}lt, {}cvs, {}clusters is complete'.format(run,lt,n_cvs,n_clusters))\n",
    "    frame = pd.DataFrame(results)\n",
    "    frame.to_pickle(output_files_dir + \"{}lt_{}cvs_{}clusters_gmrq.pickl\".format(lt,n_cvs,n_clusters))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "44013575",
   "metadata": {},
   "outputs": [],
   "source": [
    "lt_list = [50, 100, 200, 300, 400, 500]\n",
    "Parallel(n_jobs=3)(delayed(SRV_CV)(trajs=oasis_trajs, \n",
    "                                   lt=i, n_cvs=3, n_clusters=200, \n",
    "                                   para='lagtime') for i in lt_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4a44a9c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "ncvs_list = [2, 3, 4, 5]\n",
    "Parallel(n_jobs=4)(delayed(SRV_CV)(trajs=oasis_trajs, \n",
    "                                   lt=200, n_cvs=i, n_clusters=200, \n",
    "                                   para='n_cvs') for i in ncvs_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fed468fd",
   "metadata": {},
   "outputs": [],
   "source": [
    "nc_list = [50, 100, 200, 300, 400, 500]\n",
    "Parallel(n_jobs=3)(delayed(SRV_CV)(trajs=oasis_trajs, \n",
    "                                   lt=200, n_cvs=3, n_clusters=i, \n",
    "                                   para='n_clusters') for i in nc_list)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "759ad6c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1800x800 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.rcParams['font.size'] = 40\n",
    "fig, axes = plt.subplots(1, 3, figsize=(18, 8), sharey=True,\n",
    "                         gridspec_kw={'width_ratios': [6,4,5]})\n",
    "axes = axes.ravel()\n",
    "pal = {'train': '#67a9cf', 'test': '#ed5736'}\n",
    "\n",
    "para = ['lagtime', 'n_cvs', 'n_clusters']\n",
    "for i in range(len(para)):\n",
    "    file_dir = srvCV_dir + para[i] + '/'\n",
    "    data = load_pickles(file_dir + '*.pickl')\n",
    "    df = pd.concat(data, axis=0)\n",
    "    df['lagtime'] = np.int16(df['lagtime'] * 0.2)\n",
    "    sns.boxplot(data=df, x=para[i], y='score', hue='type', \n",
    "                orient=\"v\", notch=False, \n",
    "                palette=pal, linewidth=3, width=0.7, \n",
    "                sym=\".\", showfliers=False, ax=axes[i])\n",
    "    axes[i].set_ylabel(\"\")   \n",
    "    axes[i].set_ylim(2, 4)\n",
    "    axes[i].legend([], frameon=False)\n",
    "    axes[i].tick_params(axis='both', direction='in', width=2, length=8)\n",
    "    for spine in axes[i].spines.values():\n",
    "        spine.set_linewidth(4)\n",
    "\n",
    "axes[0].set_xlabel(\"lagtime (ns)\")\n",
    "axes[1].set_xlabel(\"Number of CVs\")\n",
    "axes[2].set_xlabel(\"Number of clusters\")\n",
    "fig.text(0.01, 0.55, \"GMRQ Score\", ha='center', va='center', rotation=90)\n",
    "fig.text(0.5, 0.95, \"SRVs-MSM Cross-Validation\", ha='center', va='center')\n",
    "# axes[2].legend(loc='center', bbox_to_anchor=(-0.5, 1.1), ncol=2)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.subplots_adjust(wspace=0.05, hspace=0.3)\n",
    "plt.savefig('./png/gmrq_srv.png', dpi=500, bbox_inches='tight')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e705dbf2",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}