learning_curve.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "99de795b",
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.optim as optim\n",
    "from nilearn.connectome import sym_matrix_to_vec\n",
    "from scipy.stats import pearsonr\n",
    "import torch.nn.functional as F\n",
    "import seaborn as sns\n",
    "from torch.utils.data import Dataset, DataLoader, Subset, TensorDataset\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import sys\n",
    "import pandas as pd\n",
    "import math\n",
    "from cmath import isinf\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.base import BaseEstimator\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.metrics import mean_absolute_percentage_error, r2_score\n",
    "import multiprocessing\n",
    "from sklearn.model_selection import KFold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d147112d",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c5aa9e47",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MLP(nn.Module):\n",
    "    def __init__(self, input_dim_feat = 499500, input_dim_target = 1, hidden_dim_feat = 1000, output_dim = 2, dropout_rate = 0):\n",
    "        super(MLP, self).__init__()\n",
    "        \n",
    "        # Xavier initialization for feature MLP\n",
    "        self.feat_mlp = nn.Sequential(\n",
    "            nn.BatchNorm1d(input_dim_feat),\n",
    "            nn.Linear(input_dim_feat, hidden_dim_feat),\n",
    "            nn.ReLU(),\n",
    "            nn.Dropout(p=dropout_rate),\n",
    "            nn.Linear(hidden_dim_feat, output_dim)\n",
    "        )\n",
    "        self.init_weights(self.feat_mlp)\n",
    "\n",
    "        # Xavier initialization for target MLP\n",
    "        self.target_mlp = nn.Sequential(\n",
    "            nn.BatchNorm1d(input_dim_target),\n",
    "            nn.Linear(input_dim_target, output_dim)\n",
    "        )\n",
    "        self.init_weights(self.target_mlp)\n",
    "        \n",
    "    def init_weights(self, m):\n",
    "        if isinstance(m, nn.Linear):\n",
    "            nn.init.xavier_uniform_(m.weight)\n",
    "            nn.init.constant_(m.bias, 0.0)\n",
    "        \n",
    "    def forward(self, x, y):\n",
    "        features = self.feat_mlp(x)\n",
    "        targets = self.target_mlp(y)\n",
    "        features = nn.functional.normalize(features, p=2, dim=1)\n",
    "        targets = nn.functional.normalize(targets, p=2, dim=1)\n",
    "        return features, targets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "93673abb",
   "metadata": {},
   "outputs": [],
   "source": [
    "@torch.no_grad()\n",
    "def gather_feats_targets(model, dataloader, device):\n",
    "    features = []\n",
    "    targets = []\n",
    "\n",
    "    model.eval()\n",
    "    for idx, (feat, target) in enumerate(dataloader):\n",
    "        if len(feat.shape) > 2:\n",
    "            feat = feat[:, 0, :]\n",
    "        feat = feat.to(device)\n",
    "        target = target.to(device)\n",
    "        out_feat, out_target= model(feat, target)\n",
    "        features.append(out_feat)\n",
    "        targets.append(target[:, 0])\n",
    "    \n",
    "    return torch.cat(features, 0).cpu().numpy(), torch.cat(targets, 0).cpu().numpy()\n",
    "\n",
    "class AgeEstimator(BaseEstimator):\n",
    "    \"\"\" Define the age estimator on latent space network features.\n",
    "    \"\"\"\n",
    "    def __init__(self):\n",
    "        n_jobs = multiprocessing.cpu_count()\n",
    "        self.age_estimator = GridSearchCV(\n",
    "            Ridge(), param_grid={\"alpha\": 10.**np.arange(-2, 3)}, cv=5,\n",
    "            scoring=\"r2\", n_jobs=n_jobs)\n",
    "\n",
    "    def fit(self, X, y):\n",
    "        self.age_estimator.fit(X, y)\n",
    "        return self.score(X, y), self.r2(X, y)\n",
    "\n",
    "    def predict(self, X):\n",
    "        y_pred = self.age_estimator.predict(X)\n",
    "        return y_pred\n",
    "    \n",
    "    def score(self, X, y):\n",
    "        y_pred = self.age_estimator.predict(X)\n",
    "        return mean_absolute_percentage_error(y, y_pred)\n",
    "    \n",
    "    def r2(self, X, y):\n",
    "        y_pred = self.age_estimator.predict(X)\n",
    "        return r2_score(y, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "5caa4aae",
   "metadata": {},
   "outputs": [],
   "source": [
    "class MatData(Dataset):\n",
    "    def __init__(self, path_feat, path_target, target_name, indices, transform = None, regions = None, threshold_mat = False, threshold_percent = None, random_state=42):\n",
    "        \"\"\"\n",
    "        Initializes the dataset with the capability to handle training and testing splits, \n",
    "        including multiple views for augmented data.\n",
    "        \n",
    "        Args:\n",
    "            path_feat (str): Path to the features file.\n",
    "            path_target (str): Path to the target file.\n",
    "            transform (callable): A transformation function to apply for augmentation.\n",
    "            train (bool): Whether the dataset is used for training. False will load the test set.\n",
    "            test_size (float): Proportion of the dataset to include in the test split.\n",
    "            random_state (int): Random state for reproducible train-test splits.\n",
    "        \"\"\"\n",
    "        \n",
    "        features = np.load(path_feat)[indices]\n",
    "        participant_data = pd.read_csv(path_target)\n",
    "        targets = participant_data[target_name].values[indices]\n",
    "        targets = np.expand_dims(targets, axis = 1)\n",
    "        \n",
    "        if threshold_mat:\n",
    "            thresholded_feat = []\n",
    "            for matrix in features:\n",
    "                threshold = np.percentile(matrix, threshold_percent)\n",
    "                matrix[matrix < threshold] = 0\n",
    "                thresholded_feat.append(matrix)\n",
    "            threshold_feat = np.stack(thresholded_feat)\n",
    "            features = threshold_feat\n",
    "        \n",
    "\n",
    "        self.n_sub = len(features)\n",
    "        self.n_views = 1\n",
    "        self.transform = transform\n",
    "        self.targets = targets\n",
    "        \n",
    "        vectorized_feat = np.array([sym_matrix_to_vec(mat, discard_diagonal=True) for mat in features])\n",
    "        self.n_features = vectorized_feat.shape[-1]\n",
    "        \n",
    "        if transform is not None:\n",
    "            # augmentation only in training mode\n",
    "            if transform != \"copy\":\n",
    "                augmented_features = np.array([self.transform(sample, regions = regions) for sample in features])\n",
    "\n",
    "                self.n_views = self.n_views + augmented_features.shape[1]\n",
    "                self.features = np.zeros((self.n_sub, self.n_views, self.n_features))\n",
    "                for sub in range(self.n_sub):\n",
    "                    self.features[sub, 0, :] = vectorized_feat[sub]\n",
    "                    self.features[sub, 1:, :] = augmented_features[sub]\n",
    "            else:\n",
    "                self.features = np.repeat(np.expand_dims(vectorized_feat, axis = 1), 2, axis=1)\n",
    "        else:\n",
    "            self.features = np.expand_dims(vectorized_feat, axis = 1)\n",
    "            \n",
    "    def __len__(self):\n",
    "        return self.n_sub\n",
    "\n",
    "    def __getitem__(self, idx):\n",
    "        features = self.features[idx]\n",
    "        targets = self.targets[idx]\n",
    "        features = torch.from_numpy(features).float()\n",
    "        targets = torch.from_numpy(targets).float()\n",
    "        \n",
    "        return features, targets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7aeeb48a",
   "metadata": {},
   "outputs": [],
   "source": [
    "def gaussian_kernel(x, krnl_sigma):\n",
    "    x = x - x.T\n",
    "    return torch.exp(-(x**2) / (2*(krnl_sigma**2))) / (math.sqrt(2*torch.pi)*krnl_sigma)\n",
    "\n",
    "def cauchy(x, krnl_sigma):\n",
    "        x = x - x.T\n",
    "        return  1. / (krnl_sigma*(x**2) + 1)\n",
    "\n",
    "def rbf(x, krnl_sigma):\n",
    "        x = x - x.T\n",
    "        return torch.exp(-(x**2)/(2*(krnl_sigma**2)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0f739858",
   "metadata": {},
   "outputs": [],
   "source": [
    "# loss from: https://github.com/EIDOSLAB/contrastive-brain-age-prediction/blob/master/src/losses.py\n",
    "# modified to accept input shape [bsz, n_feats]. In the age paper: [bsz, n_views, n_feats].\n",
    "class KernelizedSupCon(nn.Module):\n",
    "    \"\"\"Supervised contrastive loss: https://arxiv.org/pdf/2004.11362.pdf.\n",
    "    It also supports the unsupervised contrastive loss in SimCLR\n",
    "    Based on: https://github.com/HobbitLong/SupContrast\"\"\"\n",
    "    def __init__(self, method: str='expw', temperature: float=0.03, contrast_mode: str='all',\n",
    "                 base_temperature: float=0.07, krnl_sigma: float = 1., kernel: callable=cauchy, delta_reduction: str='sum'):\n",
    "        super().__init__()\n",
    "        self.temperature = temperature\n",
    "        self.contrast_mode = contrast_mode\n",
    "        self.base_temperature = base_temperature\n",
    "        self.method = method\n",
    "        self.kernel = kernel\n",
    "        self.krnl_sigma = krnl_sigma\n",
    "        self.delta_reduction = delta_reduction\n",
    "\n",
    "        if kernel is not None and method == 'supcon':\n",
    "            raise ValueError('Kernel must be none if method=supcon')\n",
    "        \n",
    "        if kernel is None and method != 'supcon':\n",
    "            raise ValueError('Kernel must not be none if method != supcon')\n",
    "\n",
    "        if delta_reduction not in ['mean', 'sum']:\n",
    "            raise ValueError(f\"Invalid reduction {delta_reduction}\")\n",
    "\n",
    "    def __repr__(self):\n",
    "        return f'{self.__class__.__name__} ' \\\n",
    "               f'(t={self.temperature}, ' \\\n",
    "               f'method={self.method}, ' \\\n",
    "               f'kernel={self.kernel is not None}, ' \\\n",
    "               f'delta_reduction={self.delta_reduction})'\n",
    "\n",
    "    def forward(self, features, labels=None):\n",
    "        \"\"\"Compute loss for model. If `labels` is None, \n",
    "        it degenerates to SimCLR unsupervised loss:\n",
    "        https://arxiv.org/pdf/2002.05709.pdf\n",
    "\n",
    "        Args:\n",
    "            features: hidden vector of shape [bsz, n_views, n_features]. \n",
    "                input has to be rearranged to [bsz, n_views, n_features] and labels [bsz],\n",
    "            labels: ground truth of shape [bsz].\n",
    "        Returns:\n",
    "            A loss scalar.\n",
    "        \"\"\"\n",
    "        device = features.device\n",
    "\n",
    "        if len(features.shape) != 3:\n",
    "            raise ValueError('`features` needs to be [bsz, n_views, n_feats],'\n",
    "                             '3 dimensions are required')\n",
    "\n",
    "        batch_size = features.shape[0]\n",
    "        contrast_count = features.shape[1]\n",
    "\n",
    "        if labels is None:\n",
    "            mask = torch.eye(batch_size, device=device)\n",
    "        \n",
    "        else:\n",
    "            labels = labels.view(-1, 1)\n",
    "            if labels.shape[0] != batch_size:\n",
    "                raise ValueError('Num of labels does not match num of features')\n",
    "            \n",
    "            if self.kernel is None:\n",
    "                mask = torch.eq(labels, labels.T)\n",
    "            else:\n",
    "                mask = self.kernel(labels, krnl_sigma = self.krnl_sigma)     \n",
    "        \n",
    "        contrast_feature = torch.cat(torch.unbind(features, dim=1), dim=0)\n",
    "        if self.contrast_mode == 'one':\n",
    "            anchor_feature = features[:, 0]\n",
    "            anchor_count = 1\n",
    "        elif self.contrast_mode == 'all':\n",
    "            anchor_feature = contrast_feature\n",
    "            anchor_count = contrast_count\n",
    "        else:\n",
    "            raise ValueError('Unknown mode: {}'.format(self.contrast_mode))\n",
    "\n",
    "        # Tile mask\n",
    "        mask = mask.repeat(anchor_count, contrast_count)\n",
    "        # Inverse of torch-eye to remove self-contrast (diagonal)\n",
    "        inv_diagonal = torch.scatter(\n",
    "            torch.ones_like(mask),\n",
    "            1,\n",
    "            torch.arange(batch_size*anchor_count, device=device).view(-1, 1),\n",
    "            0\n",
    "        )\n",
    "        # compute similarity\n",
    "        anchor_dot_contrast = torch.div(\n",
    "            torch.matmul(anchor_feature, contrast_feature.T),\n",
    "            self.temperature\n",
    "        )\n",
    "\n",
    "        # for numerical stability\n",
    "        logits_max, _ = torch.max(anchor_dot_contrast, dim=1, keepdim=True)\n",
    "        logits = anchor_dot_contrast - logits_max.detach()\n",
    "\n",
    "        alignment = logits \n",
    "\n",
    "        # base case is:\n",
    "        # - supcon if kernel = none \n",
    "        # - y-aware is kernel != none\n",
    "        uniformity = torch.exp(logits) * inv_diagonal \n",
    "\n",
    "        if self.method == 'threshold':\n",
    "            repeated = mask.unsqueeze(-1).repeat(1, 1, mask.shape[0]) # repeat kernel mask\n",
    "\n",
    "            delta = (mask[:, None].T - repeated.T).transpose(1, 2) # compute the difference w_k - w_j for every k,j\n",
    "            delta = (delta > 0.).float()\n",
    "\n",
    "            # for each z_i, repel only samples j s.t. K(z_i, z_j) < K(z_i, z_k)\n",
    "            uniformity = uniformity.unsqueeze(-1).repeat(1, 1, mask.shape[0])\n",
    "\n",
    "            if self.delta_reduction == 'mean':\n",
    "                uniformity = (uniformity * delta).mean(-1)\n",
    "            else:\n",
    "                uniformity = (uniformity * delta).sum(-1)\n",
    "    \n",
    "        elif self.method == 'expw':\n",
    "            # exp weight e^(s_j(1-w_j))\n",
    "            uniformity = torch.exp(logits * (1 - mask)) * inv_diagonal\n",
    "\n",
    "        uniformity = torch.log(uniformity.sum(1, keepdim=True))\n",
    "\n",
    "\n",
    "        # positive mask contains the anchor-positive pairs\n",
    "        # excluding <self,self> on the diagonal\n",
    "        positive_mask = mask * inv_diagonal\n",
    "\n",
    "        log_prob = alignment - uniformity # log(alignment/uniformity) = log(alignment) - log(uniformity)\n",
    "        log_prob = (positive_mask * log_prob).sum(1) / positive_mask.sum(1) # compute mean of log-likelihood over positive\n",
    " \n",
    "        # loss\n",
    "        loss = - (self.temperature / self.base_temperature) * log_prob\n",
    "        return loss.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c35b8b4d",
   "metadata": {},
   "outputs": [],
   "source": [
    "participants = pd.read_csv(\"participants.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a7fc00ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "def train(model, train_loader, criterion_pft, criterion_ptt, optimizer):\n",
    "    \n",
    "    model.train()\n",
    "    batch_losses = []\n",
    "    for batch_num, (features, targets) in enumerate(train_loader):\n",
    "        bsz = targets.shape[0]\n",
    "        n_views = features.shape[1]\n",
    "        n_feat = features.shape[-1]\n",
    "        \n",
    "        features = features.view(bsz * n_views, n_feat) # [bsz*2, 499500]\n",
    "        features, targets = features.to(device), targets.to(device) # [bsz, 2, 499500], [bsz, 1]\n",
    "        \n",
    "        optimizer.zero_grad()\n",
    "        out_feat, out_target = model(features, torch.cat(n_views*[targets], dim=0)) # ([bsz*5, 1], [bsz*5, 1])\n",
    "        \n",
    "        out_feat = torch.split(out_feat, [bsz]*n_views, dim=0)\n",
    "        out_feat = torch.cat([f.unsqueeze(1) for f in out_feat], dim=1) # [bsz, 5, 2]\n",
    "        \n",
    "        loss = criterion_pft(out_feat, targets) # ([bsz, 5, 2], [bsz, 1])\n",
    "        \n",
    "        out_target = torch.split(out_target, [bsz]*n_views, dim=0)\n",
    "        out_target = torch.cat([f.unsqueeze(1) for f in out_target], dim=1) # [bsz, 2, 2]\n",
    "        loss += criterion_ptt(out_target, targets) # ([bsz, 5, 2], [bsz, 1])\n",
    "        loss += torch.nn.functional.mse_loss(out_feat.view(bsz * n_views, 2), out_target.view(bsz * n_views, 2)) # mse_loss([bsz*2, 2], [bsz*2, 2])\n",
    "    \n",
    "        loss.backward()\n",
    "        # torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)\n",
    "        batch_losses.append(loss.item())\n",
    "        optimizer.step()\n",
    "    mean_loss = sum(batch_losses)/len(batch_losses)\n",
    "    print(f'Epoch {epoch} | Mean Loss {mean_loss}')\n",
    "        \n",
    "    return mean_loss\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "5524c57b",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_sizes = [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.95]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7f782af8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# val_indices = participants.index[participants['dataset'] == \"COBRE\"].values\n",
    "n_subj = len(participants)\n",
    "batch_size = 30\n",
    "\n",
    "input_dim_feat = 499500 # vectorized mat, diagonal discarded\n",
    "# the rest is arbitrary\n",
    "hidden_dim_feat = 1000\n",
    "input_dim_target = 1\n",
    "output_dim = 2\n",
    "\n",
    "lr = 0.01\n",
    "dropout_rate = 0\n",
    "weight_decay = 0\n",
    "kernel = cauchy\n",
    "epochs = 100\n",
    "\n",
    "criterion_pft = KernelizedSupCon(method='expw', temperature=0.03, base_temperature=0.07, kernel=kernel, krnl_sigma = 1, contrast_mode = 'all')\n",
    "criterion_ptt = KernelizedSupCon(method='expw', temperature=0.03, base_temperature=0.07, kernel=kernel, krnl_sigma = 1, contrast_mode = 'all')\n",
    "\n",
    "n_splits = 5\n",
    "kf = KFold(n_splits=n_splits, shuffle=True, random_state=42)\n",
    "\n",
    "results_series = []\n",
    "all_indices = np.arange(n_subj)\n",
    "\n",
    "for train_size in train_sizes:\n",
    "    \n",
    "    print(\"Train size: \", train_size)\n",
    "    num_train = int(n_subj * train_size)\n",
    "    train_indices = np.random.choice(all_indices, num_train, replace=False)\n",
    "    fold_num = 0\n",
    "    \n",
    "    for train_idx, test_idx in kf.split(train_indices):\n",
    "        \n",
    "        training_fold_indices = train_indices[train_idx]\n",
    "        validation_fold_indices = train_indices[test_idx]\n",
    "\n",
    "        train_dataset = MatData(\"matrices.npy\", \"participants.csv\", \"age\", indices=training_fold_indices)\n",
    "        train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)\n",
    "\n",
    "        validation_dataset = MatData(\"matrices.npy\", \"participants.csv\", \"age\", indices=validation_fold_indices)\n",
    "        val_loader = DataLoader(validation_dataset, batch_size=batch_size, shuffle=False)\n",
    "\n",
    "        model = MLP(input_dim_feat, input_dim_target, hidden_dim_feat, output_dim, dropout_rate).to(device)\n",
    "        optimizer = optim.Adam(model.parameters(), lr=lr, weight_decay=weight_decay)\n",
    "        for epoch in range(epochs):\n",
    "            train(model, train_loader, criterion_pft, criterion_ptt, optimizer)\n",
    "\n",
    "        # Evaluate the model on both training and testing sets\n",
    "        X_train, y_train = gather_feats_targets(model, train_loader, device)\n",
    "        estimator = AgeEstimator()\n",
    "        estimator.fit(X_train, y_train)\n",
    "        mae_train = estimator.score(X_train, y_train)\n",
    "        r2_train = estimator.r2(X_train, y_train)\n",
    "        \n",
    "        X_test, y_test = gather_feats_targets(model, val_loader, device)\n",
    "        mae_test = estimator.score(X_test, y_test)\n",
    "        r2_test = estimator.r2(X_test, y_test)\n",
    "        \n",
    "        # Record results for this fold\n",
    "        results_series.append(pd.Series({\n",
    "            \"Training Size\": num_train,\n",
    "            \"Fold Number\": int(fold_num),\n",
    "            \"MAE Train\": mae_train,\n",
    "            \"R2 Train\": r2_train,\n",
    "            \"MAE Test\": mae_test,\n",
    "            \"R2 Test\": r2_test}))\n",
    "        fold_num += 1\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "45110091",
   "metadata": {},
   "outputs": [],
   "source": [
    "results = pd.DataFrame(results_series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "d5fd3bb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "results.to_csv(\"results/cv_results.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e3882589",
   "metadata": {},
   "outputs": [],
   "source": [
    "results=pd.read_csv(\"results/cv_results.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "273100da",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>Training Size</th>\n",
       "      <th>Fold Number</th>\n",
       "      <th>MAE Train</th>\n",
       "      <th>R2 Train</th>\n",
       "      <th>MAE Test</th>\n",
       "      <th>R2 Test</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.320616</td>\n",
       "      <td>0.257167</td>\n",
       "      <td>0.306640</td>\n",
       "      <td>0.176715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>93.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.272307</td>\n",
       "      <td>0.269995</td>\n",
       "      <td>0.314143</td>\n",
       "      <td>-0.035996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>93.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.338702</td>\n",
       "      <td>0.000002</td>\n",
       "      <td>0.365380</td>\n",
       "      <td>-0.017479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>93.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.331420</td>\n",
       "      <td>0.036329</td>\n",
       "      <td>0.326681</td>\n",
       "      <td>0.099932</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>93.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.246667</td>\n",
       "      <td>0.369237</td>\n",
       "      <td>0.330757</td>\n",
       "      <td>0.222532</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>187.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.288549</td>\n",
       "      <td>0.199803</td>\n",
       "      <td>0.327753</td>\n",
       "      <td>0.046350</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>187.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.247173</td>\n",
       "      <td>0.332201</td>\n",
       "      <td>0.305040</td>\n",
       "      <td>0.060790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>7</td>\n",
       "      <td>187.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.266749</td>\n",
       "      <td>0.343352</td>\n",
       "      <td>0.344986</td>\n",
       "      <td>0.071728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>8</td>\n",
       "      <td>187.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.255858</td>\n",
       "      <td>0.334699</td>\n",
       "      <td>0.238970</td>\n",
       "      <td>0.186587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9</td>\n",
       "      <td>187.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.232313</td>\n",
       "      <td>0.447195</td>\n",
       "      <td>0.293131</td>\n",
       "      <td>0.152676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>10</td>\n",
       "      <td>280.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.246752</td>\n",
       "      <td>0.357724</td>\n",
       "      <td>0.296114</td>\n",
       "      <td>0.115880</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11</td>\n",
       "      <td>280.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.146853</td>\n",
       "      <td>0.732027</td>\n",
       "      <td>0.272262</td>\n",
       "      <td>0.313229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>12</td>\n",
       "      <td>280.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.283317</td>\n",
       "      <td>0.266801</td>\n",
       "      <td>0.268642</td>\n",
       "      <td>0.070704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>13</td>\n",
       "      <td>280.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.163439</td>\n",
       "      <td>0.676533</td>\n",
       "      <td>0.312607</td>\n",
       "      <td>0.047244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>14</td>\n",
       "      <td>280.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.285416</td>\n",
       "      <td>0.225218</td>\n",
       "      <td>0.374150</td>\n",
       "      <td>0.025030</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>15</td>\n",
       "      <td>374.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.209028</td>\n",
       "      <td>0.453858</td>\n",
       "      <td>0.291305</td>\n",
       "      <td>0.075561</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>16</td>\n",
       "      <td>374.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.276071</td>\n",
       "      <td>0.188798</td>\n",
       "      <td>0.332655</td>\n",
       "      <td>0.006045</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>17</td>\n",
       "      <td>374.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.196012</td>\n",
       "      <td>0.550216</td>\n",
       "      <td>0.275667</td>\n",
       "      <td>-0.027600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>18</td>\n",
       "      <td>374.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.266964</td>\n",
       "      <td>0.206277</td>\n",
       "      <td>0.309192</td>\n",
       "      <td>0.210061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>19</td>\n",
       "      <td>374.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.201341</td>\n",
       "      <td>0.554121</td>\n",
       "      <td>0.294132</td>\n",
       "      <td>-0.029508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>20</td>\n",
       "      <td>468.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.195929</td>\n",
       "      <td>0.505132</td>\n",
       "      <td>0.291182</td>\n",
       "      <td>0.238229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>21</td>\n",
       "      <td>468.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.256861</td>\n",
       "      <td>0.303977</td>\n",
       "      <td>0.294508</td>\n",
       "      <td>0.117936</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>22</td>\n",
       "      <td>468.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.111718</td>\n",
       "      <td>0.834775</td>\n",
       "      <td>0.265768</td>\n",
       "      <td>0.351548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>23</td>\n",
       "      <td>468.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.107329</td>\n",
       "      <td>0.803369</td>\n",
       "      <td>0.225594</td>\n",
       "      <td>0.294930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>24</td>\n",
       "      <td>468.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.049898</td>\n",
       "      <td>0.954041</td>\n",
       "      <td>0.215063</td>\n",
       "      <td>0.237799</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>25</td>\n",
       "      <td>561.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.042576</td>\n",
       "      <td>0.906988</td>\n",
       "      <td>0.235771</td>\n",
       "      <td>0.385387</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>26</td>\n",
       "      <td>561.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.045130</td>\n",
       "      <td>0.941790</td>\n",
       "      <td>0.236892</td>\n",
       "      <td>0.323066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>27</td>\n",
       "      <td>561.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.117616</td>\n",
       "      <td>0.760493</td>\n",
       "      <td>0.264699</td>\n",
       "      <td>0.329300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>28</td>\n",
       "      <td>561.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.055084</td>\n",
       "      <td>0.920741</td>\n",
       "      <td>0.221918</td>\n",
       "      <td>0.061751</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>29</td>\n",
       "      <td>561.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.157767</td>\n",
       "      <td>0.703740</td>\n",
       "      <td>0.250467</td>\n",
       "      <td>0.270442</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>30</td>\n",
       "      <td>655.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.068730</td>\n",
       "      <td>0.910643</td>\n",
       "      <td>0.260087</td>\n",
       "      <td>0.223365</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>31</td>\n",
       "      <td>655.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.236169</td>\n",
       "      <td>0.487249</td>\n",
       "      <td>0.245745</td>\n",
       "      <td>0.399340</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>32</td>\n",
       "      <td>655.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.067553</td>\n",
       "      <td>0.896034</td>\n",
       "      <td>0.239853</td>\n",
       "      <td>0.436043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>33</td>\n",
       "      <td>655.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.042245</td>\n",
       "      <td>0.946094</td>\n",
       "      <td>0.199331</td>\n",
       "      <td>0.508540</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>34</td>\n",
       "      <td>655.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.056842</td>\n",
       "      <td>0.884490</td>\n",
       "      <td>0.235411</td>\n",
       "      <td>0.326064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>35</td>\n",
       "      <td>748.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.035766</td>\n",
       "      <td>0.951238</td>\n",
       "      <td>0.197218</td>\n",
       "      <td>0.498388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>36</td>\n",
       "      <td>748.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.042459</td>\n",
       "      <td>0.952528</td>\n",
       "      <td>0.211609</td>\n",
       "      <td>0.367515</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>37</td>\n",
       "      <td>748.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.063567</td>\n",
       "      <td>0.922948</td>\n",
       "      <td>0.194133</td>\n",
       "      <td>0.504194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>38</td>\n",
       "      <td>748.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.124306</td>\n",
       "      <td>0.753209</td>\n",
       "      <td>0.234410</td>\n",
       "      <td>0.352019</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>39</td>\n",
       "      <td>748.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.036939</td>\n",
       "      <td>0.924078</td>\n",
       "      <td>0.205247</td>\n",
       "      <td>0.418990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>40</td>\n",
       "      <td>842.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.042203</td>\n",
       "      <td>0.921349</td>\n",
       "      <td>0.206384</td>\n",
       "      <td>0.481047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>41</td>\n",
       "      <td>842.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.039241</td>\n",
       "      <td>0.913930</td>\n",
       "      <td>0.204517</td>\n",
       "      <td>0.426320</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>42</td>\n",
       "      <td>842.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.050458</td>\n",
       "      <td>0.937927</td>\n",
       "      <td>0.199849</td>\n",
       "      <td>0.500108</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>43</td>\n",
       "      <td>842.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.050027</td>\n",
       "      <td>0.935052</td>\n",
       "      <td>0.187075</td>\n",
       "      <td>0.577189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>44</td>\n",
       "      <td>842.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.039057</td>\n",
       "      <td>0.908072</td>\n",
       "      <td>0.201204</td>\n",
       "      <td>0.468297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>45</td>\n",
       "      <td>889.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.047288</td>\n",
       "      <td>0.937546</td>\n",
       "      <td>0.206106</td>\n",
       "      <td>0.433290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>46</td>\n",
       "      <td>889.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.036287</td>\n",
       "      <td>0.943166</td>\n",
       "      <td>0.207399</td>\n",
       "      <td>0.546055</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>47</td>\n",
       "      <td>889.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.039207</td>\n",
       "      <td>0.946397</td>\n",
       "      <td>0.201768</td>\n",
       "      <td>0.536643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>48</td>\n",
       "      <td>889.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.040360</td>\n",
       "      <td>0.901810</td>\n",
       "      <td>0.173215</td>\n",
       "      <td>0.533037</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>49</td>\n",
       "      <td>889.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.039425</td>\n",
       "      <td>0.922188</td>\n",
       "      <td>0.196117</td>\n",
       "      <td>0.541555</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Unnamed: 0  Training Size  Fold Number  MAE Train  R2 Train  MAE Test  \\\n",
       "0            0           93.0          0.0   0.320616  0.257167  0.306640   \n",
       "1            1           93.0          1.0   0.272307  0.269995  0.314143   \n",
       "2            2           93.0          2.0   0.338702  0.000002  0.365380   \n",
       "3            3           93.0          3.0   0.331420  0.036329  0.326681   \n",
       "4            4           93.0          4.0   0.246667  0.369237  0.330757   \n",
       "5            5          187.0          0.0   0.288549  0.199803  0.327753   \n",
       "6            6          187.0          1.0   0.247173  0.332201  0.305040   \n",
       "7            7          187.0          2.0   0.266749  0.343352  0.344986   \n",
       "8            8          187.0          3.0   0.255858  0.334699  0.238970   \n",
       "9            9          187.0          4.0   0.232313  0.447195  0.293131   \n",
       "10          10          280.0          0.0   0.246752  0.357724  0.296114   \n",
       "11          11          280.0          1.0   0.146853  0.732027  0.272262   \n",
       "12          12          280.0          2.0   0.283317  0.266801  0.268642   \n",
       "13          13          280.0          3.0   0.163439  0.676533  0.312607   \n",
       "14          14          280.0          4.0   0.285416  0.225218  0.374150   \n",
       "15          15          374.0          0.0   0.209028  0.453858  0.291305   \n",
       "16          16          374.0          1.0   0.276071  0.188798  0.332655   \n",
       "17          17          374.0          2.0   0.196012  0.550216  0.275667   \n",
       "18          18          374.0          3.0   0.266964  0.206277  0.309192   \n",
       "19          19          374.0          4.0   0.201341  0.554121  0.294132   \n",
       "20          20          468.0          0.0   0.195929  0.505132  0.291182   \n",
       "21          21          468.0          1.0   0.256861  0.303977  0.294508   \n",
       "22          22          468.0          2.0   0.111718  0.834775  0.265768   \n",
       "23          23          468.0          3.0   0.107329  0.803369  0.225594   \n",
       "24          24          468.0          4.0   0.049898  0.954041  0.215063   \n",
       "25          25          561.0          0.0   0.042576  0.906988  0.235771   \n",
       "26          26          561.0          1.0   0.045130  0.941790  0.236892   \n",
       "27          27          561.0          2.0   0.117616  0.760493  0.264699   \n",
       "28          28          561.0          3.0   0.055084  0.920741  0.221918   \n",
       "29          29          561.0          4.0   0.157767  0.703740  0.250467   \n",
       "30          30          655.0          0.0   0.068730  0.910643  0.260087   \n",
       "31          31          655.0          1.0   0.236169  0.487249  0.245745   \n",
       "32          32          655.0          2.0   0.067553  0.896034  0.239853   \n",
       "33          33          655.0          3.0   0.042245  0.946094  0.199331   \n",
       "34          34          655.0          4.0   0.056842  0.884490  0.235411   \n",
       "35          35          748.0          0.0   0.035766  0.951238  0.197218   \n",
       "36          36          748.0          1.0   0.042459  0.952528  0.211609   \n",
       "37          37          748.0          2.0   0.063567  0.922948  0.194133   \n",
       "38          38          748.0          3.0   0.124306  0.753209  0.234410   \n",
       "39          39          748.0          4.0   0.036939  0.924078  0.205247   \n",
       "40          40          842.0          0.0   0.042203  0.921349  0.206384   \n",
       "41          41          842.0          1.0   0.039241  0.913930  0.204517   \n",
       "42          42          842.0          2.0   0.050458  0.937927  0.199849   \n",
       "43          43          842.0          3.0   0.050027  0.935052  0.187075   \n",
       "44          44          842.0          4.0   0.039057  0.908072  0.201204   \n",
       "45          45          889.0          0.0   0.047288  0.937546  0.206106   \n",
       "46          46          889.0          1.0   0.036287  0.943166  0.207399   \n",
       "47          47          889.0          2.0   0.039207  0.946397  0.201768   \n",
       "48          48          889.0          3.0   0.040360  0.901810  0.173215   \n",
       "49          49          889.0          4.0   0.039425  0.922188  0.196117   \n",
       "\n",
       "     R2 Test  \n",
       "0   0.176715  \n",
       "1  -0.035996  \n",
       "2  -0.017479  \n",
       "3   0.099932  \n",
       "4   0.222532  \n",
       "5   0.046350  \n",
       "6   0.060790  \n",
       "7   0.071728  \n",
       "8   0.186587  \n",
       "9   0.152676  \n",
       "10  0.115880  \n",
       "11  0.313229  \n",
       "12  0.070704  \n",
       "13  0.047244  \n",
       "14  0.025030  \n",
       "15  0.075561  \n",
       "16  0.006045  \n",
       "17 -0.027600  \n",
       "18  0.210061  \n",
       "19 -0.029508  \n",
       "20  0.238229  \n",
       "21  0.117936  \n",
       "22  0.351548  \n",
       "23  0.294930  \n",
       "24  0.237799  \n",
       "25  0.385387  \n",
       "26  0.323066  \n",
       "27  0.329300  \n",
       "28  0.061751  \n",
       "29  0.270442  \n",
       "30  0.223365  \n",
       "31  0.399340  \n",
       "32  0.436043  \n",
       "33  0.508540  \n",
       "34  0.326064  \n",
       "35  0.498388  \n",
       "36  0.367515  \n",
       "37  0.504194  \n",
       "38  0.352019  \n",
       "39  0.418990  \n",
       "40  0.481047  \n",
       "41  0.426320  \n",
       "42  0.500108  \n",
       "43  0.577189  \n",
       "44  0.468297  \n",
       "45  0.433290  \n",
       "46  0.546055  \n",
       "47  0.536643  \n",
       "48  0.533037  \n",
       "49  0.541555  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "d5569c24",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Plot the training MAE\n",
    "sns.lineplot(data=results, x=\"Training Size\", y=\"MAE Train\", label=\"MAE Train\", color=\"#0081a7\")\n",
    "\n",
    "# Plot the testing MAE\n",
    "sns.lineplot(data=results, x=\"Training Size\", y=\"MAE Test\", label=\"MAE Test\", color=\"#f07167\")\n",
    "\n",
    "# Adding some plot details\n",
    "plt.title(\"Learning curve: MAE\")\n",
    "plt.ylabel(\"Mean Absolute Error (MAE)\")\n",
    "plt.xlabel(\"Training Sizes\")\n",
    "plt.legend(title=\"Data Type\")\n",
    "\n",
    "# Show the plot\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "731bf3e2",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead\n",
      "  if pd.api.types.is_categorical_dtype(vector):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n",
      "/gpfs3/well/margulies/users/cpy397/.local/lib/python3.9/site-packages/seaborn/_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.\n",
      "  with pd.option_context('mode.use_inf_as_na', True):\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "\n",
    "# Plot the training MAE\n",
    "sns.lineplot(data=results, x=\"Training Size\", y=\"R2 Train\", label=\"R2 Train\", color=\"#0081a7\")\n",
    "\n",
    "# Plot the testing MAE\n",
    "sns.lineplot(data=results, x=\"Training Size\", y=\"R2 Test\", label=\"R2 Test\", color=\"#f07167\")\n",
    "\n",
    "# Adding some plot details\n",
    "plt.title(\"Learning curve: R2\")\n",
    "plt.ylabel(\"R2\")\n",
    "plt.xlabel(\"Training Sizes\")\n",
    "plt.legend(title=\"Data Type\")\n",
    "\n",
    "# Show the plot\n",
    "plt.show()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "66f5d96d",
   "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": "3.9.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}