iclr_plot.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "authorship_tag": "ABX9TyO1L41rIDT1BCFwhe/ZeafT",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/alps-lab/remasker/blob/main/iclr_plot.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "vJejLg2zZs_G"
      },
      "outputs": [],
      "source": [
        "import matplotlib as mpl\n",
        "from matplotlib import cycler\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import json\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "file = '/content/drive/My Drive/Colab Notebooks/data/letter-cka2.json'\n",
        "\n",
        "data = json.load(open(file))\n",
        "p\n",
        "x = np.arange(10, 601, 10)\n",
        "miss_rates = ['0.1', '0.3', '0.5', '0.7']\n",
        "ymean = []\n",
        "ystd = []\n",
        "\n",
        "for miss_rate in miss_rates:\n",
        "  y = np.array(data[miss_rate])\n",
        "  ymean.append(np.mean(y, 1))\n",
        "  ystd.append(np.std(y, 1))\n",
        "\n",
        "ymean, ystd = np.array(ymean), np.array(ystd)\n",
        "\n",
        "# Fixing random state for reproducibility\n",
        "# np.random.seed(19680801)\n",
        "\n",
        "# N = 10\n",
        "# data = (np.geomspace(1, 10, 100) + np.random.randn(N, 100)).T\n",
        "cmap = plt.cm.coolwarm\n",
        "# # mpl.rcParams['axes.prop_cycle'] = cycler(color=cmap(np.linspace(0, 1, N)))\n",
        "\n",
        "fig, ax = plt.subplots()\n",
        "for i in range(len(miss_rates)):\n",
        "  ax.plot(x, ymean[i,:], color=cmap(i/len(miss_rates)), lw=1, label=miss_rates[i])\n",
        "  ax.fill_between(x,  ymean[i,:] - ystd[i,:], ymean[i,:] + ystd[i,:], color=cmap(i/len(miss_rates)), alpha=0.25)\n",
        "\n",
        "ax.set_ylim(0.4, 0.9)\n",
        "ax.set_yticks(np.arange(0.4, 0.95, 0.05))\n",
        "\n",
        "ax.legend(title='missing ratio')\n",
        "ax.set_xlabel('Training Epochs')\n",
        "ax.set_ylabel('CKA Similarity')\n",
        "fig.savefig('/content/drive/My Drive/Colab Notebooks/data/cka.pdf')\n",
        "\n",
        "# # lines = ax.plot(data)\n",
        "# # ax.legend()\n",
        "\n",
        "# N = 4\n",
        "\n",
        "\n",
        "\n",
        "# custom_lines = [Line2D([0], [0], color=cmap(0.), lw=4),\n",
        "#                 Line2D([0], [0], color=cmap(.5), lw=4),\n",
        "#                 Line2D([0], [0], color=cmap(1.), lw=4)]\n",
        "\n",
        "# fig, ax = plt.subplots()\n",
        "# lines = ax.plot(data)\n",
        "\n",
        "# # ax.fill_between(x, y_est - y_err, y_est + y_err, alpha=0.2)\n",
        "\n",
        "# ax.legend(custom_lines, ['Cold', 'Medium', 'Hot'])\n",
        "\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!/usr/local/cuda/bin/nvcc --version\n",
        "!nvidia-smi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "sXDn8lxBi19w",
        "outputId": "74ddc17d-4e47-4208-fbcd-ba846325da7b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "nvcc: NVIDIA (R) Cuda compiler driver\n",
            "Copyright (c) 2005-2023 NVIDIA Corporation\n",
            "Built on Tue_Aug_15_22:02:13_PDT_2023\n",
            "Cuda compilation tools, release 12.2, V12.2.140\n",
            "Build cuda_12.2.r12.2/compiler.33191640_0\n",
            "/bin/bash: line 1: nvidia-smi: command not found\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib as mpl\n",
        "from matplotlib import cycler\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import json\n",
        "from google.colab import drive\n",
        "drive.mount('/content/drive')\n",
        "\n",
        "file = '/content/drive/My Drive/Colab Notebooks/data/letter-cka2.json'\n",
        "\n",
        "data = {}\n",
        "\n",
        "x = np.arange(1, 6, 1)\n",
        "miss_rates = ['5e-5', '1e-4', '2e-4']\n",
        "ymean = []\n",
        "ystd = []\n",
        "\n",
        "data['1e-4'] = np.array([71.3, 12.9, 9.8, 9.3, 9.4])\n",
        "data['2e-4'] = np.array([10.7, 8.5, 10.7, 9.6, 9.1])\n",
        "data['5e-5'] = np.array([78.7, 72.7, 51, 17.8, 8.9])\n",
        "\n",
        "# data['1e-4'] = np.array([9.3, 8.9, 9.1, 8.3, 10.3])\n",
        "# data['2e-4'] = np.array([9.3, 9.4, 11.2, 9.8, 9.4])\n",
        "# data['5e-5'] = np.array([9.7, 10.4, 8.4, 5.6, 9.1])\n",
        "\n",
        "\n",
        "\n",
        "\n",
        "for miss_rate in miss_rates:\n",
        "  y = np.array(data[miss_rate]/100)\n",
        "  ymean.append(y)\n",
        "  # ystd.append(np.std(y, 1))\n",
        "\n",
        "ymean = np.array(ymean)\n",
        "\n",
        "# Fixing random state for reproducibility\n",
        "# np.random.seed(19680801)\n",
        "\n",
        "# N = 10\n",
        "# data = (np.geomspace(1, 10, 100) + np.random.randn(N, 100)).T\n",
        "cmap = plt.cm.coolwarm\n",
        "# # mpl.rcParams['axes.prop_cycle'] = cycler(color=cmap(np.linspace(0, 1, N)))\n",
        "\n",
        "fig, ax = plt.subplots()\n",
        "for i in range(len(miss_rates)):\n",
        "  ax.plot(x, ymean[i,:], color=cmap(i/len(miss_rates)), lw=2, label=miss_rates[i])\n",
        "  # ax.fill_between(x,  ymean[i,:] - ystd[i,:], ymean[i,:] + ystd[i,:], color=cmap(i/len(miss_rates)), alpha=0.25)\n",
        "\n",
        "ax.set_ylim(0, 1.0)\n",
        "# ax.set_yticks(np.arange(0, 0.1, 1.0))\n",
        "ax.set_xlim(0.75, 5.25)\n",
        "# ax.set_xticks(np.arange(1, 1, 5))\n",
        "\n",
        "# ax.title('ASR (Untargeted Attack)')\n",
        "ax.legend(title='learning rate')\n",
        "ax.set_xlabel('Training Epochs')\n",
        "# ax.set_ylabel('Attack Success Rate (ASR)')\n",
        "\n",
        "ax.set_ylabel('Clean Accuracy (ACC)')\n",
        "\n",
        "ax.spines['right'].set_visible(False)\n",
        "ax.spines['top'].set_visible(False)\n",
        "ax.spines['left'].set_visible(True)\n",
        "ax.spines['bottom'].set_visible(True)\n",
        "\n",
        "fig.savefig('/content/drive/My Drive/Colab Notebooks/data/acc.pdf')\n",
        "\n",
        "# # lines = ax.plot(data)\n",
        "# # ax.legend()\n",
        "\n",
        "# N = 4\n",
        "\n",
        "\n",
        "\n",
        "# custom_lines = [Line2D([0], [0], color=cmap(0.), lw=4),\n",
        "#                 Line2D([0], [0], color=cmap(.5), lw=4),\n",
        "#                 Line2D([0], [0], color=cmap(1.), lw=4)]\n",
        "\n",
        "# fig, ax = plt.subplots()\n",
        "# lines = ax.plot(data)\n",
        "\n",
        "# # ax.fill_between(x, y_est - y_err, y_est + y_err, alpha=0.2)\n",
        "\n",
        "# ax.legend(custom_lines, ['Cold', 'Medium', 'Hot'])\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 472
        },
        "id": "InaNsxYcv4gT",
        "outputId": "b5f932b6-2313-4e70-c1da-362218fbc91a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    }
  ]
}