diff --git a/2_knn/knn_classification.ipynb b/2_knn/knn_classification.ipynb
index f646626c1e12d839457f69d857f8828c93904ee4..be225ab26bb2e74eb31880351f829354f5599c82 100644
--- a/2_knn/knn_classification.ipynb
+++ b/2_knn/knn_classification.ipynb
@@ -176,14 +176,26 @@
     "\n",
     "\n",
     "# 绘制结果\n",
-    "plt.scatter(x_train[:,0], x_train[:,1], c=y_train, marker='.')\n",
-    "plt.title(\"train data\")\n",
-    "plt.savefig(\"knn_train_data.pdf\")\n",
+    "for i in range(split_index):\n",
+    "    if y_train[i] == 0:\n",
+    "        plt.scatter(x_train[i,0],x_train[i,1],c = 0, marker='.')\n",
+    "    else:\n",
+    "        plt.scatter(x_train[i,0],x_train[i,1],c = 1, marker='^')   \n",
+    "plt.rcParams['figure.figsize']=(12.0, 8.0)\n",
+    "mpl.rcParams['font.family'] = 'SimHei'\n",
+    "plt.title(\"训练数据\")\n",
+    "plt.savefig(\"fig-res-train.pdf\")\n",
     "plt.show()\n",
-    "plt.scatter(x_test[:,0], x_test[:,1], c=y_test, marker='.')\n",
-    "plt.title(\"test data\")\n",
-    "plt.savefig(\"knn_test_data.pdf\")\n",
-    "plt.show()\n"
+    "\n",
+    "for i in range(data_size_all - split_index):\n",
+    "    if y_test[i] == 0:\n",
+    "        plt.scatter(x_test[i,0],x_test[i,1],c = 0, marker='.')\n",
+    "    else:\n",
+    "        plt.scatter(x_test[i,0],x_test[i,1],c = 1, marker='^')\n",
+    "plt.rcParams['figure.figsize']=(12.0, 8.0)\n",
+    "plt.title(\"测试数据\")\n",
+    "plt.savefig(\"fig-res-test.pdf\")\n",
+    "plt.show()"
    ]
   },
   {
@@ -475,7 +487,9 @@
     "for i in range(nplot):\n",
     "    img = X_digits[i].reshape(8, 8)\n",
     "    axes[i].imshow(img)\n",
-    "    axes[i].set_title(y_digits[i])\n"
+    "    axes[i].set_title(y_digits[i])\n",
+    "fig.set_size_inches(16,9)\n",
+    "fig.savefig('fig-res-digits.pdf')"
    ]
   },
   {
@@ -574,7 +588,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.5.4"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,