diff --git a/examples/notebooks/training-8gaussians-to-moons.ipynb b/examples/notebooks/training-8gaussians-to-moons.ipynb
index 003e02a..418d35a 100644
--- a/examples/notebooks/training-8gaussians-to-moons.ipynb
+++ b/examples/notebooks/training-8gaussians-to-moons.ipynb
@@ -2,7 +2,6 @@
  "cells": [
   {
    "cell_type": "markdown",
-   "id": "a7cf12e8",
    "metadata": {
     "tags": []
    },
@@ -27,7 +26,6 @@
   {
    "cell_type": "code",
    "execution_count": 1,
-   "id": "977e54ff",
    "metadata": {},
    "outputs": [],
    "source": [
@@ -53,7 +51,6 @@
   },
   {
    "cell_type": "markdown",
-   "id": "f86cc21f",
    "metadata": {
     "tags": []
    },
@@ -75,7 +72,6 @@
   {
    "cell_type": "code",
    "execution_count": 2,
-   "id": "f7eb28f3",
    "metadata": {
     "tags": []
    },
@@ -199,7 +195,6 @@
   },
   {
    "cell_type": "markdown",
-   "id": "f195ebf0",
    "metadata": {
     "tags": []
    },
@@ -221,7 +216,6 @@
   {
    "cell_type": "code",
    "execution_count": 3,
-   "id": "b14304ba",
    "metadata": {},
    "outputs": [
     {
@@ -343,7 +337,6 @@
   },
   {
    "cell_type": "markdown",
-   "id": "99ef2f00",
    "metadata": {
     "tags": []
    },
@@ -369,7 +362,6 @@
   {
    "cell_type": "code",
    "execution_count": 4,
-   "id": "0e4e0920",
    "metadata": {},
    "outputs": [
     {
@@ -456,7 +448,8 @@
     "batch_size = 256\n",
     "model = MLP(dim=dim, time_varying=True)\n",
     "optimizer = torch.optim.Adam(model.parameters())\n",
-    "FM = SchrodingerBridgeConditionalFlowMatcher(sigma=sigma)\n",
+    "# For best performance, use ot_method=\"exact\". To follow the theory, use ot_method=\"sinkhorn\"\n",
+    "FM = SchrodingerBridgeConditionalFlowMatcher(sigma=sigma, ot_method=\"exact\")\n",
     "\n",
     "start = time.time()\n",
     "for k in range(20000):\n",
@@ -491,7 +484,6 @@
   },
   {
    "cell_type": "markdown",
-   "id": "f3597e13",
    "metadata": {
     "tags": []
    },
@@ -512,7 +504,6 @@
   {
    "cell_type": "code",
    "execution_count": 5,
-   "id": "c2272ec1",
    "metadata": {},
    "outputs": [
     {
@@ -633,7 +624,6 @@
   },
   {
    "cell_type": "markdown",
-   "id": "dc89e19a",
    "metadata": {
     "tags": []
    },
@@ -655,7 +645,6 @@
   {
    "cell_type": "code",
    "execution_count": 6,
-   "id": "d047920d",
    "metadata": {},
    "outputs": [
     {
@@ -781,7 +770,6 @@
   {
    "cell_type": "code",
    "execution_count": 7,
-   "id": "74afe15e",
    "metadata": {},
    "outputs": [
     {
@@ -917,7 +905,6 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "05cc909b",
    "metadata": {},
    "outputs": [],
    "source": []
@@ -925,7 +912,6 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "ecd5bc7a",
    "metadata": {},
    "outputs": [],
    "source": []
@@ -933,7 +919,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "myenv",
+   "display_name": "riemanian_flow_matching",
    "language": "python",
    "name": "myenv"
   },
@@ -947,7 +933,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
+   "version": "3.9.17"
   }
  },
  "nbformat": 4,