diff --git a/Training_IC_on_Demand.ipynb b/Training_IC_on_Demand.ipynb
new file mode 100644
index 0000000..e375705
--- /dev/null
+++ b/Training_IC_on_Demand.ipynb
@@ -0,0 +1,133 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "f7c63b6e",
+   "metadata": {},
+   "source": [
+    "# Create ICs on Demand"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "7fefd6b2",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pylab as plt\n",
+    "\n",
+    "from models.multiresolution_flow_3d import *\n",
+    "from models.trainer import *\n",
+    "\n",
+    "%matplotlib inline"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e5aa8c20",
+   "metadata": {},
+   "source": [
+    "# Setup the model"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "3d84fb99",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "(32, 32, 32)\n",
+      "Trainer model: tm\n",
+      "load initial state\n",
+      "save initial state\n"
+     ]
+    }
+   ],
+   "source": [
+    "nlevel=5\n",
+    "\n",
+    "shape = (2**nlevel,2**nlevel,2**nlevel)\n",
+    "\n",
+    "model = multi_scale_model(nlevel=nlevel)\n",
+    "\n",
+    "tm = trainer(model)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "476e8808",
+   "metadata": {},
+   "source": [
+    "# test training with white noise"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "bba2858b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Train model...\n",
+      "test [[[3.14149864]]]\n",
+      "Training done\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "0"
+      ]
+     },
+     "execution_count": 11,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "for i in np.arange(100):\n",
+    "    x_train = np.random.normal(0,0.01,shape)+3.1415\n",
+    "    tm.train_single(x_train, silent=True)\n",
+    "tm.transfer(silent=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "75b0b971",
+   "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": "3.9.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}