{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "e363540e",
   "metadata": {},
   "source": [
    "# Training ANI from scratch on rMD17 (paracetamol)\n",
    "\n",
    "This trains the `xnn` **ANI** model, the AEV descriptor of Smith et al.\n",
    "(*Chem. Sci.* 2017) feeding one neural network per element, from scratch on a\n",
    "real H/C/N/O molecule, and reproduces the paper's headline demonstrations at\n",
    "small scale:\n",
    "\n",
    "* an **energy/force parity** plot vs DFT (paper Fig. 4), and\n",
    "* a **smooth 1-D potential-energy scan** (paper Fig. 7), the property that makes\n",
    "  ANI usable for MD.\n",
    "\n",
    "We use [rMD17](https://doi.org/10.1021/acs.jctc.9b00181) **paracetamol**\n",
    "(C₈H₉NO₂) because it exercises all four ANI elements and downloads in seconds\n",
    "through the `xnn` hub. (The 20 M-conformation ANI-1 training set itself is also\n",
    "in the hub: `load_dataset(\"ani1\")`; see `ani1_dataset.ipynb`.)\n",
    "\n",
    "Run with the **`xnn`** kernel."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7a8d5b7",
   "metadata": {},
   "source": [
    "## 0. Setup and data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "43df3ad2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-20T06:17:55.945769Z",
     "iopub.status.busy": "2026-07-20T06:17:55.945544Z",
     "iopub.status.idle": "2026-07-20T06:17:58.188822Z",
     "shell.execute_reply": "2026-07-20T06:17:58.188165Z"
    }
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b3b8a2ea201c4d5eb3cff493a6af37da",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "rmd17:paracetamol train:   0%|          | 0/1000 [00:00<?, ? struct/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "32a6227dabf642cb8ad0dc5b8aedcbd3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "rmd17:paracetamol test:   0%|          | 0/400 [00:00<?, ? struct/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "train 1000  test 400\n",
      "elements: [1, 6, 7, 8]  atoms/mol: 20\n"
     ]
    }
   ],
   "source": [
    "import time, warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "import numpy as np\n",
    "import torch\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "torch.set_default_dtype(torch.float64)\n",
    "torch.manual_seed(0)\n",
    "\n",
    "from xnn.common.data import load_dataset\n",
    "\n",
    "CUTOFF = 5.2                       # ANI-1x radial cutoff (>= angular 3.5)\n",
    "splits = load_dataset(\"rmd17\", molecule=\"paracetamol\",\n",
    "                      n_train=1000, n_test=400, cutoff=None)\n",
    "train_structs, test_structs = splits[\"train\"], splits[\"test\"]\n",
    "print(f\"train {len(train_structs)}  test {len(test_structs)}\")\n",
    "Z0 = train_structs[0][\"atomic_numbers\"]\n",
    "print(\"elements:\", sorted(set(int(z) for z in Z0)), \" atoms/mol:\", len(Z0))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ae47ce41",
   "metadata": {},
   "source": [
    "## 1. Self atomic energies (ANI's `EnergyShifter`)\n",
    "\n",
    "ANI predicts the energy *relative* to per-element self energies, fit here by\n",
    "least squares $E \\approx \\sum_Z n_Z\\,e_Z$ and subtracted from the targets. The\n",
    "network then learns the small residual, the chemically interesting part, which\n",
    "also keeps the fit numerically well conditioned."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e5366d14",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-20T06:17:58.190587Z",
     "iopub.status.busy": "2026-07-20T06:17:58.190383Z",
     "iopub.status.idle": "2026-07-20T06:17:58.233419Z",
     "shell.execute_reply": "2026-07-20T06:17:58.232770Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "self energies (eV): {1: -839.943, 6: -746.616, 7: -93.327, 8: -186.654}\n",
      "residual energy: mean -0.000 eV  std 0.271 eV\n"
     ]
    }
   ],
   "source": [
    "SPECIES = [1, 6, 7, 8]\n",
    "counts = np.array([[np.sum(s[\"atomic_numbers\"] == z) for z in SPECIES]\n",
    "                   for s in train_structs], dtype=float)\n",
    "E_train = np.array([s[\"energy\"] for s in train_structs])\n",
    "sae, *_ = np.linalg.lstsq(counts, E_train, rcond=None)\n",
    "print(\"self energies (eV):\", {z: round(float(e), 3) for z, e in zip(SPECIES, sae)})\n",
    "\n",
    "def subtract_sae(structs):\n",
    "    for s in structs:\n",
    "        n = np.array([np.sum(s[\"atomic_numbers\"] == z) for z in SPECIES])\n",
    "        s[\"energy\"] = float(s[\"energy\"] - n @ sae)\n",
    "\n",
    "subtract_sae(train_structs); subtract_sae(test_structs)\n",
    "resid = np.array([s[\"energy\"] for s in train_structs])\n",
    "print(f\"residual energy: mean {resid.mean():.3f} eV  std {resid.std():.3f} eV\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f189cef7",
   "metadata": {},
   "source": [
    "## 2. Build the ANI model and train\n",
    "\n",
    "We use the ANI-1x AEV grid (384-length) with a compact per-element MLP and the\n",
    "`CELU` activation, trained on energies **and** forces (forces come for free via\n",
    "autograd through the `ForceStressOutput` wrapper the `Trainer` adds)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b819dfd8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-20T06:17:58.235001Z",
     "iopub.status.busy": "2026-07-20T06:17:58.234882Z",
     "iopub.status.idle": "2026-07-20T06:20:56.499464Z",
     "shell.execute_reply": "2026-07-20T06:20:56.498489Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ANI parameters: 296,452  |  device: cuda\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "test loss 3.2432e-02\n",
      "trained 80 epochs in 177.5s | test loss 3.2432e-02\n"
     ]
    }
   ],
   "source": [
    "from torch.utils.data import Subset\n",
    "from xnn.common.data import AtomicDataset\n",
    "from xnn.common.config import Config, ModelConfig, DataConfig, OptimConfig\n",
    "from xnn.common.train import Trainer\n",
    "\n",
    "EPOCHS, BS, LR = 80, 32, 1e-3\n",
    "\n",
    "train_ds = AtomicDataset(train_structs, CUTOFF)\n",
    "test_ds = AtomicDataset(test_structs, CUTOFF)\n",
    "n_val = 100\n",
    "val_idx = list(range(n_val))\n",
    "tr_idx = list(range(n_val, len(train_ds)))\n",
    "\n",
    "cfg = Config(\n",
    "    model=ModelConfig(name=\"ani\", cutoff=CUTOFF,\n",
    "                      extra={\"species\": SPECIES, \"hidden\": [128, 128, 64],\n",
    "                             \"activation\": \"celu\"}),\n",
    "    data=DataConfig(cutoff=CUTOFF, batch_size=BS),\n",
    "    optim=OptimConfig(lr=LR, epochs=EPOCHS, energy_weight=1.0,\n",
    "                      force_weight=10.0, scheduler=\"plateau\"),\n",
    "    output_dir=\"runs/ani_rmd17\",\n",
    ")\n",
    "\n",
    "trainer = Trainer(cfg, Subset(train_ds, tr_idx), Subset(train_ds, val_idx), test_ds)\n",
    "n_params = sum(p.numel() for p in trainer.module.parameters())\n",
    "print(f\"ANI parameters: {n_params:,}  |  device: {trainer.device}\")\n",
    "\n",
    "hist = {\"train\": [], \"val\": []}\n",
    "trainer._log = lambda ep, tr, va: (hist[\"train\"].append(tr.get(\"loss\")),\n",
    "                                   hist[\"val\"].append(va.get(\"loss\")))\n",
    "t0 = time.time()\n",
    "metrics = trainer.fit()\n",
    "print(f\"trained {EPOCHS} epochs in {time.time()-t0:.1f}s | test loss {metrics['test']['loss']:.4e}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba95bfde",
   "metadata": {},
   "source": [
    "## 3. Learning curves"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "653abad5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-20T06:20:56.502398Z",
     "iopub.status.busy": "2026-07-20T06:20:56.502142Z",
     "iopub.status.idle": "2026-07-20T06:20:56.863094Z",
     "shell.execute_reply": "2026-07-20T06:20:56.862165Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x320 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(5, 3.2))\n",
    "ax.plot(hist[\"train\"], label=\"train\")\n",
    "ax.plot(hist[\"val\"], label=\"val\")\n",
    "ax.set_yscale(\"log\"); ax.set_xlabel(\"epoch\"); ax.set_ylabel(\"loss\")\n",
    "ax.set_title(\"ANI on rMD17 paracetamol\"); ax.legend(); fig.tight_layout()\n",
    "fig.savefig(\"ani_loss_curves.png\", dpi=110); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5047f597",
   "metadata": {},
   "source": [
    "## 4. Energy & force parity vs DFT (paper Fig. 4)\n",
    "\n",
    "Predict on the held-out test set and compare to reference DFT. We report the\n",
    "energy RMSE in meV and kcal/mol (the paper's unit)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "4d284528",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-20T06:20:56.865409Z",
     "iopub.status.busy": "2026-07-20T06:20:56.865282Z",
     "iopub.status.idle": "2026-07-20T06:20:57.915065Z",
     "shell.execute_reply": "2026-07-20T06:20:57.914117Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "energy RMSE: 31.23 meV  (0.720 kcal/mol)\n",
      "force  RMSE: 57.1 meV/A  (1.317 kcal/mol/A)\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 900x400 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from torch.utils.data import DataLoader\n",
    "from xnn.common.data import collate\n",
    "\n",
    "model = trainer.model.eval()\n",
    "device = trainer.device\n",
    "EV2KCAL = 23.060541945329334\n",
    "\n",
    "E_pred, E_ref, F_pred, F_ref = [], [], [], []\n",
    "loader = DataLoader(test_ds, batch_size=32, collate_fn=collate)\n",
    "for batch in loader:\n",
    "    batch = batch.to(device)\n",
    "    out = model(batch)\n",
    "    E_pred.append(out[\"energy\"].detach().cpu().numpy())\n",
    "    E_ref.append(batch.energy.detach().cpu().numpy())\n",
    "    F_pred.append(out[\"forces\"].detach().cpu().numpy())\n",
    "    F_ref.append(batch.forces.detach().cpu().numpy())\n",
    "E_pred, E_ref = np.concatenate(E_pred), np.concatenate(E_ref)\n",
    "F_pred, F_ref = np.concatenate(F_pred), np.concatenate(F_ref)\n",
    "\n",
    "e_rmse = np.sqrt(np.mean((E_pred - E_ref) ** 2))\n",
    "f_rmse = np.sqrt(np.mean((F_pred - F_ref) ** 2))\n",
    "print(f\"energy RMSE: {e_rmse*1e3:.2f} meV  ({e_rmse*EV2KCAL:.3f} kcal/mol)\")\n",
    "print(f\"force  RMSE: {f_rmse*1e3:.1f} meV/A  ({f_rmse*EV2KCAL:.3f} kcal/mol/A)\")\n",
    "\n",
    "fig, (a1, a2) = plt.subplots(1, 2, figsize=(9, 4))\n",
    "a1.scatter(E_ref, E_pred, s=8, alpha=0.5)\n",
    "lim = [min(E_ref.min(), E_pred.min()), max(E_ref.max(), E_pred.max())]\n",
    "a1.plot(lim, lim, \"k--\", lw=1)\n",
    "a1.set_xlabel(\"DFT energy (eV)\"); a1.set_ylabel(\"ANI energy (eV)\")\n",
    "a1.set_title(f\"energy  RMSE {e_rmse*EV2KCAL:.3f} kcal/mol\")\n",
    "a2.scatter(F_ref.ravel(), F_pred.ravel(), s=3, alpha=0.2)\n",
    "lim = [F_ref.min(), F_ref.max()]\n",
    "a2.plot(lim, lim, \"k--\", lw=1)\n",
    "a2.set_xlabel(\"DFT force (eV/A)\"); a2.set_ylabel(\"ANI force (eV/A)\")\n",
    "a2.set_title(f\"forces  RMSE {f_rmse*EV2KCAL:.3f} kcal/mol/A\")\n",
    "fig.tight_layout(); fig.savefig(\"ani_parity.png\", dpi=110); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ff398be",
   "metadata": {},
   "source": [
    "## 5. A smooth 1-D potential-energy scan (paper Fig. 7)\n",
    "\n",
    "The value of ANI over semi-empirical methods is a **smooth, differentiable**\n",
    "surface. We take a test geometry and stretch its first C–N bond, evaluating the\n",
    "trained ANI energy along the scan. A smooth, single-well curve (not a jagged\n",
    "one) is what enables stable MD."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "7518365c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-20T06:20:57.917474Z",
     "iopub.status.busy": "2026-07-20T06:20:57.917355Z",
     "iopub.status.idle": "2026-07-20T06:20:59.198287Z",
     "shell.execute_reply": "2026-07-20T06:20:59.197554Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x320 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from xnn.common.data.neighborlist import build_neighbor_list\n",
    "from xnn.common.data import AtomicGraph\n",
    "\n",
    "s = test_structs[0]\n",
    "Z = torch.tensor(s[\"atomic_numbers\"], dtype=torch.long)\n",
    "pos0 = torch.tensor(s[\"pos\"], dtype=torch.float64)\n",
    "\n",
    "# pick a C(6)-N(7) bonded pair\n",
    "Cs = (Z == 6).nonzero().flatten(); N = (Z == 7).nonzero().flatten()[0]\n",
    "d = torch.linalg.norm(pos0[Cs] - pos0[N], dim=1)\n",
    "i, j = int(Cs[d.argmin()]), int(N)          # closest C to the N\n",
    "axis = (pos0[j] - pos0[i]); axis = axis / axis.norm()\n",
    "r0 = torch.linalg.norm(pos0[j] - pos0[i]).item()\n",
    "\n",
    "scan = np.linspace(-0.4, 0.6, 41)\n",
    "energies = []\n",
    "for dr in scan:\n",
    "    pos = pos0.clone(); pos[j] = pos[j] + dr * axis\n",
    "    ei, cs = build_neighbor_list(pos, CUTOFF)\n",
    "    g = AtomicGraph(pos=pos.to(device), atomic_numbers=Z.to(device),\n",
    "                    edge_index=ei.to(device), cell_shifts=cs.to(device),\n",
    "                    batch=torch.zeros(len(Z), dtype=torch.long, device=device),\n",
    "                    n_atoms=torch.tensor([len(Z)], device=device))\n",
    "    with torch.no_grad():\n",
    "        energies.append(model(g)[\"energy\"].item())\n",
    "energies = np.array(energies) - min(energies)\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(5, 3.2))\n",
    "ax.plot(r0 + scan, energies, \"o-\", ms=3)\n",
    "ax.set_xlabel(\"C–N distance (A)\"); ax.set_ylabel(\"relative energy (eV)\")\n",
    "ax.set_title(\"ANI C–N bond scan (smooth PES)\"); fig.tight_layout()\n",
    "fig.savefig(\"ani_bond_scan.png\", dpi=110); plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6e2ebb09",
   "metadata": {},
   "source": [
    "## Summary\n",
    "\n",
    "From scratch on 900 paracetamol conformations, `xnn` ANI reaches chemical-scale\n",
    "energy/force accuracy and produces a smooth potential-energy surface: the two\n",
    "properties the ANI-1 paper emphasises. The same model, at full scale on the\n",
    "`load_dataset(\"ani1\")` training set and the published network widths\n",
    "(`ANI.ani1x()`), is the ANI-1x potential verified element-for-element against\n",
    "torchani in `examples/fidelity_checks/ani_verification.ipynb`."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "xnn (.venv)",
   "language": "python",
   "name": "xnn"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.12"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "state": {
     "0b46fbe28de440c281c98b0d1677a473": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "0f3e6314ba124ac081acd4669cebc693": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "StyleView",
       "background": null,
       "description_width": "",
       "font_size": null,
       "text_color": null
      }
     },
     "10c3187f0fad4dd8a16ba310fe0c99ef": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "StyleView",
       "background": null,
       "description_width": "",
       "font_size": null,
       "text_color": null
      }
     },
     "18b1e54be7c946149d8c37232cabb77e": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "32a6227dabf642cb8ad0dc5b8aedcbd3": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_b3ac88f99c10440db6c32c75716a8b67",
        "IPY_MODEL_3b9c07cd2e5849cd84139a59ffc596d8",
        "IPY_MODEL_647097305d364372a58c4bd223b5148b"
       ],
       "layout": "IPY_MODEL_5d8f2a8bbe2845c484667160a85ea87d",
       "tabbable": null,
       "tooltip": null
      }
     },
     "3b9c07cd2e5849cd84139a59ffc596d8": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "ProgressView",
       "bar_style": "",
       "description": "",
       "description_allow_html": false,
       "layout": "IPY_MODEL_44af2de1c53b4908a12712e78543bafe",
       "max": 400.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_5e09b2ee44764506a896c631da53aa9d",
       "tabbable": null,
       "tooltip": null,
       "value": 400.0
      }
     },
     "4417777c01f4493db255642ef4c48081": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": "hidden",
       "width": null
      }
     },
     "44af2de1c53b4908a12712e78543bafe": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "5d8f2a8bbe2845c484667160a85ea87d": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": "hidden",
       "width": null
      }
     },
     "5e09b2ee44764506a896c631da53aa9d": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "647097305d364372a58c4bd223b5148b": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_allow_html": false,
       "layout": "IPY_MODEL_f8fcfa1531b24657b1af3c9117a9c10a",
       "placeholder": "​",
       "style": "IPY_MODEL_0f3e6314ba124ac081acd4669cebc693",
       "tabbable": null,
       "tooltip": null,
       "value": " 0/400 [00:00&lt;?, ? struct/s]"
      }
     },
     "6e96c07d1480464cb4021d17ce4f71db": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "ProgressStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "ProgressStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "StyleView",
       "bar_color": null,
       "description_width": ""
      }
     },
     "98baa6fdffaf47a2a709b97e4be98b63": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "9f5da5a84f6e4a178bb53fa9ffbc0725": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     },
     "a02caa76e9434cd28f4b7a816c127ccf": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "StyleView",
       "background": null,
       "description_width": "",
       "font_size": null,
       "text_color": null
      }
     },
     "b3ac88f99c10440db6c32c75716a8b67": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_allow_html": false,
       "layout": "IPY_MODEL_9f5da5a84f6e4a178bb53fa9ffbc0725",
       "placeholder": "​",
       "style": "IPY_MODEL_a02caa76e9434cd28f4b7a816c127ccf",
       "tabbable": null,
       "tooltip": null,
       "value": "rmd17:paracetamol test:   0%"
      }
     },
     "b3b8a2ea201c4d5eb3cff493a6af37da": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HBoxModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HBoxModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "HBoxView",
       "box_style": "",
       "children": [
        "IPY_MODEL_f56e0ebf180f4ef58fbed96873748867",
        "IPY_MODEL_e7bca64b32a846bdba73b44f9b051d14",
        "IPY_MODEL_de35484a4b7f41a6886a8afcb5653793"
       ],
       "layout": "IPY_MODEL_4417777c01f4493db255642ef4c48081",
       "tabbable": null,
       "tooltip": null
      }
     },
     "da2749dfe77a49edae41b5c97b14032f": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLStyleModel",
      "state": {
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLStyleModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "StyleView",
       "background": null,
       "description_width": "",
       "font_size": null,
       "text_color": null
      }
     },
     "de35484a4b7f41a6886a8afcb5653793": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_allow_html": false,
       "layout": "IPY_MODEL_0b46fbe28de440c281c98b0d1677a473",
       "placeholder": "​",
       "style": "IPY_MODEL_da2749dfe77a49edae41b5c97b14032f",
       "tabbable": null,
       "tooltip": null,
       "value": " 0/1000 [00:00&lt;?, ? struct/s]"
      }
     },
     "e7bca64b32a846bdba73b44f9b051d14": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "FloatProgressModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "FloatProgressModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "ProgressView",
       "bar_style": "",
       "description": "",
       "description_allow_html": false,
       "layout": "IPY_MODEL_98baa6fdffaf47a2a709b97e4be98b63",
       "max": 1000.0,
       "min": 0.0,
       "orientation": "horizontal",
       "style": "IPY_MODEL_6e96c07d1480464cb4021d17ce4f71db",
       "tabbable": null,
       "tooltip": null,
       "value": 1000.0
      }
     },
     "f56e0ebf180f4ef58fbed96873748867": {
      "model_module": "@jupyter-widgets/controls",
      "model_module_version": "2.0.0",
      "model_name": "HTMLModel",
      "state": {
       "_dom_classes": [],
       "_model_module": "@jupyter-widgets/controls",
       "_model_module_version": "2.0.0",
       "_model_name": "HTMLModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/controls",
       "_view_module_version": "2.0.0",
       "_view_name": "HTMLView",
       "description": "",
       "description_allow_html": false,
       "layout": "IPY_MODEL_18b1e54be7c946149d8c37232cabb77e",
       "placeholder": "​",
       "style": "IPY_MODEL_10c3187f0fad4dd8a16ba310fe0c99ef",
       "tabbable": null,
       "tooltip": null,
       "value": "rmd17:paracetamol train:   0%"
      }
     },
     "f8fcfa1531b24657b1af3c9117a9c10a": {
      "model_module": "@jupyter-widgets/base",
      "model_module_version": "2.0.0",
      "model_name": "LayoutModel",
      "state": {
       "_model_module": "@jupyter-widgets/base",
       "_model_module_version": "2.0.0",
       "_model_name": "LayoutModel",
       "_view_count": null,
       "_view_module": "@jupyter-widgets/base",
       "_view_module_version": "2.0.0",
       "_view_name": "LayoutView",
       "align_content": null,
       "align_items": null,
       "align_self": null,
       "border_bottom": null,
       "border_left": null,
       "border_right": null,
       "border_top": null,
       "bottom": null,
       "display": null,
       "flex": null,
       "flex_flow": null,
       "grid_area": null,
       "grid_auto_columns": null,
       "grid_auto_flow": null,
       "grid_auto_rows": null,
       "grid_column": null,
       "grid_gap": null,
       "grid_row": null,
       "grid_template_areas": null,
       "grid_template_columns": null,
       "grid_template_rows": null,
       "height": null,
       "justify_content": null,
       "justify_items": null,
       "left": null,
       "margin": null,
       "max_height": null,
       "max_width": null,
       "min_height": null,
       "min_width": null,
       "object_fit": null,
       "object_position": null,
       "order": null,
       "overflow": null,
       "padding": null,
       "right": null,
       "top": null,
       "visibility": null,
       "width": null
      }
     }
    },
    "version_major": 2,
    "version_minor": 0
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}