results_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c894e82b",
   "metadata": {},
   "source": [
    "# Results Analysis\n",
    "\n",
    "This notebook uses the saved results objects to analyse the results of the trained models over multiple runs. It will look at each of the datasets in turn, with each method studied within."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2faf85be",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Imports\n",
    "import os\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import json\n",
    "%matplotlib inline\n",
    "\n",
    "def get_files(results_folder, dataset):\n",
    "    all_files = list(set([os.path.join(results_folder, f) for f in os.listdir(results_folder) if (dataset+\"_\") in f]))\n",
    "    return all_files\n",
    "\n",
    "# function for loading and creating a pandas dataframe of the results and the hyperparameters\n",
    "def process_results2pandas(results_folder, dataset):\n",
    "    \"\"\"Creates a pandas table from the saved results dicts\n",
    "    related to the input dataset name\n",
    "    \n",
    "    Args:\n",
    "        results_folder (str or Path)\n",
    "        dataset = (str) string name of dataset to pick out, should be one of:\n",
    "            drugcomb, drugcombdb, drugbankddi, OncoPolyPharmacology, twosides\n",
    "    \n",
    "    \"\"\"\n",
    "    df_columns = [\"dataset_name\", \"emb_decomposition\", \"emb_dimension\", \"emb_num_epochs\", \"model_name\", \"model_num_epochs\", \"model_batch_size\", \"train_size\", \"random_seed\", \"results_auc\", \"results_losses\"]\n",
    "    df_rows = []\n",
    "    all_files = get_files(results_folder, dataset)\n",
    "    for fpath in all_files:\n",
    "        with open(fpath) as fh:\n",
    "            results_dict = json.load(fh) \n",
    "        f = os.path.basename(fpath)\n",
    "        f = f.replace(\".json\", \"\")\n",
    "        \n",
    "        if len(f.split(\"_\")) == 10:\n",
    "            dataset_name, emb_decomposition, emb_dimension, emb_num_epochs, model_name, model_num_epochs, _, model_batch_size, train_size, random_seed = f.split(\"_\")\n",
    "        else:\n",
    "            dataset_name, model_name, model_num_epochs, _, model_batch_size, train_size, random_seed = f.split(\"_\")\n",
    "            emb_decomposition, emb_dimension, emb_num_epochs = 0,0,0\n",
    "            \n",
    "        # Reset Types\n",
    "        dataset_name = dataset_name\n",
    "        emb_decomposition = emb_decomposition\n",
    "        emb_dimension = int(emb_dimension)\n",
    "        emb_num_epochs = int(emb_num_epochs)\n",
    "        model_name = model_name\n",
    "        model_num_epochs = int(model_num_epochs)\n",
    "        model_batch_size = int(model_batch_size)\n",
    "        train_size = float(train_size)\n",
    "        random_seed = int(random_seed)\n",
    "        \n",
    "        results_auc = results_dict['evaluation']['roc_auc']\n",
    "        results_losses = np.array(results_dict['losses'])\n",
    "    \n",
    "        # Populate table\n",
    "        row = [dataset_name, emb_decomposition, emb_dimension, emb_num_epochs, model_name, model_num_epochs, model_batch_size, train_size, random_seed, results_auc, results_losses]\n",
    "        df_rows.append(row)\n",
    "    \n",
    "    df = pd.DataFrame(df_rows, columns=df_columns)\n",
    "    return df\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1434a113",
   "metadata": {},
   "source": [
    "# Study hyperparameters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "d0531161",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "embdim = 64\n",
    "embepochs = 1000\n",
    "model_num_epochs = 250"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0d9c801e",
   "metadata": {},
   "source": [
    "# DrugCombDB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "336630cf",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "### DROnly\n",
      "DROnly_sp: auc: 0.711 +- 0.004 with 5 samples\n",
      "DROnly_wl3: auc: 0.763 +- 0.002 with 5 samples\n",
      "\n",
      "### DeepDDSDR\n",
      "DeepDDSDR_sp: auc: 0.79 +- 0.003 with 5 samples\n",
      "DeepDDSDR_wl3: auc: 0.799 +- 0.004 with 5 samples\n",
      "\n",
      "### DeepDRSynergy\n",
      "DeepDRSynergy_sp: auc: 0.813 +- 0.003 with 5 samples\n",
      "DeepDRSynergy_wl3: auc: 0.814 +- 0.004 with 5 samples\n",
      "\n",
      "### DeepDrugDR\n",
      "DeepDrugDR_sp: auc: 0.743 +- 0.0 with 5 samples\n",
      "DeepDrugDR_wl3: auc: 0.743 +- 0.001 with 5 samples\n",
      "\n",
      "### EPGCNDSDR\n",
      "EPGCNDSDR_sp: auc: 0.704 +- 0.001 with 5 samples\n",
      "EPGCNDSDR_wl3: auc: 0.711 +- 0.002 with 5 samples\n",
      "\n",
      "### MatchMakerDR\n",
      "MatchMakerDR_sp: auc: 0.784 +- 0.004 with 5 samples\n",
      "MatchMakerDR_wl3: auc: 0.783 +- 0.002 with 5 samples\n"
     ]
    }
   ],
   "source": [
    "df = process_results2pandas(\"results\", \"drugcombdb\")\n",
    "\n",
    "# Sample down\n",
    "df = df[((df['emb_dimension']==embdim) | (df['emb_dimension']==0)) & ((df['emb_num_epochs']==embepochs) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "\n",
    "# Summarise\n",
    "for model_name in sorted(list(set(df.model_name))):\n",
    "    if model_name in [\"DROnly\", \"DeepDRSynergy\", \"EPGCNDSDR\", \"DeepDDSDR\", \"MatchMakerDR\", \"DeepDrugDR\"]:\n",
    "        model_specific_df = df[(df['model_name']==model_name)]\n",
    "        print(f\"\\n### {model_name}\")\n",
    "        for decomposition in sorted(list(set(model_specific_df.emb_decomposition) - set([\"wl1\", \"wl2\"]))):\n",
    "            ddf = model_specific_df[(model_specific_df['emb_decomposition']==decomposition)]\n",
    "            mean_auc = np.around(np.mean(ddf.results_auc), 3)\n",
    "            std_auc = np.around(np.std(ddf.results_auc), 3)\n",
    "            num_emp_samples = len(ddf.results_auc)\n",
    "            print(f\"{model_name}_{decomposition}: auc: {mean_auc} +- {std_auc} with {num_emp_samples} samples\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b886df9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ## Performance Histograms\n",
    "# sns.catplot(x = \"model_name\", y = \"results_auc\", kind=\"bar\", data=df)\n",
    "# plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cc75eae",
   "metadata": {},
   "source": [
    "# DrugComb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "dce7eae9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "### DROnly\n",
      "DROnly_sp: auc: 0.621 +- 0.002 with 5 samples\n",
      "DROnly_wl3: auc: 0.651 +- 0.002 with 5 samples\n",
      "\n",
      "### DeepDDSDR\n",
      "DeepDDSDR_sp: auc: 0.694 +- 0.002 with 5 samples\n",
      "DeepDDSDR_wl3: auc: 0.7 +- 0.002 with 5 samples\n",
      "\n",
      "### DeepDRSynergy\n",
      "DeepDRSynergy_sp: auc: 0.74 +- 0.004 with 5 samples\n",
      "DeepDRSynergy_wl3: auc: 0.738 +- 0.002 with 5 samples\n",
      "\n",
      "### DeepDrugDR\n",
      "DeepDrugDR_sp: auc: 0.648 +- 0.0 with 5 samples\n",
      "DeepDrugDR_wl3: auc: 0.648 +- 0.001 with 5 samples\n",
      "\n",
      "### EPGCNDSDR\n",
      "EPGCNDSDR_sp: auc: 0.622 +- 0.001 with 5 samples\n",
      "EPGCNDSDR_wl3: auc: 0.627 +- 0.001 with 5 samples\n",
      "\n",
      "### MatchMakerDR\n",
      "MatchMakerDR_sp: auc: 0.714 +- 0.003 with 5 samples\n",
      "MatchMakerDR_wl3: auc: 0.714 +- 0.005 with 5 samples\n"
     ]
    }
   ],
   "source": [
    "df = process_results2pandas(\"results\", \"drugcomb\")\n",
    "\n",
    "# Sample down\n",
    "df = df[((df['emb_dimension']==embdim) | (df['emb_dimension']==0)) & ((df['emb_num_epochs']==embepochs) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "\n",
    "# Summarise\n",
    "for model_name in sorted(list(set(df.model_name))):\n",
    "    if model_name in [\"DROnly\", \"DeepDRSynergy\", \"EPGCNDSDR\", \"DeepDDSDR\", \"MatchMakerDR\", \"DeepDrugDR\"]:\n",
    "        model_specific_df = df[(df['model_name']==model_name)]\n",
    "        print(f\"\\n### {model_name}\")\n",
    "        for decomposition in sorted(list(set(model_specific_df.emb_decomposition) - set([\"wl1\", \"wl2\"]))):\n",
    "            ddf = model_specific_df[(model_specific_df['emb_decomposition']==decomposition)]\n",
    "            mean_auc = np.around(np.mean(ddf.results_auc), 3)\n",
    "            std_auc = np.around(np.std(ddf.results_auc), 3)\n",
    "            num_emp_samples = len(ddf.results_auc)\n",
    "            print(f\"{model_name}_{decomposition}: auc: {mean_auc} +- {std_auc} with {num_emp_samples} samples\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "443769ba",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ## Performance Histograms\n",
    "# sns.catplot(x = \"model_name\", y = \"results_auc\", kind=\"bar\", data=df)\n",
    "# plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4c99ed59",
   "metadata": {},
   "source": [
    "# DrugBankDDI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "0ec5cab4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "### DROnly\n",
      "DROnly_sp: auc: 0.71 +- 0.005 with 5 samples\n",
      "DROnly_wl3: auc: 0.809 +- 0.005 with 5 samples\n",
      "\n",
      "### DeepDDSDR\n",
      "DeepDDSDR_sp: auc: 0.987 +- 0.001 with 5 samples\n",
      "DeepDDSDR_wl3: auc: 0.989 +- 0.0 with 5 samples\n",
      "\n",
      "### DeepDRSynergy\n",
      "DeepDRSynergy_sp: auc: 0.988 +- 0.001 with 5 samples\n",
      "DeepDRSynergy_wl3: auc: 0.988 +- 0.0 with 5 samples\n",
      "\n",
      "### DeepDrugDR\n",
      "DeepDrugDR_sp: auc: 0.863 +- 0.001 with 5 samples\n",
      "DeepDrugDR_wl3: auc: 0.863 +- 0.0 with 5 samples\n",
      "\n",
      "### EPGCNDSDR\n",
      "EPGCNDSDR_sp: auc: 0.73 +- 0.003 with 5 samples\n",
      "EPGCNDSDR_wl3: auc: 0.741 +- 0.004 with 5 samples\n",
      "\n",
      "### MatchMakerDR\n",
      "MatchMakerDR_sp: auc: 0.991 +- 0.001 with 5 samples\n",
      "MatchMakerDR_wl3: auc: 0.992 +- 0.0 with 5 samples\n"
     ]
    }
   ],
   "source": [
    "df = process_results2pandas(\"results\", \"drugbankddi\")\n",
    "\n",
    "# Sample down\n",
    "df = df[((df['emb_dimension']==embdim) | (df['emb_dimension']==0)) & ((df['emb_num_epochs']==embepochs) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "\n",
    "# Summarise\n",
    "for model_name in sorted(list(set(df.model_name))):\n",
    "    if model_name in [\"DROnly\", \"DeepDRSynergy\", \"EPGCNDSDR\", \"DeepDDSDR\", \"MatchMakerDR\", \"DeepDrugDR\"]:\n",
    "        model_specific_df = df[(df['model_name']==model_name)]\n",
    "        print(f\"\\n### {model_name}\")\n",
    "        for decomposition in sorted(list(set(model_specific_df.emb_decomposition) - set([\"wl1\", \"wl2\"]))):\n",
    "            ddf = model_specific_df[(model_specific_df['emb_decomposition']==decomposition)]\n",
    "            mean_auc = np.around(np.mean(ddf.results_auc), 3)\n",
    "            std_auc = np.around(np.std(ddf.results_auc), 3)\n",
    "            num_emp_samples = len(ddf.results_auc)\n",
    "            print(f\"{model_name}_{decomposition}: auc: {mean_auc} +- {std_auc} with {num_emp_samples} samples\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f111031e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ## Performance Histograms\n",
    "# sns.catplot(x = \"model_name\", y = \"results_auc\", kind=\"bar\", data=df)\n",
    "# plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a3849b05",
   "metadata": {},
   "outputs": [],
   "source": [
    "# sdf = df[(df[\"model_name\"]==\"DeepDDSDR\")]\n",
    "# sdf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "405abae9",
   "metadata": {},
   "source": [
    "# TwoSides"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "51d48e74",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "### DROnly\n",
      "DROnly_sp: auc: 0.823 +- 0.005 with 5 samples\n",
      "DROnly_wl3: auc: 0.917 +- 0.002 with 5 samples\n",
      "\n",
      "### DeepDDSDR\n",
      "DeepDDSDR_sp: auc: 0.943 +- 0.001 with 5 samples\n",
      "DeepDDSDR_wl3: auc: 0.944 +- 0.001 with 5 samples\n",
      "\n",
      "### DeepDRSynergy\n",
      "DeepDRSynergy_sp: auc: 0.935 +- 0.0 with 5 samples\n",
      "DeepDRSynergy_wl3: auc: 0.934 +- 0.002 with 5 samples\n",
      "\n",
      "### DeepDrugDR\n",
      "DeepDrugDR_sp: auc: 0.926 +- 0.0 with 5 samples\n",
      "DeepDrugDR_wl3: auc: 0.926 +- 0.001 with 5 samples\n",
      "\n",
      "### EPGCNDSDR\n",
      "EPGCNDSDR_sp: auc: 0.808 +- 0.002 with 5 samples\n",
      "EPGCNDSDR_wl3: auc: 0.822 +- 0.006 with 5 samples\n",
      "\n",
      "### MatchMakerDR\n",
      "MatchMakerDR_sp: auc: 0.928 +- 0.002 with 5 samples\n",
      "MatchMakerDR_wl3: auc: 0.93 +- 0.001 with 5 samples\n"
     ]
    }
   ],
   "source": [
    "df = process_results2pandas(\"results\", \"twosides\")\n",
    "\n",
    "# Sample down\n",
    "df = df[((df['emb_dimension']==embdim) | (df['emb_dimension']==0)) & ((df['emb_num_epochs']==embepochs) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "\n",
    "# Summarise\n",
    "for model_name in sorted(list(set(df.model_name))):\n",
    "    if model_name in [\"DROnly\", \"DeepDRSynergy\", \"EPGCNDSDR\", \"DeepDDSDR\", \"MatchMakerDR\", \"DeepDrugDR\"]:\n",
    "        model_specific_df = df[(df['model_name']==model_name)]\n",
    "        print(f\"\\n### {model_name}\")\n",
    "        for decomposition in sorted(list(set(model_specific_df.emb_decomposition) - set([\"wl1\", \"wl2\"]))):\n",
    "            ddf = model_specific_df[(model_specific_df['emb_decomposition']==decomposition)]\n",
    "            mean_auc = np.around(np.mean(ddf.results_auc), 3)\n",
    "            std_auc = np.around(np.std(ddf.results_auc), 3)\n",
    "            num_emp_samples = len(ddf.results_auc)\n",
    "            print(f\"{model_name}_{decomposition}: auc: {mean_auc} +- {std_auc} with {num_emp_samples} samples\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "447dcf21",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ## Performance Histograms\n",
    "# sns.catplot(x = \"model_name\", y = \"results_auc\", kind=\"bar\", data=df)\n",
    "# plt.xticks(rotation=45)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "36a01fde",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c6729cd4",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "75541f59",
   "metadata": {},
   "source": [
    "### Utilities for ablation study"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "eefeeaef",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_files_nodataset(results_folder):\n",
    "    all_files = list(set([os.path.join(results_folder, f) for f in os.listdir(results_folder)]))\n",
    "    return all_files\n",
    "\n",
    "# function for loading and creating a pandas dataframe of the results and the hyperparameters\n",
    "def process_results2pandas_nodataset(results_folder):\n",
    "    \"\"\"Creates a pandas table from the saved results dicts\n",
    "    related to the input dataset name\n",
    "    \n",
    "    Args:\n",
    "        results_folder (str or Path)\n",
    "        dataset = (str) string name of dataset to pick out, should be one of:\n",
    "            drugcomb, drugcombdb, drugbankddi, OncoPolyPharmacology, twosides\n",
    "    \n",
    "    \"\"\"\n",
    "    df_columns = [\"dataset_name\", \"emb_decomposition\", \"emb_dimension\", \"emb_num_epochs\", \"model_name\", \"model_num_epochs\", \"model_batch_size\", \"train_size\", \"random_seed\", \"results_auc\", \"results_losses\"]\n",
    "    df_rows = []\n",
    "    all_files = get_files_nodataset(results_folder)\n",
    "    for fpath in all_files:\n",
    "        with open(fpath) as fh:\n",
    "            results_dict = json.load(fh) \n",
    "        f = os.path.basename(fpath)\n",
    "        f = f.replace(\".json\", \"\")\n",
    "        \n",
    "        if len(f.split(\"_\")) == 10:\n",
    "            dataset_name, emb_decomposition, emb_dimension, emb_num_epochs, model_name, model_num_epochs, _, model_batch_size, train_size, random_seed = f.split(\"_\")\n",
    "        else:\n",
    "            dataset_name, model_name, model_num_epochs, _, model_batch_size, train_size, random_seed = f.split(\"_\")\n",
    "            emb_decomposition, emb_dimension, emb_num_epochs = 0,0,0\n",
    "            \n",
    "        # Reset Types\n",
    "        dataset_name = dataset_name\n",
    "        emb_decomposition = emb_decomposition\n",
    "        emb_dimension = int(emb_dimension)\n",
    "        emb_num_epochs = int(emb_num_epochs)\n",
    "        model_name = model_name\n",
    "        model_num_epochs = int(model_num_epochs)\n",
    "        model_batch_size = int(model_batch_size)\n",
    "        train_size = float(train_size)\n",
    "        random_seed = int(random_seed)\n",
    "        \n",
    "        results_auc = results_dict['evaluation']['roc_auc']\n",
    "        results_losses = np.array(results_dict['losses'])\n",
    "    \n",
    "        # Populate table\n",
    "        row = [dataset_name, emb_decomposition, emb_dimension, emb_num_epochs, model_name, model_num_epochs, model_batch_size, train_size, random_seed, results_auc, results_losses]\n",
    "        df_rows.append(row)\n",
    "    \n",
    "    df = pd.DataFrame(df_rows, columns=df_columns)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed3817a6",
   "metadata": {},
   "source": [
    "# Ablation Study\n",
    "\n",
    "Here we perform an ablation study over the two hyperparameters introduced in our distributed representations\n",
    "1. User specified dimensionality of distributed representations\n",
    "2. User specified number of training epochs for the skipgram model\n",
    "\n",
    "Here we explore the effects of varying these values on the downstream performance"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6690b9e1",
   "metadata": {},
   "source": [
    "## Number of dimensions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ff2de266",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Study hyperparameters\n",
    "embdims = [8, 16, 32, 64, 128, 256, 512, 1024]\n",
    "embepochs = 1000\n",
    "model_num_epochs = 250\n",
    "model_name = \"DeepDRSynergy\"\n",
    "decompositions = [\"wl3\", \"sp\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0aed9b04",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the table of results with all datasets, sp/wl3 decompositions, deepsynergydr\n",
    "df = process_results2pandas_nodataset(\"results\")\n",
    "df = df[((df['emb_dimension'].isin(embdims)) | (df['emb_dimension']==0)) & ((df['emb_num_epochs']==embepochs) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "df = df[(df['model_name'] == model_name) & (df['emb_decomposition'].isin(decompositions))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "9f906e30",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2270.21x500 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(font_scale=1.5)\n",
    "g = sns.relplot(data = df, kind=\"line\", \n",
    "                x=\"emb_dimension\", \n",
    "                y=\"results_auc\", \n",
    "                col=\"dataset_name\", \n",
    "                col_order=[\"drugcombdb\", \"drugcomb\", \"drugbankddi\", \"twosides\"], \n",
    "                hue=\"emb_decomposition\",\n",
    "                hue_order=[\"wl3\", \"sp\"])\n",
    "# ax0 = g.axes[0][0]\n",
    "# ax0.axhline(0.933, ls=\"--\", c=\"r\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3ab9d1c0",
   "metadata": {},
   "source": [
    "### Individual dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "29933e1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the table of results with all datasets, sp/wl3 decompositions, deepsynergydr\n",
    "dataset = \"drugcomb\"\n",
    "df = process_results2pandas(\"results\", dataset)\n",
    "df = df[((df['emb_dimension'].isin(embdims)) | (df['emb_dimension']==0)) & ((df['emb_num_epochs']==embepochs) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "df = df[(df['model_name'] == model_name) & (df['emb_decomposition'].isin(decompositions))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e1f5170f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson r: -0.5732417999049775\n",
      "p-value: 1.692873790498619e-06\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import pearsonr\n",
    "x = df['emb_dimension']\n",
    "y = df['results_auc']\n",
    "r, p = pearsonr(x,y)\n",
    "print(f\"Pearson r: {r}\")\n",
    "print(f\"p-value: {p}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1b580232",
   "metadata": {},
   "source": [
    "## Number of training epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "deb78564",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Study hyperparameters\n",
    "embdim = 64\n",
    "embepochs = [200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800, 2000]\n",
    "model_num_epochs = 250\n",
    "model_name = \"DeepDRSynergy\"\n",
    "decompositions = [\"wl3\", \"sp\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d9bf3da8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the table of results with all datasets, sp/wl3 decompositions, deepsynergydr, and embedding dimension==64\n",
    "df = process_results2pandas_nodataset(\"results\")\n",
    "df = df[((df['emb_dimension']==embdim) | (df['emb_dimension']==0)) & ((df['emb_num_epochs'].isin(embepochs)) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "df = df[(df['model_name'] == model_name) & (df['emb_decomposition'].isin(decompositions))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ff60fbff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2270.21x500 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.set(font_scale=1.5)\n",
    "g = sns.relplot(data = df, kind=\"line\", \n",
    "                x=\"emb_num_epochs\", \n",
    "                y=\"results_auc\", \n",
    "                col=\"dataset_name\", \n",
    "                col_order=[\"drugcombdb\", \"drugcomb\", \"drugbankddi\", \"twosides\"], \n",
    "                hue=\"emb_decomposition\",\n",
    "                hue_order=[\"wl3\", \"sp\"])\n",
    "# ax0 = g.axes[0][0]\n",
    "# ax0.axhline(0.933, ls=\"--\", c=\"r\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48eaccd5",
   "metadata": {},
   "source": [
    "### Individual dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "3240b7e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = \"drugbankddi\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e3993003",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the table of results with specified dataset, sp/wl3 decompositions, deepsynergydr, and embedding dimension==64\n",
    "df = process_results2pandas(\"results\", dataset)\n",
    "df = df[((df['emb_dimension']==embdim) | (df['emb_dimension']==0)) & ((df['emb_num_epochs'].isin(embepochs)) | (df['emb_num_epochs']==0)) & (df['model_num_epochs']==model_num_epochs)]\n",
    "df = df[(df['model_name'] == model_name) & (df['emb_decomposition'].isin(decompositions))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d7a4eabd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pearson r: 0.1966647081148902\n",
      "p-value: 0.049865196582319186\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import pearsonr\n",
    "x = df['emb_num_epochs']\n",
    "y = df['results_auc']\n",
    "r, p = pearsonr(x,y)\n",
    "print(f\"Pearson r: {r}\")\n",
    "print(f\"p-value: {p}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2be612d5",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b30a2992",
   "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": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}