deploying_lambda.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Introduction to the MNIST data\n",
    "\n",
    "For this classification project, we will be using the MNIST data set that contain 70,000 images of handwritten digits. In this data set, each row represents an image and each column a pixel from a 28 by 28 pixel image. The MNIST dataset is widely used to train classifiers and can be fetched using the helper function sklearn.datasets.fetch_openml.\n",
    "\n",
    "The code below will download the data, and sample 20,000 rows from the original dataset. I will be reducing the number of rows in the dataset to decrease model size and to reduce build time for this project. The code below will also plot the first image in the data set which we can see is the number eight."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-0.5, 27.5, 27.5, -0.5)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_openml\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "\n",
    "# Load data\n",
    "mnist = fetch_openml(\"mnist_784\", version=1)\n",
    "\n",
    "# Randomly sample 20000 rows from the original dataset\n",
    "mnist_data = (\n",
    "    mnist\n",
    "    .data\n",
    "    .sample(n=20000, random_state=42, axis=0, replace=False)\n",
    ")\n",
    "\n",
    "# Slice target by the same row sampling\n",
    "target = (\n",
    "    mnist\n",
    "    .target\n",
    "    .loc[mnist_data.index].astype('uint8')\n",
    ")\n",
    "\n",
    "# Reshape values to be 28x28\n",
    "some_digit_image = (\n",
    "    mnist_data\n",
    "    .iloc[0]\n",
    "    .values\n",
    "    .reshape(28,28)\n",
    "    .astype('float32')\n",
    ")\n",
    "plt.imshow(some_digit_image, cmap = \"binary\")\n",
    "plt.axis(\"off\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "# 2. Training a K-Nearest Neighbors Classifier\n",
    "\n",
    "First we will split the data into training and test set then train a K-nearest neighbour classifier using thescikit-learn library."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cross Validation Scores: [0.956875 0.96375  0.95625  0.953125 0.955    0.955    0.958125 0.95375\n",
      " 0.960625 0.958125]\n",
      "Average accuracy: 0.9570625000000001\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.neighbors import  KNeighborsClassifier\n",
    "from sklearn.model_selection import cross_val_score\n",
    "import numpy as np\n",
    "\n",
    "# Function to train KNN Classifier and show scores\n",
    "def train_knn_model(features:np.array, target:np.array):\n",
    "\n",
    "    # Train KNN Classifier\n",
    "    knnclf = KNeighborsClassifier(weights='distance', n_neighbors=4)\n",
    "    knnclf.fit(features, target)\n",
    "    scores = cross_val_score(\n",
    "        knnclf, features, target, scoring='accuracy', cv=10\n",
    "    )\n",
    "    print(f'Cross Validation Scores: {scores}')\n",
    "    print(f'Average accuracy: {np.mean(scores)}')\n",
    "    return knnclf, scores\n",
    "\n",
    "# Split data to training and test set\n",
    "train_features, test_features, train_target, test_target = train_test_split(\n",
    "        mnist_data, target, test_size = 0.2, random_state = 42\n",
    ")\n",
    "knnclf, scores = train_knn_model(train_features, train_target)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The model achieves a decent average accuracy of 96% from cross validation. Let's evaluate the model's performance on the test_features data set and plot a confusion matrix with the show_cm function as shown below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.95725\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
    "from sklearn.metrics import accuracy_score\n",
    "\n",
    "def show_cm(y_true, y_pred, labels):\n",
    "\n",
    "    # Display Confusion matrix and show accuracy scores\n",
    "    conf_mat = confusion_matrix(y_true, y_pred, labels=labels)\n",
    "    disp = ConfusionMatrixDisplay(confusion_matrix=conf_mat, display_labels=labels)\n",
    "    score = accuracy_score(y_true, y_pred)\n",
    "    print(f'Accuracy: {score}')\n",
    "    disp.plot();\n",
    "\n",
    "# Make predictions\n",
    "test_target_pred = knnclf.predict(test_features)\n",
    "# Show confusion matrix\n",
    "show_cm(test_target, test_target_pred, range(10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Base on the accuracy on the test data set, we can see that our model fits the data. We get very similar prediction accuracy when comparing accuracies between the training and testing set. \n",
    "\n",
    "Furthermore, a confusion matrix, like above, is very effective in helping visualise the gaps in the model's performance. It will help us understand the kind of errors that the classifier is making.\n",
    "\n",
    "The matrix indicates that there were 16 instances where the number 4 was misidentified for the number 9, and 12 instances where the number 8 was misidentified for the number 5. \n",
    "\n",
    "Looking at the images below, it is possible to see why some of these errors may occur as the number 4 and 9 do share some similar features. Likewise for the number 8 and 5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def show_digits(pixel_vals):\n",
    "\n",
    "    some_digit_image = (\n",
    "        pixel_vals\n",
    "        .values\n",
    "        .reshape(28,28)\n",
    "        .astype('float32')\n",
    "    )\n",
    "    plt.imshow(some_digit_image, cmap = \"binary\")\n",
    "    plt.axis(\"off\")\n",
    "\n",
    "fours = train_features[train_target == 4]\n",
    "nines = train_features[train_target == 9]\n",
    "eights = train_features[train_target == 8]\n",
    "fives = train_features[train_target == 5]\n",
    "\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.subplot(221); show_digits(fours.iloc[0])\n",
    "plt.subplot(222); show_digits(nines.iloc[0])\n",
    "plt.subplot(223); show_digits(eights.iloc[1])\n",
    "plt.subplot(224); show_digits(fives.iloc[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This insight is not going to affect model deployment on AWS but will help guide strategies to further improve the model. \n",
    "For now, we will save the model locally to be containerised as part of the lambda function using Docker."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['app/knnclf.joblib']"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import joblib\n",
    "\n",
    "joblib.dump(knnclf, 'app/knnclf.joblib')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "# 3. Initialising AWS S3 Bucket\n",
    "\n",
    "The image below illustrates the overall resource infrastructure that will need to be deployed to support our lambda function. There are three key resources requirements for our application:\n",
    "\n",
    "1. S3 Bucket to store data.\n",
    "2. API gateway to manage HTTP requests.\n",
    "3. Lambda function containing the predictive logic.\n",
    "\n",
    "![](overview.png)\n",
    "\n",
    "#### Serverless deployment of ML models\n",
    "\n",
    "1. Test data is uploaded to a S3 bucket.\n",
    "\n",
    "2. To initiate the lambda function, a POST HTTP request is sent through the Amazon API Gateway. \n",
    "\n",
    "3. Initialisation of the lambda function executes code that downloads the data from the S3 bucket and performs predictions. \n",
    "\n",
    "4. A HTTP response is returned to client with the predictions as a data payload.\n",
    "\n",
    "The Lambda function will contain Python code that performs a prediction based on the test_features dataset stored on a S3 bucket. \n",
    "\n",
    "Therefore, we will first need to initialise a S3 bucket where we can host our data.\n",
    "\n",
    "To do so, we will be interacting with AWS using the AWS Python SDK `boto3`. This package contains all the dependencies we require to integrate Python projects with AWS. \n",
    "\n",
    "Let's initialise a S3 bucket with the code below.\n",
    "\n",
    "***Note:*** *The `bucket_name` has to be unique therefore you will have to replace the bucket_name with a name that is not taken.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "ename": "BucketAlreadyOwnedByYou",
     "evalue": "An error occurred (BucketAlreadyOwnedByYou) when calling the CreateBucket operation: Your previous request to create the named bucket succeeded and you already own it.",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mBucketAlreadyOwnedByYou\u001b[0m                   Traceback (most recent call last)",
      "\u001b[0;32m/var/folders/kj/2sm6n_8n4qz6nc6n10vljkbr0000gn/T/ipykernel_27841/1589641224.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     14\u001b[0m \u001b[0mregion\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'eu-west-2'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     15\u001b[0m \u001b[0mbucket_name\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m'lh-lambda-buckets-2022'\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 16\u001b[0;31m \u001b[0mcreate_bucket\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mregion\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbucket_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/var/folders/kj/2sm6n_8n4qz6nc6n10vljkbr0000gn/T/ipykernel_27841/1589641224.py\u001b[0m in \u001b[0;36mcreate_bucket\u001b[0;34m(region, bucket_name)\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m     \u001b[0ms3\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mboto3\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclient\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m's3'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 6\u001b[0;31m     response = s3.create_bucket(\n\u001b[0m\u001b[1;32m      7\u001b[0m         \u001b[0mBucket\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mbucket_name\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      8\u001b[0m         CreateBucketConfiguration={\n",
      "\u001b[0;32m~/Library/Caches/pypoetry/virtualenvs/aws-lambdas-deploy-0XbIEOKZ-py3.9/lib/python3.9/site-packages/botocore/client.py\u001b[0m in \u001b[0;36m_api_call\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    389\u001b[0m                     \"%s() only accepts keyword arguments.\" % py_operation_name)\n\u001b[1;32m    390\u001b[0m             \u001b[0;31m# The \"self\" in this scope is referring to the BaseClient.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 391\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_api_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moperation_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    392\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    393\u001b[0m         \u001b[0m_api_call\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpy_operation_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Library/Caches/pypoetry/virtualenvs/aws-lambdas-deploy-0XbIEOKZ-py3.9/lib/python3.9/site-packages/botocore/client.py\u001b[0m in \u001b[0;36m_make_api_call\u001b[0;34m(self, operation_name, api_params)\u001b[0m\n\u001b[1;32m    717\u001b[0m             \u001b[0merror_code\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparsed_response\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Error\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m{\u001b[0m\u001b[0;34m}\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mget\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Code\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    718\u001b[0m             \u001b[0merror_class\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexceptions\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_code\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0merror_code\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 719\u001b[0;31m             \u001b[0;32mraise\u001b[0m \u001b[0merror_class\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mparsed_response\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moperation_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    720\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    721\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mparsed_response\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mBucketAlreadyOwnedByYou\u001b[0m: An error occurred (BucketAlreadyOwnedByYou) when calling the CreateBucket operation: Your previous request to create the named bucket succeeded and you already own it."
     ]
    }
   ],
   "source": [
    "import boto3\n",
    "\n",
    "def create_bucket(region:str, bucket_name:str) -> dict:\n",
    "\n",
    "    s3 = boto3.client('s3')\n",
    "    response = s3.create_bucket(\n",
    "        Bucket=bucket_name,\n",
    "        CreateBucketConfiguration={\n",
    "            'LocationConstraint':region\n",
    "        }\n",
    "    )\n",
    "    return response\n",
    "\n",
    "region = 'eu-west-2'\n",
    "bucket_name = 'lh-lambda-buckets-2022'\n",
    "create_bucket(region, bucket_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Upload Test data to S3\n",
    "\n",
    "The S3 bucket will host our test_features data set which we can call in our lambda function to perform a prediction.\n",
    "\n",
    "To save an object currently in our workspace, we will be making use of BytesIO function from the io library. This will enable us to temporary store the test_features data set in a file object. This file object can be uploaded onto a S3 bucket by calling the .upload_fileobj function.\n",
    "\n",
    "The bucket variable defines the destination S3 bucket and the key variable will define the file path in the bucket. The `bucket` and `key` variables will form part of the data payload in the POST HTTP request to our lambda function.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [],
   "source": [
    "from io import BytesIO\n",
    "import joblib\n",
    "import boto3\n",
    "\n",
    "def UploadToS3(data, bucket:str, key:str):\n",
    "\n",
    "    with BytesIO() as f:\n",
    "        joblib.dump(data, f)\n",
    "        f.seek(0)\n",
    "        (\n",
    "            boto3\n",
    "            .client('s3')\n",
    "            .upload_fileobj(Bucket=bucket, Key=key, Fileobj=f)\n",
    "        )\n",
    "\n",
    "bucket_name = 'lh-lambda-buckets-202222'\n",
    "key =  'validation/test_features.joblib'\n",
    "UploadToS3(test_features, bucket_name, key)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### List all objects in S3 bucket\n",
    "\n",
    "We can check if the objects have been uploaded with the helper function below. list_s3_objects will list all objects in the defined bucket."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['validation/test_features.joblib']"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import boto3\n",
    "\n",
    "def listS3Objects(bucket:str) -> list:\n",
    "\n",
    "     # Connect to s3 resource\n",
    "    s3 = boto3.resource('s3')\n",
    "    my_bucket = s3.Bucket(bucket)\n",
    "\n",
    "    # List all object keys in s3 bucket\n",
    "    obj_list = [object_summary.key for object_summary in my_bucket.objects.all()]\n",
    "    return obj_list\n",
    "\n",
    "listS3Objects('lh-lambda-buckets-2022')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have now successfully initialised a S3 bucket to store the test_feature data. The next two key resources, API Gateway and lambda function, will be deployed using AWS Serverless Application Model (SAM). Refer to main blog for details."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<hr>\n",
    "\n",
    "# 4. Deploying and Testing AWS Lambda Functions with SAM"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 4.4. Building and testing the application locally.\n",
    "\n",
    "AWS SAM provide functionality to build and locally test applications before deployment. \n",
    "Ensure docker is running. \n",
    "\n",
    "In a terminal window, navigate to the project directory and build the application in SAM.\n",
    "\n",
    "a) Ensure docker is running. In a terminal window, navigate to the project directory and build the application in SAM.\n",
    "\n",
    "`sam build -t template_no_auth.yaml`\n",
    "\n",
    "b) Locally deploy the dockerised lambda function.\n",
    "\n",
    "`sam local start-api`\n",
    "\n",
    "c) Locally invoke the function at http://127.0.0.1:3000/predict. Your URL may differ.\n",
    "\n",
    "***Note***: *The `bucket` and `key` variable which references the test_feature data set on S3 will need to be passed as part of the data payload in the POST HTTP request.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.95725\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import requests\n",
    "import json\n",
    "import numpy as np\n",
    "\n",
    "bucket_name = 'lh-lambda-buckets-202222'\n",
    "key =  'validation/test_features.joblib'\n",
    "\n",
    "data = {\n",
    "    'bucket':bucket_name,\n",
    "    'key':key,\n",
    "}\n",
    "\n",
    "headers = {\n",
    "    'Content-type': \"application/json\"\n",
    "}\n",
    "\n",
    "# Main code for post HTTP request\n",
    "url = \"http://127.0.0.1:3000/predict\"\n",
    "response = requests.request(\"POST\", url, headers=headers, data=json.dumps(data))\n",
    "lambda_predictions = np.array(response.json())\n",
    "show_cm(test_target, lambda_predictions, range(10))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 4.5 Deploying on AWS Lambda\n",
    "\n",
    "As easy as it was to deploy locally, SAM will also handle all the heavy lifting to deploy on AWS Lambda.\n",
    "\n",
    "a) Build the application in SAM.\n",
    "\n",
    "`sam build -t template_no_auth.yaml`\n",
    "\n",
    "b) Deploy the application.\n",
    "\n",
    "`sam deploy --guided`\n",
    "\n",
    "Follow the prompts that guides you through the deployment configurations. Most of the settings I used were the default value with a few exceptions. See main blog for details.\n",
    "\n",
    "c) Invoke your function by replacing the URL in the code below with the URL from the output out of a successfully deployed stack."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<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>pixel1</th>\n",
       "      <th>pixel2</th>\n",
       "      <th>pixel3</th>\n",
       "      <th>pixel4</th>\n",
       "      <th>pixel5</th>\n",
       "      <th>pixel6</th>\n",
       "      <th>pixel7</th>\n",
       "      <th>pixel8</th>\n",
       "      <th>pixel9</th>\n",
       "      <th>pixel10</th>\n",
       "      <th>...</th>\n",
       "      <th>pixel775</th>\n",
       "      <th>pixel776</th>\n",
       "      <th>pixel777</th>\n",
       "      <th>pixel778</th>\n",
       "      <th>pixel779</th>\n",
       "      <th>pixel780</th>\n",
       "      <th>pixel781</th>\n",
       "      <th>pixel782</th>\n",
       "      <th>pixel783</th>\n",
       "      <th>pixel784</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24301</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52585</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13415</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28171</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>62085</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35438</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3431</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39641</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29551</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68803</th>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>4000 rows × 784 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       pixel1  pixel2  pixel3  pixel4  pixel5  pixel6  pixel7  pixel8  pixel9  \\\n",
       "24301     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "52585     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "13415     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "28171     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "62085     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "...       ...     ...     ...     ...     ...     ...     ...     ...     ...   \n",
       "35438     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "3431      0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "39641     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "29551     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "68803     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0     0.0   \n",
       "\n",
       "       pixel10  ...  pixel775  pixel776  pixel777  pixel778  pixel779  \\\n",
       "24301      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "52585      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "13415      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "28171      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "62085      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "...        ...  ...       ...       ...       ...       ...       ...   \n",
       "35438      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "3431       0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "39641      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "29551      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "68803      0.0  ...       0.0       0.0       0.0       0.0       0.0   \n",
       "\n",
       "       pixel780  pixel781  pixel782  pixel783  pixel784  \n",
       "24301       0.0       0.0       0.0       0.0       0.0  \n",
       "52585       0.0       0.0       0.0       0.0       0.0  \n",
       "13415       0.0       0.0       0.0       0.0       0.0  \n",
       "28171       0.0       0.0       0.0       0.0       0.0  \n",
       "62085       0.0       0.0       0.0       0.0       0.0  \n",
       "...         ...       ...       ...       ...       ...  \n",
       "35438       0.0       0.0       0.0       0.0       0.0  \n",
       "3431        0.0       0.0       0.0       0.0       0.0  \n",
       "39641       0.0       0.0       0.0       0.0       0.0  \n",
       "29551       0.0       0.0       0.0       0.0       0.0  \n",
       "68803       0.0       0.0       0.0       0.0       0.0  \n",
       "\n",
       "[4000 rows x 784 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 576x576 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def show_digits(pixel_vals):\n",
    "\n",
    "    some_digit_image = (\n",
    "        pixel_vals\n",
    "        .values\n",
    "        .reshape(28,28)\n",
    "        .astype('float32')\n",
    "    )\n",
    "    plt.imshow(some_digit_image, cmap = \"binary\")\n",
    "    plt.axis(\"off\")\n",
    "\n",
    "fours = train_features[train_target == 4]\n",
    "nines = train_features[train_target == 9]\n",
    "eights = train_features[train_target == 8]\n",
    "fives = train_features[train_target == 5]\n",
    "\n",
    "plt.figure(figsize=(8,8))\n",
    "plt.subplot(221); show_digits(fours.iloc[0])\n",
    "plt.subplot(222); show_digits(nines.iloc[0])\n",
    "plt.subplot(223); show_digits(eights.iloc[1])\n",
    "plt.subplot(224); show_digits(fives.iloc[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.95725\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import requests\n",
    "import json\n",
    "import numpy as np\n",
    "\n",
    "bucket_name = 'lh-lambda-buckets-202222'\n",
    "key =  'validation/test_features.joblib'\n",
    "\n",
    "data = {\n",
    "    'bucket':bucket_name,\n",
    "    'key':key,\n",
    "}\n",
    "\n",
    "headers = {\n",
    "    'Content-type': \"application/json\"\n",
    "}\n",
    "\n",
    "# Main code for post HTTP request\n",
    "url = \"https://qjiim4z1db.execute-api.eu-west-2.amazonaws.com/dev/predict\"\n",
    "response = requests.request(\"POST\", url, headers=headers, json=data)\n",
    "lambda_predictions = np.array(response.json())\n",
    "show_cm(test_target, lambda_predictions, range(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[NbConvertApp] Converting notebook deploying_lambda.ipynb to html\n",
      "[NbConvertApp] Writing 412187 bytes to deploying_lambda.html\n"
     ]
    }
   ],
   "source": [
    "!jupyter nbconvert deploying_lambda.ipynb --to html --template classic"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "919d5182d6e251f85cfea4f0d38d6cd16b39376b792c9ccd6a31cc8ff7d08ddb"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}