fig_12.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "## Divergence of the weights on Mnist\n",
        "\n",
        "Skeleton borrowed from Flax tutorials distributed under Apache License 2.0: https://github.com/google/flax/tree/main/examples/mnist\n",
        "\n"
      ],
      "metadata": {
        "id": "t1DcxGxbRjMW"
      }
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "inJ9bV636dRx",
        "tags": [
          "skip-execution"
        ],
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ccb2b7db-e4c0-48cb-a436-d0dec584240c"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[K     |████████████████████████████████| 202 kB 5.4 MB/s \n",
            "\u001b[K     |████████████████████████████████| 9.1 MB 30.3 MB/s \n",
            "\u001b[K     |████████████████████████████████| 596 kB 52.4 MB/s \n",
            "\u001b[K     |████████████████████████████████| 145 kB 62.9 MB/s \n",
            "\u001b[K     |████████████████████████████████| 217 kB 54.1 MB/s \n",
            "\u001b[K     |████████████████████████████████| 51 kB 3.4 MB/s \n",
            "\u001b[K     |████████████████████████████████| 72 kB 634 kB/s \n",
            "\u001b[?25h"
          ]
        }
      ],
      "source": [
        "!pip install -q flax"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Z7MuGFB16E8m"
      },
      "outputs": [],
      "source": [
        "import jax\n",
        "import jax.numpy as jnp                # JAX NumPy\n",
        "\n",
        "from flax import linen as nn           # The Linen API\n",
        "from flax.training import train_state  # Useful dataclass to keep train state\n",
        "\n",
        "import numpy as np                     # Ordinary NumPy\n",
        "import optax                           # Optimizers\n",
        "import tensorflow as tf"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tf.config.experimental.set_visible_devices([], 'GPU')\n",
        "use_float64 = True\n",
        "if use_float64:\n",
        "  from jax.config import config\n",
        "  config.update(\"jax_enable_x64\", True)\n",
        "  dtype = jnp.float64\n",
        "else:\n",
        "  dtype = jnp.float32"
      ],
      "metadata": {
        "id": "hy_Zmuv5a9tK"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_s1lXBBO66dc"
      },
      "outputs": [],
      "source": [
        "class CNN(nn.Module):\n",
        "  \"\"\"A simple CNN model.\"\"\"\n",
        "\n",
        "  @nn.compact\n",
        "  def __call__(self, x):\n",
        "    x = nn.Conv(features=32, kernel_size=(3, 3), dtype=dtype)(x)\n",
        "    x = nn.relu(x)\n",
        "    x = nn.avg_pool(x, window_shape=(2, 2), strides=(2, 2))\n",
        "    x = nn.Conv(features=64, kernel_size=(3, 3), dtype=dtype)(x)\n",
        "    x = nn.relu(x)\n",
        "    x = nn.avg_pool(x, window_shape=(2, 2), strides=(2, 2))\n",
        "    x = x.reshape((x.shape[0], -1))  # flatten\n",
        "    x = nn.Dense(features=256, dtype=dtype)(x)\n",
        "    x = nn.relu(x)\n",
        "    x = nn.Dense(features=10, dtype=dtype)(x)\n",
        "    return x"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Zcb_ebU87G7s"
      },
      "outputs": [],
      "source": [
        "def cross_entropy_loss(*, logits, labels):\n",
        "  labels_onehot = jax.nn.one_hot(labels, num_classes=10)\n",
        "  return optax.softmax_cross_entropy(logits=logits, labels=labels_onehot).mean()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "KvuEA8Tw-MYa"
      },
      "outputs": [],
      "source": [
        "def compute_metrics(*, logits, labels):\n",
        "  loss = cross_entropy_loss(logits=logits, labels=labels)\n",
        "  accuracy = jnp.mean(jnp.argmax(logits, -1) == labels)\n",
        "  metrics = {\n",
        "      'loss': loss,\n",
        "      'accuracy': accuracy,\n",
        "  }\n",
        "  return metrics"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "IOeWiS_b-p8O"
      },
      "outputs": [],
      "source": [
        "def get_datasets():\n",
        "  \"\"\"Load MNIST train and test datasets into memory.\"\"\"\n",
        "  (x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()\n",
        "  train_ds = {'image':x_train.reshape((-1,28,28,1)), 'label':y_train}\n",
        "  test_ds = {'image':x_test.reshape((-1,28,28,1)), 'label':y_test}\n",
        "  if dtype == jnp.float32:\n",
        "    train_ds['image'] = jnp.float32(train_ds['image']) / 255.\n",
        "    test_ds['image'] = jnp.float32(test_ds['image']) / 255.\n",
        "  else:\n",
        "    train_ds['image'] = jnp.float64(train_ds['image']) / 255.\n",
        "    test_ds['image'] = jnp.float64(test_ds['image']) / 255.\n",
        "  train_ds['label'] = jnp.array(train_ds['label'])\n",
        "  test_ds['label'] = jnp.array(test_ds['label'])\n",
        "  return train_ds, test_ds"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "QadyBPbWBEAT"
      },
      "outputs": [],
      "source": [
        "def create_train_state(rng, tx):\n",
        "  \"\"\"Creates initial `TrainState`.\"\"\"\n",
        "  cnn = CNN()\n",
        "  params = cnn.init(rng, jnp.ones([1, 28, 28, 1]))['params']\n",
        "  return train_state.TrainState.create(\n",
        "      apply_fn=cnn.apply, params=params, tx=tx)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Ng11cdMf-z0x"
      },
      "outputs": [],
      "source": [
        "@jax.jit\n",
        "def train_step(state, batch):\n",
        "  \"\"\"Train for a single step.\"\"\"\n",
        "  def loss_fn(params):\n",
        "    logits = CNN().apply({'params': params}, batch['image'])\n",
        "    loss = cross_entropy_loss(logits=logits, labels=batch['label'])\n",
        "    return loss, logits\n",
        "  grad_fn = jax.value_and_grad(loss_fn, has_aux=True)\n",
        "  (_, logits), grads = grad_fn(state.params)\n",
        "  state = state.apply_gradients(grads=grads)\n",
        "  metrics = compute_metrics(logits=logits, labels=batch['label'])\n",
        "  return state, metrics"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "w1J9i6alBv_u"
      },
      "outputs": [],
      "source": [
        "def eval_step(params, batch):\n",
        "  logits = CNN().apply({'params': params}, batch['image'])\n",
        "  return compute_metrics(logits=logits, labels=batch['label'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "7ipyJ-JGCNqP"
      },
      "outputs": [],
      "source": [
        "def train_epoch(state, train_ds, batch_size, epoch, rng):\n",
        "  \"\"\"Train for a single epoch.\"\"\"\n",
        "  train_ds_size = len(train_ds['image'])\n",
        "  steps_per_epoch = train_ds_size // batch_size\n",
        "\n",
        "  perms = jax.random.permutation(rng, train_ds_size)\n",
        "  perms = perms[:steps_per_epoch*batch_size]  # drop last batch\n",
        "  train_ds['image'] = train_ds['image'][perms]\n",
        "  train_ds['label'] = train_ds['label'][perms]\n",
        "  batch_metrics = []\n",
        "  for batch_idx in range(0, train_ds_size, batch_size):\n",
        "    batch = {k: v[batch_idx:batch_idx+batch_size, ...] for k, v in train_ds.items()}\n",
        "    state, metrics = train_step(state, batch)\n",
        "    batch_metrics.append(metrics)\n",
        "  del train_ds\n",
        "\n",
        "  # compute mean of metrics across each batch in epoch.\n",
        "  # batch_metrics_np = jax.device_get(batch_metrics)\n",
        "  epoch_metrics_np = {\n",
        "      k: jnp.mean(jnp.array([metrics[k] for metrics in batch_metrics]))\n",
        "      for k in batch_metrics[0]}\n",
        "\n",
        "  print('train epoch: %d, loss: %.4f, accuracy: %.2f' % (\n",
        "      epoch, epoch_metrics_np['loss'], epoch_metrics_np['accuracy'] * 100))\n",
        "\n",
        "  return state"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "_dKahNmMCr5q"
      },
      "outputs": [],
      "source": [
        "def eval_model(params, test_ds):\n",
        "  metrics = eval_step(params, test_ds)\n",
        "  # metrics = jax.device_get(metrics)\n",
        "  summary = jax.tree_util.tree_map(lambda x: x.item(), metrics)\n",
        "  return summary['loss'], summary['accuracy']"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import jax.tree_util as tu\n",
        "\n",
        "def has_kernel(tree):\n",
        "  return 'kernel' in tree\n",
        "\n",
        "def spectral_norm(kernel):\n",
        "  kernel = kernel.reshape((-1, kernel.shape[-1]))\n",
        "  return jnp.linalg.norm(kernel)\n",
        "\n",
        "def eval_spectral_norm(params_tree, weight_norm=spectral_norm):\n",
        "  ker_layers = tu.tree_leaves(params_tree, has_kernel)\n",
        "  kernels = [layer['kernel'] for layer in ker_layers]\n",
        "  norms = [weight_norm(kernel) for kernel in kernels]\n",
        "  return [float(norm) for norm in norms]"
      ],
      "metadata": {
        "id": "_Fmvf5abCMhC"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6CLXnP3KHptR",
        "outputId": "cb8eee08-840c-4278-be43-d85a815b68ff"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Downloading data from https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz\n",
            "11493376/11490434 [==============================] - 0s 0us/step\n",
            "11501568/11490434 [==============================] - 0s 0us/step\n"
          ]
        }
      ],
      "source": [
        "train_ds, test_ds = get_datasets()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "n53xh_B8Ht_W"
      },
      "outputs": [],
      "source": [
        "init_rng = jax.random.PRNGKey(0)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "Mj6OfdEEIU-o"
      },
      "outputs": [],
      "source": [
        "learning_rate = 0.1\n",
        "momentum = 0.9"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "0nxgS5Z5IsT_"
      },
      "outputs": [],
      "source": [
        "num_epochs = 25\n",
        "batch_size = 32"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ugGlV3u6Iq1A",
        "outputId": "83897cc9-3a99-47c6-d9ff-410f89283517"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " test epoch: 0, loss: 2.30950, accuracy: 12.86 norms: 5.58,8.03,16.00,3.22\n",
            "train epoch: 1, loss: 0.1204, accuracy: 96.40\n",
            " test epoch: 1, loss: 0.05265, accuracy: 98.35 norms: 6.23,12.56,30.02,3.94\n",
            "train epoch: 2, loss: 0.0411, accuracy: 98.73\n",
            " test epoch: 2, loss: 0.02823, accuracy: 99.03 norms: 6.41,14.52,38.72,4.47\n",
            "train epoch: 3, loss: 0.0282, accuracy: 99.14\n",
            " test epoch: 3, loss: 0.03171, accuracy: 98.93 norms: 6.51,16.09,46.38,4.98\n",
            "train epoch: 4, loss: 0.0218, accuracy: 99.31\n",
            " test epoch: 4, loss: 0.03209, accuracy: 99.03 norms: 6.58,17.56,53.63,5.50\n",
            "train epoch: 5, loss: 0.0149, accuracy: 99.53\n",
            " test epoch: 5, loss: 0.03353, accuracy: 98.99 norms: 6.66,19.01,60.54,5.99\n",
            "train epoch: 6, loss: 0.0117, accuracy: 99.64\n",
            " test epoch: 6, loss: 0.02716, accuracy: 99.15 norms: 6.72,20.27,66.42,6.45\n",
            "train epoch: 7, loss: 0.0103, accuracy: 99.64\n",
            " test epoch: 7, loss: 0.03238, accuracy: 98.98 norms: 6.79,21.55,72.74,6.92\n",
            "train epoch: 8, loss: 0.0082, accuracy: 99.74\n",
            " test epoch: 8, loss: 0.03077, accuracy: 99.14 norms: 6.83,22.61,77.58,7.26\n",
            "train epoch: 9, loss: 0.0064, accuracy: 99.80\n",
            " test epoch: 9, loss: 0.03668, accuracy: 99.10 norms: 6.88,23.60,82.24,7.65\n",
            "train epoch: 10, loss: 0.0051, accuracy: 99.82\n",
            " test epoch: 10, loss: 0.03901, accuracy: 98.99 norms: 6.93,24.73,86.90,8.02\n",
            "train epoch: 11, loss: 0.0053, accuracy: 99.82\n",
            " test epoch: 11, loss: 0.02993, accuracy: 99.22 norms: 6.97,25.53,91.03,8.32\n",
            "train epoch: 12, loss: 0.0041, accuracy: 99.88\n",
            " test epoch: 12, loss: 0.05639, accuracy: 98.84 norms: 7.01,26.28,94.46,8.63\n",
            "train epoch: 13, loss: 0.0034, accuracy: 99.87\n",
            " test epoch: 13, loss: 0.03908, accuracy: 99.21 norms: 7.03,27.06,98.16,8.93\n",
            "train epoch: 14, loss: 0.0039, accuracy: 99.89\n",
            " test epoch: 14, loss: 0.03689, accuracy: 99.27 norms: 7.05,27.94,101.68,9.23\n",
            "train epoch: 15, loss: 0.0038, accuracy: 99.88\n",
            " test epoch: 15, loss: 0.04385, accuracy: 99.07 norms: 7.07,28.74,105.05,9.52\n",
            "train epoch: 16, loss: 0.0041, accuracy: 99.88\n",
            " test epoch: 16, loss: 0.05672, accuracy: 98.95 norms: 7.06,29.34,108.10,9.75\n",
            "train epoch: 17, loss: 0.0026, accuracy: 99.91\n",
            " test epoch: 17, loss: 0.04520, accuracy: 99.15 norms: 7.10,29.87,110.83,9.98\n",
            "train epoch: 18, loss: 0.0030, accuracy: 99.91\n",
            " test epoch: 18, loss: 0.04570, accuracy: 99.03 norms: 7.12,30.81,113.86,10.25\n",
            "train epoch: 19, loss: 0.0022, accuracy: 99.93\n",
            " test epoch: 19, loss: 0.03520, accuracy: 99.29 norms: 7.16,31.27,116.29,10.46\n",
            "train epoch: 20, loss: 0.0017, accuracy: 99.95\n",
            " test epoch: 20, loss: 0.04372, accuracy: 99.23 norms: 7.18,31.79,118.53,10.70\n",
            "train epoch: 21, loss: 0.0032, accuracy: 99.90\n",
            " test epoch: 21, loss: 0.04395, accuracy: 99.20 norms: 7.20,32.60,121.98,11.03\n",
            "train epoch: 22, loss: 0.0031, accuracy: 99.91\n",
            " test epoch: 22, loss: 0.03308, accuracy: 99.36 norms: 7.17,33.14,124.30,11.20\n",
            "train epoch: 23, loss: 0.0022, accuracy: 99.93\n",
            " test epoch: 23, loss: 0.04960, accuracy: 99.19 norms: 7.23,33.82,126.62,11.41\n",
            "train epoch: 24, loss: 0.0022, accuracy: 99.95\n",
            " test epoch: 24, loss: 0.06666, accuracy: 98.79 norms: 7.25,34.36,128.53,11.59\n",
            "train epoch: 25, loss: 0.0022, accuracy: 99.93\n",
            " test epoch: 25, loss: 0.04628, accuracy: 99.18 norms: 7.25,34.76,130.92,11.81\n",
            " test epoch: 0, loss: 2.32381, accuracy: 5.53 norms: 5.70,7.98,16.00,3.11\n",
            "train epoch: 1, loss: 0.1406, accuracy: 95.75\n",
            " test epoch: 1, loss: 0.04840, accuracy: 98.61 norms: 9.33,14.48,28.13,10.70\n",
            "train epoch: 2, loss: 0.0513, accuracy: 98.44\n",
            " test epoch: 2, loss: 0.04643, accuracy: 98.57 norms: 10.71,16.44,31.67,11.84\n",
            "train epoch: 3, loss: 0.0409, accuracy: 98.75\n",
            " test epoch: 3, loss: 0.03667, accuracy: 98.89 norms: 11.74,18.11,34.54,12.59\n",
            "train epoch: 4, loss: 0.0257, accuracy: 99.21\n",
            " test epoch: 4, loss: 0.03089, accuracy: 99.13 norms: 12.51,19.15,36.44,13.27\n",
            "train epoch: 5, loss: 0.0227, accuracy: 99.33\n",
            " test epoch: 5, loss: 0.03069, accuracy: 99.07 norms: 12.98,20.06,38.31,13.80\n",
            "train epoch: 6, loss: 0.0200, accuracy: 99.39\n",
            " test epoch: 6, loss: 0.03531, accuracy: 99.01 norms: 13.52,20.94,40.12,14.33\n",
            "train epoch: 7, loss: 0.0170, accuracy: 99.48\n",
            " test epoch: 7, loss: 0.04346, accuracy: 98.79 norms: 13.95,21.79,41.75,14.85\n",
            "train epoch: 8, loss: 0.0127, accuracy: 99.60\n",
            " test epoch: 8, loss: 0.05188, accuracy: 98.88 norms: 14.41,22.49,42.95,15.33\n",
            "train epoch: 9, loss: 0.0143, accuracy: 99.55\n",
            " test epoch: 9, loss: 0.06706, accuracy: 98.61 norms: 14.83,23.27,44.50,15.74\n",
            "train epoch: 10, loss: 0.0101, accuracy: 99.68\n",
            " test epoch: 10, loss: 0.07081, accuracy: 98.33 norms: 15.34,23.93,45.70,16.22\n",
            "train epoch: 11, loss: 0.0123, accuracy: 99.60\n",
            " test epoch: 11, loss: 0.04227, accuracy: 98.89 norms: 15.79,24.68,47.18,16.60\n",
            "train epoch: 12, loss: 0.0149, accuracy: 99.57\n",
            " test epoch: 12, loss: 0.04783, accuracy: 98.93 norms: 16.12,25.46,48.86,16.92\n",
            "train epoch: 13, loss: 0.0119, accuracy: 99.65\n",
            " test epoch: 13, loss: 0.04114, accuracy: 99.08 norms: 16.50,26.08,50.17,17.26\n",
            "train epoch: 14, loss: 0.0112, accuracy: 99.66\n",
            " test epoch: 14, loss: 0.04129, accuracy: 98.97 norms: 16.87,26.72,51.43,17.57\n",
            "train epoch: 15, loss: 0.0130, accuracy: 99.64\n",
            " test epoch: 15, loss: 0.05911, accuracy: 98.72 norms: 17.18,27.43,52.82,17.85\n",
            "train epoch: 16, loss: 0.0075, accuracy: 99.79\n",
            " test epoch: 16, loss: 0.04183, accuracy: 99.06 norms: 17.42,27.80,53.54,18.03\n",
            "train epoch: 17, loss: 0.0043, accuracy: 99.87\n",
            " test epoch: 17, loss: 0.05256, accuracy: 99.14 norms: 17.64,28.07,54.05,18.23\n",
            "train epoch: 18, loss: 0.0120, accuracy: 99.67\n",
            " test epoch: 18, loss: 0.07481, accuracy: 98.60 norms: 18.11,28.88,55.72,18.50\n",
            "train epoch: 19, loss: 0.0124, accuracy: 99.69\n",
            " test epoch: 19, loss: 0.05153, accuracy: 98.96 norms: 18.38,29.45,56.96,18.67\n",
            "train epoch: 20, loss: 0.0045, accuracy: 99.86\n",
            " test epoch: 20, loss: 0.06548, accuracy: 98.92 norms: 18.63,29.75,57.49,18.86\n",
            "train epoch: 21, loss: 0.0119, accuracy: 99.71\n",
            " test epoch: 21, loss: 0.06305, accuracy: 98.89 norms: 18.99,30.49,58.85,19.07\n",
            "train epoch: 22, loss: 0.0149, accuracy: 99.62\n",
            " test epoch: 22, loss: 0.07052, accuracy: 98.84 norms: 19.59,31.35,60.52,19.31\n",
            "train epoch: 23, loss: 0.0216, accuracy: 99.44\n",
            " test epoch: 23, loss: 0.07519, accuracy: 98.72 norms: 20.24,32.57,62.95,19.65\n",
            "train epoch: 24, loss: 0.0117, accuracy: 99.69\n",
            " test epoch: 24, loss: 0.05753, accuracy: 98.90 norms: 20.52,33.16,64.11,19.92\n",
            "train epoch: 25, loss: 0.0109, accuracy: 99.72\n",
            " test epoch: 25, loss: 0.08891, accuracy: 98.45 norms: 20.84,33.73,65.25,20.09\n"
          ]
        }
      ],
      "source": [
        "import collections\n",
        "\n",
        "opt_lst = {\n",
        "    'adam': optax.adam(1e-3),\n",
        "    'sgd': optax.sgd(learning_rate, momentum),\n",
        "}\n",
        "metrics = collections.defaultdict(list)\n",
        "\n",
        "for opt_name, tx in opt_lst.items():\n",
        "  rng, init_rng = jax.random.split(init_rng)\n",
        "  state = create_train_state(rng, tx)\n",
        "  test_loss, test_accuracy = eval_model(state.params, test_ds)\n",
        "  norms = eval_spectral_norm(state.params)\n",
        "  metrics[opt_name].append(max(norms))\n",
        "  norms = ','.join([f'{norm:.2f}' for norm in norms])\n",
        "  print(f' test epoch: {0}, loss: {test_loss:.5f}, accuracy: {test_accuracy * 100:.2f} norms: {norms}')\n",
        "  for epoch in range(1, num_epochs + 1):\n",
        "    # Use a separate PRNG key to permute image data during shuffling\n",
        "    rng, input_rng = jax.random.split(rng)\n",
        "    # Run an optimization step over a training batch\n",
        "    state = train_epoch(state, train_ds, batch_size, epoch, input_rng)\n",
        "    # Evaluate on the test set after each training epoch \n",
        "    test_loss, test_accuracy = eval_model(state.params, test_ds)\n",
        "    norms = eval_spectral_norm(state.params)\n",
        "    metrics[opt_name].append(max(norms))\n",
        "    norms = ','.join([f'{norm:.2f}' for norm in norms])\n",
        "    print(f' test epoch: {epoch}, loss: {test_loss:.5f}, accuracy: {test_accuracy * 100:.2f} norms: {norms}')"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pandas as pd\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt"
      ],
      "metadata": {
        "id": "OocjzbqGhswR"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.DataFrame.from_dict(metrics)\n",
        "df"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 865
        },
        "id": "Lwzx7c0hhybj",
        "outputId": "0953ac08-c35e-46fb-d500-682421b3b4e3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          adam        sgd\n",
              "0    16.000542  15.997117\n",
              "1    30.018700  28.129322\n",
              "2    38.716732  31.667624\n",
              "3    46.378811  34.543392\n",
              "4    53.627213  36.436890\n",
              "5    60.544743  38.306351\n",
              "6    66.419792  40.115341\n",
              "7    72.736778  41.754349\n",
              "8    77.584129  42.954498\n",
              "9    82.240593  44.504028\n",
              "10   86.896568  45.698036\n",
              "11   91.030251  47.183315\n",
              "12   94.460213  48.857704\n",
              "13   98.156151  50.165382\n",
              "14  101.683945  51.425743\n",
              "15  105.046448  52.823753\n",
              "16  108.097076  53.540913\n",
              "17  110.833031  54.051525\n",
              "18  113.864517  55.717289\n",
              "19  116.291763  56.955685\n",
              "20  118.534225  57.486092\n",
              "21  121.980392  58.846817\n",
              "22  124.302681  60.519264\n",
              "23  126.617371  62.945007\n",
              "24  128.531158  64.108963\n",
              "25  130.917816  65.246796"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fad7d99a-d67f-437c-a951-944e39584b74\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>adam</th>\n",
              "      <th>sgd</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>16.000542</td>\n",
              "      <td>15.997117</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>30.018700</td>\n",
              "      <td>28.129322</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>38.716732</td>\n",
              "      <td>31.667624</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>46.378811</td>\n",
              "      <td>34.543392</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>53.627213</td>\n",
              "      <td>36.436890</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>60.544743</td>\n",
              "      <td>38.306351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>66.419792</td>\n",
              "      <td>40.115341</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>72.736778</td>\n",
              "      <td>41.754349</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>77.584129</td>\n",
              "      <td>42.954498</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>82.240593</td>\n",
              "      <td>44.504028</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>86.896568</td>\n",
              "      <td>45.698036</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>91.030251</td>\n",
              "      <td>47.183315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>94.460213</td>\n",
              "      <td>48.857704</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>98.156151</td>\n",
              "      <td>50.165382</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>101.683945</td>\n",
              "      <td>51.425743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>105.046448</td>\n",
              "      <td>52.823753</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>108.097076</td>\n",
              "      <td>53.540913</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>110.833031</td>\n",
              "      <td>54.051525</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>113.864517</td>\n",
              "      <td>55.717289</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>116.291763</td>\n",
              "      <td>56.955685</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>118.534225</td>\n",
              "      <td>57.486092</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>121.980392</td>\n",
              "      <td>58.846817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>124.302681</td>\n",
              "      <td>60.519264</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>126.617371</td>\n",
              "      <td>62.945007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>128.531158</td>\n",
              "      <td>64.108963</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>130.917816</td>\n",
              "      <td>65.246796</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fad7d99a-d67f-437c-a951-944e39584b74')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-fad7d99a-d67f-437c-a951-944e39584b74 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-fad7d99a-d67f-437c-a951-944e39584b74');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "index = 'Epoch'\n",
        "vals = r'Spectral Norm $\\mathcal{M}^t$'\n",
        "cols = 'Optimizer'\n",
        "df[index] = df.index\n",
        "num_bits = 64 if use_float64 else 32\n",
        "df = df.rename(columns={'sgd':f'SGD {num_bits} bits', 'adam':f'Adam {num_bits} bits'})\n",
        "dfm = df.melt(index, var_name=cols, value_name=vals)\n",
        "dfm"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "IgldoHw3h5Db",
        "outputId": "765e8dfc-2b1f-49b6-f28c-be46299e572a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    Epoch     Optimizer  Spectral Norm $\\mathcal{M}^t$\n",
              "0       0  Adam 64 bits                      16.000542\n",
              "1       1  Adam 64 bits                      30.018700\n",
              "2       2  Adam 64 bits                      38.716732\n",
              "3       3  Adam 64 bits                      46.378811\n",
              "4       4  Adam 64 bits                      53.627213\n",
              "5       5  Adam 64 bits                      60.544743\n",
              "6       6  Adam 64 bits                      66.419792\n",
              "7       7  Adam 64 bits                      72.736778\n",
              "8       8  Adam 64 bits                      77.584129\n",
              "9       9  Adam 64 bits                      82.240593\n",
              "10     10  Adam 64 bits                      86.896568\n",
              "11     11  Adam 64 bits                      91.030251\n",
              "12     12  Adam 64 bits                      94.460213\n",
              "13     13  Adam 64 bits                      98.156151\n",
              "14     14  Adam 64 bits                     101.683945\n",
              "15     15  Adam 64 bits                     105.046448\n",
              "16     16  Adam 64 bits                     108.097076\n",
              "17     17  Adam 64 bits                     110.833031\n",
              "18     18  Adam 64 bits                     113.864517\n",
              "19     19  Adam 64 bits                     116.291763\n",
              "20     20  Adam 64 bits                     118.534225\n",
              "21     21  Adam 64 bits                     121.980392\n",
              "22     22  Adam 64 bits                     124.302681\n",
              "23     23  Adam 64 bits                     126.617371\n",
              "24     24  Adam 64 bits                     128.531158\n",
              "25     25  Adam 64 bits                     130.917816\n",
              "26      0   SGD 64 bits                      15.997117\n",
              "27      1   SGD 64 bits                      28.129322\n",
              "28      2   SGD 64 bits                      31.667624\n",
              "29      3   SGD 64 bits                      34.543392\n",
              "30      4   SGD 64 bits                      36.436890\n",
              "31      5   SGD 64 bits                      38.306351\n",
              "32      6   SGD 64 bits                      40.115341\n",
              "33      7   SGD 64 bits                      41.754349\n",
              "34      8   SGD 64 bits                      42.954498\n",
              "35      9   SGD 64 bits                      44.504028\n",
              "36     10   SGD 64 bits                      45.698036\n",
              "37     11   SGD 64 bits                      47.183315\n",
              "38     12   SGD 64 bits                      48.857704\n",
              "39     13   SGD 64 bits                      50.165382\n",
              "40     14   SGD 64 bits                      51.425743\n",
              "41     15   SGD 64 bits                      52.823753\n",
              "42     16   SGD 64 bits                      53.540913\n",
              "43     17   SGD 64 bits                      54.051525\n",
              "44     18   SGD 64 bits                      55.717289\n",
              "45     19   SGD 64 bits                      56.955685\n",
              "46     20   SGD 64 bits                      57.486092\n",
              "47     21   SGD 64 bits                      58.846817\n",
              "48     22   SGD 64 bits                      60.519264\n",
              "49     23   SGD 64 bits                      62.945007\n",
              "50     24   SGD 64 bits                      64.108963\n",
              "51     25   SGD 64 bits                      65.246796"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e975c41c-d424-4097-b9ce-b451e9edf7be\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Epoch</th>\n",
              "      <th>Optimizer</th>\n",
              "      <th>Spectral Norm $\\mathcal{M}^t$</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>16.000542</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>30.018700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>38.716732</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>46.378811</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>53.627213</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>5</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>60.544743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>6</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>66.419792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>7</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>72.736778</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>8</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>77.584129</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>9</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>82.240593</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>10</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>86.896568</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>11</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>91.030251</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>12</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>94.460213</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>13</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>98.156151</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>14</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>101.683945</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>15</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>105.046448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>16</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>108.097076</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>17</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>110.833031</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>18</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>113.864517</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>19</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>116.291763</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>20</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>118.534225</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>21</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>121.980392</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>22</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>124.302681</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>23</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>126.617371</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>24</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>128.531158</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>25</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>130.917816</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>0</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>15.997117</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>1</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>28.129322</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>2</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>31.667624</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>3</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>34.543392</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>4</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>36.436890</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>5</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>38.306351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>6</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>40.115341</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>7</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>41.754349</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>8</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>42.954498</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>9</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>44.504028</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>10</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>45.698036</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>11</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>47.183315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>12</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>48.857704</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39</th>\n",
              "      <td>13</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>50.165382</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>14</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>51.425743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>15</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>52.823753</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>16</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>53.540913</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>17</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>54.051525</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>18</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>55.717289</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>19</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>56.955685</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>20</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>57.486092</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>21</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>58.846817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>22</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>60.519264</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>23</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>62.945007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>24</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>64.108963</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>25</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>65.246796</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e975c41c-d424-4097-b9ce-b451e9edf7be')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e975c41c-d424-4097-b9ce-b451e9edf7be button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e975c41c-d424-4097-b9ce-b451e9edf7be');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(22*0.35, 7*0.35), dpi=300)\n",
        "g = sns.lineplot(x=index, y=vals, hue=cols, data=dfm)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "4l7-qibui4OT",
        "outputId": "14faf7cb-7b28-4e91-d6be-b4c13b34a8e8"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 2310x735 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dfm.to_csv(f'divergence_{num_bits}.csv')"
      ],
      "metadata": {
        "id": "zYrQvrCOkb6h"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Merge float32 and float64"
      ],
      "metadata": {
        "id": "hwryPJJ2SYXs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "dfm32 = pd.read_csv('divergence_32.csv')\n",
        "dfm32 = dfm32.drop(columns=dfm32.columns[0])\n",
        "dfm32"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "AliMGpYGrKKa",
        "outputId": "687d4783-c66e-4103-dea6-a0c4fd1679eb"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    Epoch     Optimizer  Spectral Norm $\\mathcal{M}^t$\n",
              "0       0  Adam 32 bits                      16.000542\n",
              "1       1  Adam 32 bits                      30.324541\n",
              "2       2  Adam 32 bits                      39.110847\n",
              "3       3  Adam 32 bits                      46.799461\n",
              "4       4  Adam 32 bits                      53.995857\n",
              "5       5  Adam 32 bits                      60.609196\n",
              "6       6  Adam 32 bits                      67.160538\n",
              "7       7  Adam 32 bits                      72.790085\n",
              "8       8  Adam 32 bits                      78.248039\n",
              "9       9  Adam 32 bits                      83.362534\n",
              "10     10  Adam 32 bits                      87.579994\n",
              "11     11  Adam 32 bits                      92.458290\n",
              "12     12  Adam 32 bits                      96.275658\n",
              "13     13  Adam 32 bits                     100.106903\n",
              "14     14  Adam 32 bits                     103.636566\n",
              "15     15  Adam 32 bits                     106.705444\n",
              "16     16  Adam 32 bits                     109.947365\n",
              "17     17  Adam 32 bits                     112.859909\n",
              "18     18  Adam 32 bits                     115.064293\n",
              "19     19  Adam 32 bits                     119.020874\n",
              "20     20  Adam 32 bits                     120.998405\n",
              "21     21  Adam 32 bits                     123.794853\n",
              "22     22  Adam 32 bits                     126.488556\n",
              "23     23  Adam 32 bits                     127.919289\n",
              "24     24  Adam 32 bits                     130.712921\n",
              "25     25  Adam 32 bits                     132.860794\n",
              "26      0   SGD 32 bits                      15.997117\n",
              "27      1   SGD 32 bits                      28.052641\n",
              "28      2   SGD 32 bits                      31.984524\n",
              "29      3   SGD 32 bits                      34.430344\n",
              "30      4   SGD 32 bits                      36.623337\n",
              "31      5   SGD 32 bits                      38.541431\n",
              "32      6   SGD 32 bits                      40.228794\n",
              "33      7   SGD 32 bits                      41.909302\n",
              "34      8   SGD 32 bits                      43.694351\n",
              "35      9   SGD 32 bits                      45.016376\n",
              "36     10   SGD 32 bits                      46.591915\n",
              "37     11   SGD 32 bits                      48.045300\n",
              "38     12   SGD 32 bits                      49.242592\n",
              "39     13   SGD 32 bits                      50.212055\n",
              "40     14   SGD 32 bits                      51.885036\n",
              "41     15   SGD 32 bits                      54.390228\n",
              "42     16   SGD 32 bits                      55.528408\n",
              "43     17   SGD 32 bits                      56.383167\n",
              "44     18   SGD 32 bits                      57.773670\n",
              "45     19   SGD 32 bits                      58.836021\n",
              "46     20   SGD 32 bits                      59.666168\n",
              "47     21   SGD 32 bits                      60.684925\n",
              "48     22   SGD 32 bits                      62.115803\n",
              "49     23   SGD 32 bits                      63.598408\n",
              "50     24   SGD 32 bits                      65.401291\n",
              "51     25   SGD 32 bits                      66.728004"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-5e4b8c82-4c6f-4c48-8bbc-76442a89f2dd\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Epoch</th>\n",
              "      <th>Optimizer</th>\n",
              "      <th>Spectral Norm $\\mathcal{M}^t$</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>16.000542</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>30.324541</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>39.110847</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>46.799461</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>53.995857</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>5</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>60.609196</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>6</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>67.160538</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>7</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>72.790085</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>8</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>78.248039</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>9</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>83.362534</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>10</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>87.579994</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>11</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>92.458290</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>12</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>96.275658</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>13</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>100.106903</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>14</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>103.636566</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>15</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>106.705444</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>16</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>109.947365</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>17</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>112.859909</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>18</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>115.064293</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>19</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>119.020874</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>20</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>120.998405</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>21</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>123.794853</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>22</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>126.488556</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>23</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>127.919289</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>24</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>130.712921</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>25</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>132.860794</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>0</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>15.997117</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>1</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>28.052641</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>2</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>31.984524</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>3</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>34.430344</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>4</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>36.623337</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>5</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>38.541431</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>6</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>40.228794</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>7</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>41.909302</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>8</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>43.694351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>9</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>45.016376</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>10</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>46.591915</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>11</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>48.045300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>12</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>49.242592</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39</th>\n",
              "      <td>13</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>50.212055</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>14</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>51.885036</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>15</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>54.390228</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>16</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>55.528408</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>17</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>56.383167</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>18</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>57.773670</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>19</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>58.836021</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>20</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>59.666168</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>21</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>60.684925</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>22</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>62.115803</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>23</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>63.598408</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>24</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>65.401291</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>25</td>\n",
              "      <td>SGD 32 bits</td>\n",
              "      <td>66.728004</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5e4b8c82-4c6f-4c48-8bbc-76442a89f2dd')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-5e4b8c82-4c6f-4c48-8bbc-76442a89f2dd button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-5e4b8c82-4c6f-4c48-8bbc-76442a89f2dd');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dfm64 = pd.read_csv('divergence_64.csv')\n",
        "dfm64 = dfm64.drop(columns=dfm64.columns[0])\n",
        "dfm64"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "RDUPGGrwSfVx",
        "outputId": "40d35e7f-706c-4a1e-e864-cbd38efc2788"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    Epoch     Optimizer  Spectral Norm $\\mathcal{M}^t$\n",
              "0       0  Adam 64 bits                      16.000542\n",
              "1       1  Adam 64 bits                      30.018700\n",
              "2       2  Adam 64 bits                      38.716732\n",
              "3       3  Adam 64 bits                      46.378811\n",
              "4       4  Adam 64 bits                      53.627213\n",
              "5       5  Adam 64 bits                      60.544743\n",
              "6       6  Adam 64 bits                      66.419792\n",
              "7       7  Adam 64 bits                      72.736778\n",
              "8       8  Adam 64 bits                      77.584129\n",
              "9       9  Adam 64 bits                      82.240593\n",
              "10     10  Adam 64 bits                      86.896568\n",
              "11     11  Adam 64 bits                      91.030251\n",
              "12     12  Adam 64 bits                      94.460213\n",
              "13     13  Adam 64 bits                      98.156151\n",
              "14     14  Adam 64 bits                     101.683945\n",
              "15     15  Adam 64 bits                     105.046448\n",
              "16     16  Adam 64 bits                     108.097076\n",
              "17     17  Adam 64 bits                     110.833031\n",
              "18     18  Adam 64 bits                     113.864517\n",
              "19     19  Adam 64 bits                     116.291763\n",
              "20     20  Adam 64 bits                     118.534225\n",
              "21     21  Adam 64 bits                     121.980392\n",
              "22     22  Adam 64 bits                     124.302681\n",
              "23     23  Adam 64 bits                     126.617371\n",
              "24     24  Adam 64 bits                     128.531158\n",
              "25     25  Adam 64 bits                     130.917816\n",
              "26      0   SGD 64 bits                      15.997117\n",
              "27      1   SGD 64 bits                      28.129322\n",
              "28      2   SGD 64 bits                      31.667624\n",
              "29      3   SGD 64 bits                      34.543392\n",
              "30      4   SGD 64 bits                      36.436890\n",
              "31      5   SGD 64 bits                      38.306351\n",
              "32      6   SGD 64 bits                      40.115341\n",
              "33      7   SGD 64 bits                      41.754349\n",
              "34      8   SGD 64 bits                      42.954498\n",
              "35      9   SGD 64 bits                      44.504028\n",
              "36     10   SGD 64 bits                      45.698036\n",
              "37     11   SGD 64 bits                      47.183315\n",
              "38     12   SGD 64 bits                      48.857704\n",
              "39     13   SGD 64 bits                      50.165382\n",
              "40     14   SGD 64 bits                      51.425743\n",
              "41     15   SGD 64 bits                      52.823753\n",
              "42     16   SGD 64 bits                      53.540913\n",
              "43     17   SGD 64 bits                      54.051525\n",
              "44     18   SGD 64 bits                      55.717289\n",
              "45     19   SGD 64 bits                      56.955685\n",
              "46     20   SGD 64 bits                      57.486092\n",
              "47     21   SGD 64 bits                      58.846817\n",
              "48     22   SGD 64 bits                      60.519264\n",
              "49     23   SGD 64 bits                      62.945007\n",
              "50     24   SGD 64 bits                      64.108963\n",
              "51     25   SGD 64 bits                      65.246796"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1694b188-dd45-4fb0-95ad-7460418862e4\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Epoch</th>\n",
              "      <th>Optimizer</th>\n",
              "      <th>Spectral Norm $\\mathcal{M}^t$</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>16.000542</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>30.018700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>38.716732</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>46.378811</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>53.627213</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>5</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>60.544743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>6</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>66.419792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>7</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>72.736778</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>8</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>77.584129</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>9</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>82.240593</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>10</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>86.896568</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>11</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>91.030251</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>12</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>94.460213</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>13</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>98.156151</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>14</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>101.683945</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>15</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>105.046448</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>16</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>108.097076</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>17</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>110.833031</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>18</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>113.864517</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>19</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>116.291763</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>20</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>118.534225</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>21</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>121.980392</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>22</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>124.302681</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>23</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>126.617371</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>24</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>128.531158</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>25</td>\n",
              "      <td>Adam 64 bits</td>\n",
              "      <td>130.917816</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>0</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>15.997117</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>1</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>28.129322</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>2</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>31.667624</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>3</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>34.543392</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>4</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>36.436890</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>5</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>38.306351</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>6</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>40.115341</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>7</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>41.754349</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>8</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>42.954498</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>9</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>44.504028</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>10</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>45.698036</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>11</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>47.183315</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>12</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>48.857704</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39</th>\n",
              "      <td>13</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>50.165382</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>14</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>51.425743</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>15</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>52.823753</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>16</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>53.540913</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>17</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>54.051525</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>18</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>55.717289</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>19</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>56.955685</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>20</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>57.486092</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>21</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>58.846817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>22</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>60.519264</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>23</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>62.945007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>24</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>64.108963</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>25</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>65.246796</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1694b188-dd45-4fb0-95ad-7460418862e4')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1694b188-dd45-4fb0-95ad-7460418862e4 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1694b188-dd45-4fb0-95ad-7460418862e4');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dfm_mixed = pd.concat([dfm32, dfm64], axis=0, ignore_index=True)\n",
        "dfm_mixed"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 423
        },
        "id": "wyIQOB3MLpHq",
        "outputId": "9f9f8a16-aa1e-4536-bb15-b956e6ddd243"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "     Epoch     Optimizer  Spectral Norm $\\mathcal{M}^t$\n",
              "0        0  Adam 32 bits                      16.000542\n",
              "1        1  Adam 32 bits                      30.324541\n",
              "2        2  Adam 32 bits                      39.110847\n",
              "3        3  Adam 32 bits                      46.799461\n",
              "4        4  Adam 32 bits                      53.995857\n",
              "..     ...           ...                            ...\n",
              "99      21   SGD 64 bits                      58.846817\n",
              "100     22   SGD 64 bits                      60.519264\n",
              "101     23   SGD 64 bits                      62.945007\n",
              "102     24   SGD 64 bits                      64.108963\n",
              "103     25   SGD 64 bits                      65.246796\n",
              "\n",
              "[104 rows x 3 columns]"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-445894a5-369f-4fc2-ad43-197bb2f4ba23\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Epoch</th>\n",
              "      <th>Optimizer</th>\n",
              "      <th>Spectral Norm $\\mathcal{M}^t$</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>16.000542</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>30.324541</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>2</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>39.110847</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>3</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>46.799461</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>4</td>\n",
              "      <td>Adam 32 bits</td>\n",
              "      <td>53.995857</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>21</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>58.846817</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>100</th>\n",
              "      <td>22</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>60.519264</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>101</th>\n",
              "      <td>23</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>62.945007</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>102</th>\n",
              "      <td>24</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>64.108963</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>103</th>\n",
              "      <td>25</td>\n",
              "      <td>SGD 64 bits</td>\n",
              "      <td>65.246796</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>104 rows × 3 columns</p>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-445894a5-369f-4fc2-ad43-197bb2f4ba23')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-445894a5-369f-4fc2-ad43-197bb2f4ba23 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-445894a5-369f-4fc2-ad43-197bb2f4ba23');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "fig, ax = plt.subplots(figsize=(22*0.35, 7*0.35), dpi=300)\n",
        "g = sns.lineplot(x=index, y=vals, hue=cols, data=dfm_mixed)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 295
        },
        "id": "Y-zJ3h4TMBGS",
        "outputId": "f24cca37-6c4c-421a-89bd-c8d2eb9273c5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 2310x735 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        ""
      ],
      "metadata": {
        "id": "HEw7vrxrMLG_"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "collapsed_sections": [],
      "name": "Mnist_divergence.ipynb",
      "provenance": []
    },
    "interpreter": {
      "hash": "50100d07a2a27af6847cdedde67ae5e97c9798a9e1a9ae21eed9ecf69a9f619c"
    },
    "jupytext": {
      "formats": "ipynb,md:myst",
      "main_language": "python"
    },
    "kernelspec": {
      "display_name": "Python 3.8.11 ('.venv': venv)",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "name": "python",
      "version": "3.8.11"
    },
    "gpuClass": "standard"
  },
  "nbformat": 4,
  "nbformat_minor": 0
}