tutorials/SDCD_basic_usage.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Stable Differentiable Causal Discovery (SDCD) Tutorial\n",
    "\n",
    "In this brief tutorial, we will go over the main API for the `sdcd` package. SDCD is a differentiable causal discovery method designed to scale reliably to hundreds and thousands of variables. The algorithm is implemented in vanilla PyTorch and also uses the `networkx` package. \n",
    "\n",
    "(Note, under the default pip installation, only the packages necessary to run the `SDCD` model are installed. If you would also like to run other causal discovery methods implemented in this package, you must also install the `benchmark` extra via the `pip install sdcd[benchmark]` command.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Imports\n",
    "from sdcd.models import SDCD\n",
    "from sdcd.utils import create_intervention_dataset"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will start by simulating some data for the tutorial. If you have your own data, you can ignore this step."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Simulate Data\n",
    "from sdcd.simulated_data import random_model_gaussian_global_variance # For demonstration\n",
    "\n",
    "n = 200\n",
    "n_per_intervention = 50\n",
    "d = 20\n",
    "n_edges = 20\n",
    "\n",
    "true_causal_model = random_model_gaussian_global_variance(\n",
    "    d,\n",
    "    n_edges,\n",
    "    dag_type=\"ER\",\n",
    "    scale=0.5,\n",
    "    hard=True,\n",
    ")\n",
    "X_df = true_causal_model.generate_dataframe_from_all_distributions(\n",
    "    n_samples_control=n,\n",
    "    n_samples_per_intervention=n_per_intervention,\n",
    ")\n",
    "X_df.iloc[:, :-1] = (X_df.iloc[:, :-1] - X_df.iloc[:, :-1].mean()) / X_df.iloc[\n",
    "    :, :-1\n",
    "].std() # Normalize the data"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The input data should be formatted as a Pandas Dataframe where each row corresponds to one observation and each column corresponds to one variable. There should be an additional column reports which variable(s) was intervened on for the given observation. Here it is labeled as `perturbation_label`. For rows that do not have any interventions, the value should be set to `\"obs\"`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>8</th>\n",
       "      <th>9</th>\n",
       "      <th>...</th>\n",
       "      <th>11</th>\n",
       "      <th>12</th>\n",
       "      <th>13</th>\n",
       "      <th>14</th>\n",
       "      <th>15</th>\n",
       "      <th>16</th>\n",
       "      <th>17</th>\n",
       "      <th>18</th>\n",
       "      <th>19</th>\n",
       "      <th>perturbation_label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1.042648</td>\n",
       "      <td>0.226783</td>\n",
       "      <td>-1.871494</td>\n",
       "      <td>-1.581519</td>\n",
       "      <td>2.149227</td>\n",
       "      <td>-0.458693</td>\n",
       "      <td>0.714175</td>\n",
       "      <td>3.091564</td>\n",
       "      <td>-0.509956</td>\n",
       "      <td>0.036024</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.486337</td>\n",
       "      <td>1.109146</td>\n",
       "      <td>1.863365</td>\n",
       "      <td>0.409856</td>\n",
       "      <td>0.858165</td>\n",
       "      <td>0.190376</td>\n",
       "      <td>0.407933</td>\n",
       "      <td>0.380345</td>\n",
       "      <td>0.605214</td>\n",
       "      <td>obs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-1.068825</td>\n",
       "      <td>0.559836</td>\n",
       "      <td>0.854967</td>\n",
       "      <td>0.205361</td>\n",
       "      <td>1.143803</td>\n",
       "      <td>0.896259</td>\n",
       "      <td>1.228846</td>\n",
       "      <td>-0.800072</td>\n",
       "      <td>-0.918887</td>\n",
       "      <td>-1.052896</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.304848</td>\n",
       "      <td>1.779307</td>\n",
       "      <td>-0.951121</td>\n",
       "      <td>-0.049833</td>\n",
       "      <td>-0.104780</td>\n",
       "      <td>0.651097</td>\n",
       "      <td>0.105443</td>\n",
       "      <td>0.042832</td>\n",
       "      <td>0.458566</td>\n",
       "      <td>obs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.178734</td>\n",
       "      <td>-0.770220</td>\n",
       "      <td>0.989042</td>\n",
       "      <td>-0.676152</td>\n",
       "      <td>-0.534547</td>\n",
       "      <td>-0.174431</td>\n",
       "      <td>1.071692</td>\n",
       "      <td>-1.091981</td>\n",
       "      <td>-0.098621</td>\n",
       "      <td>0.191706</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.111603</td>\n",
       "      <td>-0.645817</td>\n",
       "      <td>-1.211867</td>\n",
       "      <td>1.330255</td>\n",
       "      <td>-0.226251</td>\n",
       "      <td>0.530106</td>\n",
       "      <td>-0.267283</td>\n",
       "      <td>-0.685813</td>\n",
       "      <td>0.020803</td>\n",
       "      <td>obs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.359658</td>\n",
       "      <td>-1.081468</td>\n",
       "      <td>0.904916</td>\n",
       "      <td>0.500605</td>\n",
       "      <td>0.384816</td>\n",
       "      <td>-0.414948</td>\n",
       "      <td>-0.396487</td>\n",
       "      <td>-0.713363</td>\n",
       "      <td>-0.001175</td>\n",
       "      <td>-0.623004</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.637798</td>\n",
       "      <td>-0.967510</td>\n",
       "      <td>-0.544244</td>\n",
       "      <td>-0.324714</td>\n",
       "      <td>0.335550</td>\n",
       "      <td>0.454699</td>\n",
       "      <td>-0.320341</td>\n",
       "      <td>-0.104311</td>\n",
       "      <td>0.478039</td>\n",
       "      <td>obs</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.906305</td>\n",
       "      <td>-0.653814</td>\n",
       "      <td>-1.006582</td>\n",
       "      <td>-0.506965</td>\n",
       "      <td>1.263883</td>\n",
       "      <td>-1.640321</td>\n",
       "      <td>-0.273144</td>\n",
       "      <td>0.785179</td>\n",
       "      <td>-0.078615</td>\n",
       "      <td>-1.359076</td>\n",
       "      <td>...</td>\n",
       "      <td>1.403845</td>\n",
       "      <td>-0.450875</td>\n",
       "      <td>0.264944</td>\n",
       "      <td>-0.603409</td>\n",
       "      <td>1.056394</td>\n",
       "      <td>-1.742338</td>\n",
       "      <td>0.175205</td>\n",
       "      <td>0.388329</td>\n",
       "      <td>-0.720161</td>\n",
       "      <td>obs</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 21 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          0         1         2         3         4         5         6  \\\n",
       "0 -1.042648  0.226783 -1.871494 -1.581519  2.149227 -0.458693  0.714175   \n",
       "1 -1.068825  0.559836  0.854967  0.205361  1.143803  0.896259  1.228846   \n",
       "2 -0.178734 -0.770220  0.989042 -0.676152 -0.534547 -0.174431  1.071692   \n",
       "3 -0.359658 -1.081468  0.904916  0.500605  0.384816 -0.414948 -0.396487   \n",
       "4  0.906305 -0.653814 -1.006582 -0.506965  1.263883 -1.640321 -0.273144   \n",
       "\n",
       "          7         8         9  ...        11        12        13        14  \\\n",
       "0  3.091564 -0.509956  0.036024  ... -1.486337  1.109146  1.863365  0.409856   \n",
       "1 -0.800072 -0.918887 -1.052896  ... -1.304848  1.779307 -0.951121 -0.049833   \n",
       "2 -1.091981 -0.098621  0.191706  ... -0.111603 -0.645817 -1.211867  1.330255   \n",
       "3 -0.713363 -0.001175 -0.623004  ... -0.637798 -0.967510 -0.544244 -0.324714   \n",
       "4  0.785179 -0.078615 -1.359076  ...  1.403845 -0.450875  0.264944 -0.603409   \n",
       "\n",
       "         15        16        17        18        19  perturbation_label  \n",
       "0  0.858165  0.190376  0.407933  0.380345  0.605214                 obs  \n",
       "1 -0.104780  0.651097  0.105443  0.042832  0.458566                 obs  \n",
       "2 -0.226251  0.530106 -0.267283 -0.685813  0.020803                 obs  \n",
       "3  0.335550  0.454699 -0.320341 -0.104311  0.478039                 obs  \n",
       "4  1.056394 -1.742338  0.175205  0.388329 -0.720161                 obs  \n",
       "\n",
       "[5 rows x 21 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_df.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, to construct a torch dataset with the appropriate tensors, we use the utility function `sdcd.utils.create_intervention_dataset`.\n",
    " \n",
    "(Note, we also have a function `sdcd.utils.create_dataset_anndata` for users working with the AnnData data format.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<torch.utils.data.dataset.TensorDataset at 0x28a8772b0>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_dataset = create_intervention_dataset(X_df, perturbation_colname=\"perturbation_label\")\n",
    "X_dataset"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, we are ready to train the SDCD model. Here, we will set `finetune=True` which runs the algorithm until the final adjacency matrix is estimated and binarized, then proceeds with a final stage of training the model until it converges for the fixed, discretized adjacency matrix. If you only care about the predicted graph, you may set this to `False` to shorten the runtime."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: loss=27.26, gamma=0.00\n",
      "Epoch 100: loss=9.29, gamma=0.00\n",
      "Epoch 200: loss=7.22, gamma=0.00\n",
      "Epoch 300: loss=6.15, gamma=0.00\n",
      "Epoch 400: loss=5.41, gamma=0.00\n",
      "Epoch 500: loss=4.80, gamma=0.00\n",
      "Epoch 600: loss=4.41, gamma=0.00\n",
      "Epoch 700: loss=4.04, gamma=0.00\n",
      "Epoch 800: loss=3.75, gamma=0.00\n",
      "Epoch 900: loss=3.45, gamma=0.00\n",
      "Epoch 1000: loss=3.21, gamma=0.00\n",
      "Epoch 1100: loss=2.93, gamma=0.00\n",
      "Epoch 1200: loss=2.77, gamma=0.00\n",
      "Epoch 1300: loss=2.61, gamma=0.00\n",
      "Epoch 1400: loss=2.52, gamma=0.00\n",
      "Epoch 1500: loss=2.46, gamma=0.00\n",
      "Epoch 1600: loss=2.37, gamma=0.00\n",
      "Epoch 1700: loss=2.29, gamma=0.00\n",
      "Epoch 1800: loss=2.19, gamma=0.00\n",
      "Epoch 1900: loss=2.09, gamma=0.00\n",
      "Fraction of possible edges in mask: 0.1675\n",
      "Epoch 0: loss=28.21, gamma=0.00\n",
      "Epoch 100: loss=17.03, gamma=0.50\n",
      "Epoch 200: loss=12.50, gamma=1.00\n",
      "Epoch 300: loss=12.00, gamma=1.50\n",
      "Epoch 400: loss=12.17, gamma=2.00\n",
      "Epoch 500: loss=12.37, gamma=2.50\n",
      "Epoch 600: loss=12.70, gamma=3.00\n",
      "Epoch 700: loss=12.95, gamma=3.50\n",
      "Epoch 800: loss=13.24, gamma=4.00\n",
      "Epoch 900: loss=13.51, gamma=4.50\n",
      "Epoch 1000: loss=13.70, gamma=5.00\n",
      "Epoch 1100: loss=14.00, gamma=5.50\n",
      "Epoch 1200: loss=14.20, gamma=6.00\n",
      "Epoch 1300: loss=14.55, gamma=6.50\n",
      "Epoch 1400: loss=14.63, gamma=7.00\n",
      "Epoch 1500: loss=14.85, gamma=7.50\n",
      "Epoch 1600: loss=15.30, gamma=8.00\n",
      "Epoch 1700: loss=15.53, gamma=8.50\n",
      "Epoch 1800: loss=15.67, gamma=9.00\n",
      "Epoch 1900: loss=15.70, gamma=9.50\n",
      "Fixing adjacency matrix.\n",
      "Beginning finetune.\n",
      "Epoch 0: loss=26.42, gamma=0.00\n",
      "Epoch 100: loss=13.35, gamma=0.00\n",
      "Epoch 200: loss=12.87, gamma=0.00\n",
      "Epoch 300: loss=12.49, gamma=0.00\n",
      "Epoch 400: loss=12.25, gamma=0.00\n",
      "Epoch 500: loss=12.05, gamma=0.00\n",
      "Epoch 600: loss=11.92, gamma=0.00\n",
      "Epoch 700: loss=11.75, gamma=0.00\n",
      "Epoch 800: loss=11.71, gamma=0.00\n",
      "Epoch 900: loss=11.58, gamma=0.00\n",
      "Epoch 1000: loss=11.51, gamma=0.00\n",
      "Epoch 1100: loss=11.44, gamma=0.00\n",
      "Epoch 1200: loss=11.38, gamma=0.00\n",
      "Epoch 1300: loss=11.29, gamma=0.00\n",
      "Epoch 1400: loss=11.24, gamma=0.00\n",
      "Epoch 1500: loss=11.20, gamma=0.00\n",
      "Epoch 1600: loss=11.19, gamma=0.00\n",
      "Epoch 1700: loss=11.14, gamma=0.00\n",
      "Epoch 1800: loss=11.11, gamma=0.00\n",
      "Epoch 1900: loss=11.07, gamma=0.00\n",
      "Finished training in 85.46118497848511 seconds.\n"
     ]
    }
   ],
   "source": [
    "model = SDCD()\n",
    "model.train(X_dataset, finetune=True)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lastly, we can recover the predicted adjacency matrix, both before thresholding and after thresholding, and compute likelihoods respect to a given observation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 0 0 0 1]\n",
      " [0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 0 1 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0]\n",
      " [0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0]\n",
      " [0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0]\n",
      " [0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0]\n",
      " [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0]]\n"
     ]
    }
   ],
   "source": [
    "adj_matrix = model.get_adjacency_matrix(threshold=True)\n",
    "print(adj_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# We can visualize the graph with the networkx package (Note: you will need matplotlib for this step)\n",
    "import networkx as nx\n",
    "G = nx.from_numpy_array(adj_matrix, create_using=nx.DiGraph())\n",
    "nx.draw_circular(G, with_labels=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 1.6727905e-03 0.0000000e+00\n",
      "  0.0000000e+00 7.8354633e-01 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.3813855e-03\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 1.9577078e-03 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 2.6522236e+00 0.0000000e+00 0.0000000e+00\n",
      "  5.0185114e-01 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.7044384e-03\n",
      "  0.0000000e+00 0.0000000e+00 1.6327979e-03 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 4.3727842e-01]\n",
      " [0.0000000e+00 1.8689970e-02 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 7.7737542e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 2.8749868e-01 7.4654472e-01\n",
      "  0.0000000e+00 4.9055874e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 2.1733830e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  6.2809849e-01 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 1.1433051e-03 0.0000000e+00 0.0000000e+00 3.5454746e-04\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 1.6724280e-03 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.6190677e-03\n",
      "  0.0000000e+00 0.0000000e+00 1.2286949e-03 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 3.5387962e+00 0.0000000e+00 1.3754174e-01\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.8482170e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 7.1594781e-01 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 4.5380104e-01]\n",
      " [0.0000000e+00 5.1711901e-04 0.0000000e+00 0.0000000e+00 1.9851405e-02\n",
      "  0.0000000e+00 4.4309711e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 5.2044072e+00 0.0000000e+00 2.2932219e-03\n",
      "  0.0000000e+00 0.0000000e+00 1.3206527e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 4.2797357e-01\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 9.5271561e-03 0.0000000e+00\n",
      "  0.0000000e+00 6.8612874e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 2.0644292e-01 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  6.5747511e-01 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [4.1176522e-01 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.4522611e-01\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 7.0423633e-01\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 2.4101284e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [4.5351207e-04 7.2255603e-04 0.0000000e+00 0.0000000e+00 1.1438043e-03\n",
      "  0.0000000e+00 7.7758020e-01 0.0000000e+00 1.2574311e-02 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.4328014e-03\n",
      "  0.0000000e+00 0.0000000e+00 2.7702439e-03 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.2340996e-03\n",
      "  0.0000000e+00 0.0000000e+00 4.9472046e-01 0.0000000e+00 2.5839218e-01\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 2.6131721e+00 0.0000000e+00 4.8997840e-03 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 1.7420980e-01 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 6.1521608e-01 0.0000000e+00 0.0000000e+00 5.2604266e-03\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 2.8826244e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [8.3698623e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00 6.8660494e-04\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 6.7929155e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  5.3244369e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 7.3441479e-04 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 1.2605757e-01\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 3.3022836e-02\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]\n",
      " [7.0602121e-04 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00\n",
      "  0.0000000e+00 0.0000000e+00 0.0000000e+00 7.7821594e-04 0.0000000e+00\n",
      "  0.0000000e+00 3.5236080e+00 0.0000000e+00 0.0000000e+00 0.0000000e+00]]\n"
     ]
    }
   ],
   "source": [
    "# To retrieve the unthresholded adjacency matrix\n",
    "adj_matrix = model.get_adjacency_matrix(threshold=False)\n",
    "print(adj_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8.566566569010417"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.compute_nll(X_dataset) # Reports average negative log-likelihood"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This concludes the tutorial. Please raise an issue on the Github repo if you come across any issues, and we hope you find this package useful to your research!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "causalperturb",
   "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": "3.10.13"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}