ExploringHiDimLatentSpace_Generate_LS_with_CVAE.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "name": "ExploringHiDimLatentSpace - Generate LS with CVAE",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Hackathorn/ExploringHiDimSpaces/blob/master/ExploringHiDimLatentSpace_Generate_LS_with_CVAE.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "Ndo4ERqnwQOU"
      },
      "source": [
        "##### Copyright 2020 Bolder Technology, Inc. which was derived from https://www.tensorflow.org/tutorials/generative/cvae\n",
        "##### Copyright 2018 The TensorFlow Authors."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "cellView": "form",
        "colab_type": "code",
        "id": "MTKwbguKwT4R",
        "colab": {}
      },
      "source": [
        "#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
        "# you may not use this file except in compliance with the License.\n",
        "# You may obtain a copy of the License at\n",
        "#\n",
        "# https://www.apache.org/licenses/LICENSE-2.0\n",
        "#\n",
        "# Unless required by applicable law or agreed to in writing, software\n",
        "# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
        "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
        "# See the License for the specific language governing permissions and\n",
        "# limitations under the License."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "xfNT-mlFwxVM"
      },
      "source": [
        "# Convolutional Variational Autoencoder Using the MNIST Handwritten Numbers"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "ITZuApL56Mny"
      },
      "source": [
        "![evolution of output during training](https://tensorflow.org/images/autoencoders/cvae.gif)\n",
        "\n",
        "This notebook demonstrates how to generate images of handwritten digits by training a Variational Autoencoder ([1](https://arxiv.org/abs/1312.6114), [2](https://arxiv.org/abs/1401.4082)).\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "P-JuIu2N_SQf",
        "colab": {}
      },
      "source": [
        "# to generate gifs\n",
        "!pip install imageio"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "e1_Y75QXJS6h"
      },
      "source": [
        "## Import TensorFlow and other libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "YfIk2es3hJEd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "outputId": "0028131f-b5f9-4007-e460-7d8f9a866616"
      },
      "source": [
        "from __future__ import absolute_import, division, print_function, unicode_literals\n",
        "\n",
        "try:\n",
        "  # %tensorflow_version only exists in Colab.\n",
        "  %tensorflow_version 2.x\n",
        "except Exception:\n",
        "  pass\n",
        "import tensorflow as tf\n",
        "\n",
        "import os\n",
        "import time\n",
        "import numpy as np\n",
        "import glob\n",
        "import matplotlib.pyplot as plt\n",
        "import PIL\n",
        "import imageio\n",
        "\n",
        "from IPython import display"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "TensorFlow 2.x selected.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "iYn4MdZnKCey"
      },
      "source": [
        "## Load the MNIST dataset\n",
        "Each MNIST image is originally a vector of 784 integers, each of which is between 0-255 and represents the intensity of a pixel. We model each pixel with a Bernoulli distribution in our model, and we statically binarize the dataset."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "a4fYMGxGhrna",
        "colab": {}
      },
      "source": [
        "(train_images, _), (test_images, _) = tf.keras.datasets.mnist.load_data()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "L-dm_Qt09ui7",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 294
        },
        "outputId": "705208fd-bf93-46e8-94d6-698edde01b6c"
      },
      "source": [
        "# before binarizing the pixels, what is the intensity distribution\n",
        "b = {}\n",
        "for pixel in train_images.flatten():\n",
        "  b[pixel] = b.get(pixel, 0) + 1\n",
        "\n",
        "b[0] = 0  # pixel = 0 has high count\n",
        "plt.bar(b.keys(), b.values(), width=5, color='g')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<BarContainer object of 256 artists>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAEDCAYAAAAlRP8qAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\npEYY6JLUCANdkhrxv17N2eCRH9viAAAAAElFTkSuQmCC\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "NFC2ghIdiZYE",
        "colab": {}
      },
      "source": [
        "train_images = train_images.reshape(train_images.shape[0], 28, 28, 1).astype('float32')\n",
        "test_images = test_images.reshape(test_images.shape[0], 28, 28, 1).astype('float32')\n",
        "\n",
        "# Normalizing the images to the range of [0., 1.]\n",
        "train_images /= 255.\n",
        "test_images /= 255.\n",
        "\n",
        "# Binarization\n",
        "train_images[train_images >= .5] = 1.\n",
        "train_images[train_images < .5] = 0.\n",
        "test_images[test_images >= .5] = 1.\n",
        "test_images[test_images < .5] = 0."
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "S4PIDhoDLbsZ",
        "colab": {}
      },
      "source": [
        "TRAIN_BUF = 60000\n",
        "BATCH_SIZE = 100\n",
        "\n",
        "TEST_BUF = 10000"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "PIGN6ouoQxt3"
      },
      "source": [
        "## Use *tf.data* to create batches and shuffle the dataset"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "-yKCCQOoJ7cn",
        "colab": {}
      },
      "source": [
        "train_dataset = tf.data.Dataset.from_tensor_slices(train_images).shuffle(TRAIN_BUF).batch(BATCH_SIZE)\n",
        "test_dataset = tf.data.Dataset.from_tensor_slices(test_images).shuffle(TEST_BUF).batch(BATCH_SIZE)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "THY-sZMiQ4UV"
      },
      "source": [
        "## Wire up the generative and inference network with *tf.keras.Sequential*\n",
        "\n",
        "In our VAE example, we use two small ConvNets for the generative and inference network. Since these neural nets are small, we use `tf.keras.Sequential` to simplify our code. Let $x$ and $z$ denote the observation and latent variable respectively in the following descriptions.\n",
        "\n",
        "### Generative Network\n",
        "This defines the generative model which takes a latent encoding as input, and outputs the parameters for a conditional distribution of the observation, i.e. $p(x|z)$. Additionally, we use a unit Gaussian prior $p(z)$ for the latent variable.\n",
        "\n",
        "### Inference Network\n",
        "This defines an approximate posterior distribution $q(z|x)$, which takes as input an observation and outputs a set of parameters for the conditional distribution of the latent representation. In this example, we simply model this distribution as a diagonal Gaussian. In this case, the inference network outputs the mean and log-variance parameters of a factorized Gaussian (log-variance instead of the variance directly is for numerical stability).\n",
        "\n",
        "### Reparameterization Trick\n",
        "During optimization, we can sample from $q(z|x)$ by first sampling from a unit Gaussian, and then multiplying by the standard deviation and adding the mean. This ensures the gradients could pass through the sample to the inference network parameters.\n",
        "\n",
        "### Network architecture\n",
        "For the inference network, we use two convolutional layers followed by a fully-connected layer. In the generative network, we mirror this architecture by using a fully-connected layer followed by three convolution transpose layers (a.k.a. deconvolutional layers in some contexts). Note, it's common practice to avoid using batch normalization when training VAEs, since the additional stochasticity due to using mini-batches may aggravate instability on top of the stochasticity from sampling."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "VGLbvBEmjK0a",
        "colab": {}
      },
      "source": [
        "class CVAE(tf.keras.Model):\n",
        "  def __init__(self, latent_dim):\n",
        "    super(CVAE, self).__init__()\n",
        "    self.latent_dim = latent_dim\n",
        "    self.inference_net = tf.keras.Sequential(\n",
        "      [\n",
        "          tf.keras.layers.InputLayer(input_shape=(28, 28, 1)),\n",
        "          tf.keras.layers.Conv2D(\n",
        "              filters=32, kernel_size=3, strides=(2, 2), activation='relu'),\n",
        "          tf.keras.layers.Conv2D(\n",
        "              filters=64, kernel_size=3, strides=(2, 2), activation='relu'),\n",
        "          tf.keras.layers.Flatten(),\n",
        "          # No activation\n",
        "          tf.keras.layers.Dense(latent_dim + latent_dim),\n",
        "      ]\n",
        "    )\n",
        "\n",
        "    self.generative_net = tf.keras.Sequential(\n",
        "        [\n",
        "          tf.keras.layers.InputLayer(input_shape=(latent_dim,)),\n",
        "          tf.keras.layers.Dense(units=7*7*32, activation=tf.nn.relu),\n",
        "          tf.keras.layers.Reshape(target_shape=(7, 7, 32)),\n",
        "          tf.keras.layers.Conv2DTranspose(\n",
        "              filters=64,\n",
        "              kernel_size=3,\n",
        "              strides=(2, 2),\n",
        "              padding=\"SAME\",\n",
        "              activation='relu'),\n",
        "          tf.keras.layers.Conv2DTranspose(\n",
        "              filters=32,\n",
        "              kernel_size=3,\n",
        "              strides=(2, 2),\n",
        "              padding=\"SAME\",\n",
        "              activation='relu'),\n",
        "          # No activation\n",
        "          tf.keras.layers.Conv2DTranspose(\n",
        "              filters=1, kernel_size=3, strides=(1, 1), padding=\"SAME\"),\n",
        "        ]\n",
        "    )\n",
        "\n",
        "  @tf.function\n",
        "  def sample(self, eps=None):\n",
        "    if eps is None:\n",
        "      eps = tf.random.normal(shape=(100, self.latent_dim))\n",
        "    return self.decode(eps, apply_sigmoid=True)\n",
        "\n",
        "  def encode(self, x):\n",
        "    mean, logvar = tf.split(self.inference_net(x), num_or_size_splits=2, axis=1)\n",
        "    return mean, logvar\n",
        "\n",
        "  def reparameterize(self, mean, logvar):\n",
        "    eps = tf.random.normal(shape=mean.shape)\n",
        "    return eps * tf.exp(logvar * .5) + mean\n",
        "\n",
        "  def decode(self, z, apply_sigmoid=False):\n",
        "    logits = self.generative_net(z)\n",
        "    if apply_sigmoid:\n",
        "      probs = tf.sigmoid(logits)\n",
        "      return probs\n",
        "\n",
        "    return logits"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "0FMYgY_mPfTi"
      },
      "source": [
        "## Define the loss function and the optimizer\n",
        "\n",
        "VAEs train by maximizing the evidence lower bound (ELBO) on the marginal log-likelihood:\n",
        "\n",
        "$$\\log p(x) \\ge \\text{ELBO} = \\mathbb{E}_{q(z|x)}\\left[\\log \\frac{p(x, z)}{q(z|x)}\\right].$$\n",
        "\n",
        "In practice, we optimize the single sample Monte Carlo estimate of this expectation:\n",
        "\n",
        "$$\\log p(x| z) + \\log p(z) - \\log q(z|x),$$\n",
        "where $z$ is sampled from $q(z|x)$.\n",
        "\n",
        "**Note**: we could also analytically compute the KL term, but here we incorporate all three terms in the Monte Carlo estimator for simplicity."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "iWCn_PVdEJZ7",
        "colab": {}
      },
      "source": [
        "optimizer = tf.keras.optimizers.Adam(1e-4)\n",
        "\n",
        "def log_normal_pdf(sample, mean, logvar, raxis=1):\n",
        "  log2pi = tf.math.log(2. * np.pi)\n",
        "  return tf.reduce_sum(\n",
        "      -.5 * ((sample - mean) ** 2. * tf.exp(-logvar) + logvar + log2pi),\n",
        "      axis=raxis)\n",
        "\n",
        "@tf.function\n",
        "def compute_loss(model, x):\n",
        "  mean, logvar = model.encode(x)\n",
        "  z = model.reparameterize(mean, logvar)\n",
        "  x_logit = model.decode(z)\n",
        "\n",
        "  cross_ent = tf.nn.sigmoid_cross_entropy_with_logits(logits=x_logit, labels=x)\n",
        "  logpx_z = -tf.reduce_sum(cross_ent, axis=[1, 2, 3])\n",
        "  logpz = log_normal_pdf(z, 0., 0.)\n",
        "  logqz_x = log_normal_pdf(z, mean, logvar)\n",
        "  return -tf.reduce_mean(logpx_z + logpz - logqz_x)\n",
        "\n",
        "@tf.function\n",
        "def compute_apply_gradients(model, x, optimizer):\n",
        "  with tf.GradientTape() as tape:\n",
        "    loss = compute_loss(model, x)\n",
        "  gradients = tape.gradient(loss, model.trainable_variables)\n",
        "  optimizer.apply_gradients(zip(gradients, model.trainable_variables))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "Rw1fkAczTQYh"
      },
      "source": [
        "## Training\n",
        "\n",
        "* We start by iterating over the dataset\n",
        "* During each iteration, we pass the image to the encoder to obtain a set of mean and log-variance parameters of the approximate posterior $q(z|x)$\n",
        "* We then apply the *reparameterization trick* to sample from $q(z|x)$\n",
        "* Finally, we pass the reparameterized samples to the decoder to obtain the logits of the generative distribution $p(x|z)$\n",
        "* **Note:** Since we use the dataset loaded by keras with 60k datapoints in the training set and 10k datapoints in the test set, our resulting ELBO on the test set is slightly higher than reported results in the literature which uses dynamic binarization of Larochelle's MNIST.\n",
        "\n",
        "## Generate Images\n",
        "\n",
        "* After training, it is time to generate some images\n",
        "* We start by sampling a set of latent vectors from the unit Gaussian prior distribution $p(z)$\n",
        "* The generator will then convert the latent sample $z$ to logits of the observation, giving a distribution $p(x|z)$\n",
        "* Here we plot the probabilities of Bernoulli distributions\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "NS2GWywBbAWo",
        "colab": {}
      },
      "source": [
        "epochs = 100\n",
        "latent_dim = 50\n",
        "num_examples_to_generate = 16\n",
        "\n",
        "# keeping the random vector constant for generation (prediction) so\n",
        "# it will be easier to see the improvement.\n",
        "random_vector_for_generation = tf.random.normal(\n",
        "    shape=[num_examples_to_generate, latent_dim])\n",
        "model = CVAE(latent_dim)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "RmdVsmvhPxyy",
        "colab": {}
      },
      "source": [
        "def generate_and_save_images(model, epoch, test_input):\n",
        "  predictions = model.sample(test_input)\n",
        "  fig = plt.figure(figsize=(4,4))\n",
        "\n",
        "  for i in range(predictions.shape[0]):\n",
        "      plt.subplot(4, 4, i+1)\n",
        "      plt.imshow(predictions[i, :, :, 0], cmap='gray')\n",
        "      plt.axis('off')\n",
        "\n",
        "  # tight_layout minimizes the overlap between 2 sub-plots\n",
        "  plt.savefig('image_at_epoch_{:04d}.png'.format(epoch))\n",
        "  plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "2M7LmLtGEMQJ",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "outputId": "3087abbe-97e9-4672-d167-1f93d51460fb"
      },
      "source": [
        "generate_and_save_images(model, 0, random_vector_for_generation)\n",
        "\n",
        "for epoch in range(1, epochs + 1):\n",
        "  start_time = time.time()\n",
        "  for train_x in train_dataset:\n",
        "    compute_apply_gradients(model, train_x, optimizer)\n",
        "  end_time = time.time()\n",
        "\n",
        "  if epoch % 1 == 0:\n",
        "    loss = tf.keras.metrics.Mean()\n",
        "    for test_x in test_dataset:\n",
        "      loss(compute_loss(model, test_x))\n",
        "    elbo = -loss.result()\n",
        "    display.clear_output(wait=False)\n",
        "    print('Epoch: {}, Test set ELBO: {}, '\n",
        "          'time elapse for current epoch {}'.format(epoch,\n",
        "                                                    elbo,\n",
        "                                                    end_time - start_time))\n",
        "    generate_and_save_images(\n",
        "        model, epoch, random_vector_for_generation)"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch: 100, Test set ELBO: -78.06375885009766, time elapse for current epoch 3.855811834335327\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOwAAADnCAYAAAAdFLrXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhP8H4Gt+xz3at54AAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 288x288 with 16 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "P4M_vIbUi7c0"
      },
      "source": [
        "### Display an image using the epoch number"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "WfO5wCdclHGL",
        "colab": {}
      },
      "source": [
        "def display_image(epoch_no):\n",
        "  return PIL.Image.open('image_at_epoch_{:04d}.png'.format(epoch_no))"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "5x3q9_Oe5q0A",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 266
        },
        "outputId": "6eed7e83-0b2b-412a-fc02-ef1d451b0ea1"
      },
      "source": [
        "plt.imshow(display_image(epochs))\n",
        "plt.axis('off')# Display images"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(-0.5, 287.5, 287.5, -0.5)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAOcAAADnCAYAAADl9EEgAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "NywiH3nL8guF"
      },
      "source": [
        "### Generate a GIF of all the saved images."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "IGKQgENQ8lEI",
        "colab": {}
      },
      "source": [
        "anim_file = 'cvae.gif'\n",
        "\n",
        "with imageio.get_writer(anim_file, mode='I') as writer:\n",
        "  filenames = glob.glob('image*.png')\n",
        "  filenames = sorted(filenames)\n",
        "  last = -1\n",
        "  for i,filename in enumerate(filenames):\n",
        "    frame = 2*(i**0.5)\n",
        "    if round(frame) > round(last):\n",
        "      last = frame\n",
        "    else:\n",
        "      continue\n",
        "    image = imageio.imread(filename)\n",
        "    writer.append_data(image)\n",
        "  image = imageio.imread(filename)\n",
        "  writer.append_data(image)\n",
        "\n",
        "import IPython\n",
        "if IPython.version_info >= (6,2,0,''):\n",
        "  display.Image(filename=anim_file)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "colab_type": "text",
        "id": "yQXO_dlXkKsT"
      },
      "source": [
        "If you're working in Colab you can download the animation with the code below:"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab_type": "code",
        "id": "4fSJS3m5HLFM",
        "colab": {}
      },
      "source": [
        "try:\n",
        "  from google.colab import files\n",
        "except ImportError:\n",
        "   pass\n",
        "else:\n",
        "  files.download(anim_file)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RYweoy_oU-v_",
        "colab_type": "code",
        "outputId": "8d12da1c-82aa-490b-d845-232f2422ecad",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 133
        }
      },
      "source": [
        "YEA............."
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "error",
          "ename": "SyntaxError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;36m  File \u001b[0;32m\"<ipython-input-18-82f16f7080bd>\"\u001b[0;36m, line \u001b[0;32m1\u001b[0m\n\u001b[0;31m    YEA.............\u001b[0m\n\u001b[0m         ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hdSEpZGzVdDk",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}