From fc759eb7696ed9344403b3e0b0901ca11f77be4c Mon Sep 17 00:00:00 2001
From: EC2 Default User <ec2-user@ip-10-0-0-171.ec2.internal>
Date: Thu, 16 Mar 2023 00:51:43 +0000
Subject: [PATCH] updating ml evaluation

---
 ml_training/Evaluate_Results.ipynb | 375 +++++++++------------------
 ml_training/evaluate.py            | 402 +++++++++++++++++------------
 ml_training/test_e2e.py            |   9 +-
 3 files changed, 365 insertions(+), 421 deletions(-)

diff --git a/ml_training/Evaluate_Results.ipynb b/ml_training/Evaluate_Results.ipynb
index cfc022d..ae46652 100644
--- a/ml_training/Evaluate_Results.ipynb
+++ b/ml_training/Evaluate_Results.ipynb
@@ -12,7 +12,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 1,
    "id": "b17fcf27",
    "metadata": {},
    "outputs": [],
@@ -21,8 +21,11 @@
     "import json\n",
     "import boto3\n",
     "import numpy as np\n",
+    "import seaborn as sns\n",
     "import matplotlib.pyplot as plt\n",
     "\n",
+    "from sklearn.metrics import confusion_matrix\n",
+    "\n",
     "def process_results(bucket_name, prefix, severe_threshold=0.45):\n",
     "    s3_client = boto3.client('s3')\n",
     "    s3_resource = boto3.resource('s3')\n",
@@ -54,14 +57,14 @@
     "\n",
     "            os.remove(fname)\n",
     "        \n",
-    "        severe_chips_results.append(np.array([severe_count, total_chips, fname]))\n",
+    "        severe_chips_results.append(np.array([severe_count, total_chips])) #, fname]))\n",
     "\n",
     "    return np.array(severe_chips_results)"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 2,
    "id": "80dec071",
    "metadata": {},
    "outputs": [
@@ -70,14 +73,14 @@
      "output_type": "stream",
      "text": [
       "Mild\n",
-      "{'84204T_001', '84485T_002', '92316', '92310 2', '84449T_001', '84483T_004', '84439T_002', '92308 2', '92308', '92311', '84449T_003', '92314', '84244T_004', '92309', '92313', '84250T_004', '92310', '84464T_001', '92312', '92341', '84247T_002', '84487T_003', '84245T_003', '84463T_004', '92315', '84251T_002'}\n",
+      "{'84439T_002', '84487T_003', '84483T_004', '92315', '92313', '84245T_003', '84449T_001', '92311', '84464T_001', '92314', '84478T_001', '84463T_001', '84204T_001', '84444T_002', '92308 2', '84250T_004', '92310', '92310 2', '84248T_004', '84247T_002', '92309', '84251T_002', '84438T_003', '84244T_004', '84485T_002', '84463T_004', '92312', '84449T_003', '92341', '92308', '84285T_001', '84250T_001', '92316'}\n",
       "\n",
       "Moderate\n",
-      "{'92324', '92322', '92327', '92323', '92317', '92319'}\n",
+      "{'84459T_004', '84460T_003', '84437T_003', '84047T_003', '84421T_003', '92317', '92324', '92327', '92323', '84422T_002', '84475T_002', '92319', '84422T_004', '92322'}\n",
       "\n",
       "too many indices for array: array is 1-dimensional, but 2 were indexed\n",
       "Severe\n",
-      "{'92330 2', '84278T_004', '84468T_002', '84456T_003', '92334', '84443T_001', '92337', '84461T_001', '84436T_003', '84282T_003', '92329 2', '92332 2', '84436T_001', '92338', '84452T_002', '84442T_004', '84280T_003', '84479T_003', '84461T_004', '92339', '92331', '92330', '92332', '92335', '84443T_003', '84479T_004', '92331 2', '92329', '92333'}\n",
+      "{'84461T_001', '84442T_004', '92330 2', '92332 2', '92332', '84477T_001', '92329', '92337', '92339', '84282T_003', '92329 2', '84284T_001', '92330', '92334', '84436T_003', '84436T_001', '92331', '92333', '92338', '84443T_001', '84282T_004', '84479T_003', '84461T_004', '92331 2', '84456T_003', '84468T_002', '84443T_003', '84280T_003', '92335', '84452T_002', '84479T_004', '84278T_004'}\n",
       "\n"
      ]
     }
@@ -95,211 +98,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
-   "id": "1deaf18c",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "['825.0' '4997' './84204T_001.tif_29999.json']\n",
-      "['749.0' '4266' './84485T_002.tif_19999.json']\n",
-      "['806.0' '4101' './92316.tif_24999.json']\n",
-      "['978.0' '5649' './92310 2.tif_24999.json']\n",
-      "['827.0' '3184' './84449T_001.tif_19999.json']\n",
-      "['1585.0' '11216' './84483T_004.tif_19999.json']\n",
-      "['1199.0' '7856' './84439T_002.tif_24999.json']\n",
-      "['1652.0' '8210' './92308 2.tif_29999.json']\n",
-      "['2997.0' '15373' './92308.tif_29999.json']\n",
-      "['3165.0' '13066' './92311.tif_29999.json']\n",
-      "['652.0' '5939' './84449T_003.tif_24999.json']\n",
-      "['735.0' '4069' './92314.tif_29999.json']\n",
-      "['127.0' '723' './84244T_004.tif_14999.json']\n",
-      "['132.0' '995' './92309.tif_09999.json']\n",
-      "['1120.0' '5781' './92313.tif_19999.json']\n",
-      "['344.0' '1802' './84250T_004.tif_24999.json']\n",
-      "['1936.0' '9893' './92310.tif_09999.json']\n",
-      "['642.0' '3670' './84464T_001.tif_24999.json']\n",
-      "['1236.0' '6286' './92312.tif_19999.json']\n",
-      "['1321.0' '7558' './92341.tif_29999.json']\n",
-      "['84.0' '518' './84247T_002.tif_09999.json']\n",
-      "['1556.0' '8896' './84487T_003.tif_29999.json']\n",
-      "['1131.0' '4223' './84245T_003.tif_19999.json']\n",
-      "['91.0' '216' './84463T_004.tif_09999.json']\n",
-      "['376.0' '2363' './92315.tif_14999.json']\n",
-      "['685.0' '2223' './84251T_002.tif_09999.json']\n",
-      "./84449T_003.tif_24999.json 0.10978279171577707\n"
-     ]
-    }
-   ],
-   "source": [
-    "#Lowest for preloading - ['132.0' '995' './92309.tif_09999.json']\n",
-    "#Demoing - ./84449T_003.tif_24999.json 0.10978279171577707\n",
-    "#1 more solid MILD\n",
-    "#['376.0' '2363' './92315.tif_14999.json']\n",
-    "\n",
-    "#1 correct MODERATE\n",
-    "#['1305.0' '5307' './92317.tif_14999.json'] 0.2459016393442623\n",
-    "#1 MODERATE classified as SEVERE\n",
-    "#['3767.0' '11972' './92319.tif_24999.json'] 0.31465085198797194\n",
-    "\n",
-    "#1 solid SEVERE\n",
-    "#['5018.0' '8811' './92338.tif_29999.json']\n",
-    "#Highest SEVERE\n",
-    "#./92332.tif_09999.json 9061.0\n",
-    "\n",
-    "lowest_count = np.inf\n",
-    "lowest_fname = ''\n",
-    "\n",
-    "for stats in mild_image_severe_predictions:\n",
-    "    print(stats)\n",
-    "    severe_count = float(stats[0])\n",
-    "    total = int(stats[1])\n",
-    "    fname = stats[2]\n",
-    "\n",
-    "    severe_percentage = severe_count/total\n",
-    "\n",
-    "    if severe_percentage < lowest_count:\n",
-    "        lowest_count = severe_percentage\n",
-    "        lowest_fname = fname\n",
-    "\n",
-    "print(lowest_fname, lowest_count)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "id": "07bdd8e3",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "['2402.0' '8314' './92330 2.tif_24999.json']\n",
-      "['358.0' '1175' './84278T_004.tif_04999.json']\n",
-      "['3931.0' '9001' './84468T_002.tif_24999.json']\n",
-      "['2394.0' '6302' './84456T_003.tif_19999.json']\n",
-      "['3004.0' '9726' './92334.tif_24999.json']\n",
-      "['6734.0' '12945' './84443T_001.tif_29999.json']\n",
-      "['1981.0' '7454' './92337.tif_24999.json']\n",
-      "['4858.0' '9118' './84461T_001.tif_24999.json']\n",
-      "['3701.0' '6046' './84436T_003.tif_14999.json']\n",
-      "['2954.0' '5078' './84282T_003.tif_29999.json']\n",
-      "['1663.0' '10701' './92329 2.tif_19999.json']\n",
-      "['6988.0' '8105' './92332 2.tif_24999.json']\n",
-      "['3917.0' '6104' './84436T_001.tif_24999.json']\n",
-      "['5018.0' '8811' './92338.tif_29999.json']\n",
-      "['7082.0' '8501' './84452T_002.tif_19999.json']\n",
-      "['7081.0' '17092' './84442T_004.tif_29999.json']\n",
-      "['2419.0' '5465' './84280T_003.tif_14999.json']\n",
-      "['3630.0' '7764' './84479T_003.tif_24999.json']\n",
-      "['6696.0' '11449' './84461T_004.tif_29999.json']\n",
-      "['1350.0' '3824' './92339.tif_24999.json']\n",
-      "['8494.0' '24521' './92331.tif_29999.json']\n",
-      "['5940.0' '20232' './92330.tif_29999.json']\n",
-      "['9061.0' '10733' './92332.tif_09999.json']\n",
-      "['2667.0' '5518' './92335.tif_19999.json']\n",
-      "['6420.0' '12604' './84443T_003.tif_29999.json']\n",
-      "['2950.0' '6925' './84479T_004.tif_24999.json']\n",
-      "['4649.0' '16911' './92331 2.tif_29999.json']\n",
-      "['4231.0' '18403' './92329.tif_19999.json']\n",
-      "['2902.0' '7403' './92333.tif_24999.json']\n",
-      "./92332.tif_09999.json 9061.0\n"
-     ]
-    }
-   ],
-   "source": [
-    "highest_count = 0\n",
-    "highest_fname = ''\n",
-    "\n",
-    "for stats in severe_image_severe_predictions:\n",
-    "    print(stats)\n",
-    "    severe_count = float(stats[0])\n",
-    "    total = int(stats[1])\n",
-    "    fname = stats[2]\n",
-    "\n",
-    "    severe_percentage = severe_count/total\n",
-    "\n",
-    "    if severe_count > highest_count:\n",
-    "        highest_count = severe_count\n",
-    "        highest_fname = fname\n",
-    "\n",
-    "print(highest_fname, highest_count)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "id": "c5850b1c",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "['2143.0' '7779' './92324.tif_29999.json'] 0.27548528088443247\n",
-      "['1481.0' '8586' './92322.tif_24999.json'] 0.17249010016305613\n",
-      "['2382.0' '8357' './92327.tif_29999.json'] 0.28503051334210844\n",
-      "['30.0' '54' './92323.tif_24999.json'] 0.5555555555555556\n",
-      "['1305.0' '5307' './92317.tif_14999.json'] 0.2459016393442623\n",
-      "['3767.0' '11972' './92319.tif_24999.json'] 0.31465085198797194\n"
-     ]
-    }
-   ],
-   "source": [
-    "for stats in moderate_image_severe_predictions:\n",
-    "    severe_count = float(stats[0])\n",
-    "    total = int(stats[1])\n",
-    "    fname = stats[2]\n",
-    "\n",
-    "    severe_percentage = severe_count/total\n",
-    "    \n",
-    "    print(stats, severe_percentage)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "0153733e",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "7c0a52ae",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "91a80dfa",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "1dd08979",
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 69,
+   "execution_count": 21,
    "id": "2fa887ea",
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 1000x500 with 1 Axes>"
       ]
@@ -337,8 +142,8 @@
     "    )\n",
     "\n",
     "#Plot thresholds\n",
-    "mild_threshold = 1050\n",
-    "severe_theshold = 1600\n",
+    "mild_threshold = 1000\n",
+    "severe_theshold = 1850\n",
     "\n",
     "y = np.arange(0, 4)\n",
     "threshold_line = np.ones_like(y)\n",
@@ -354,13 +159,13 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 72,
+   "execution_count": 30,
    "id": "ef4f7de1",
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 1000x500 with 1 Axes>"
       ]
@@ -399,8 +204,8 @@
     "    )\n",
     "\n",
     "#Plot thresholds\n",
-    "mild_threshold = 0.16\n",
-    "severe_theshold = 0.215\n",
+    "mild_threshold = 0.25\n",
+    "severe_theshold = 0.285\n",
     "\n",
     "y = np.arange(0, 9)\n",
     "threshold_line = np.ones_like(y)\n",
@@ -416,7 +221,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 73,
+   "execution_count": 49,
    "id": "158f82e8",
    "metadata": {},
    "outputs": [
@@ -424,17 +229,41 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Chip Counts\n",
-      "Accuracy: 69.23076923076923%\n",
-      "Mild Accuracy: 70.0%\n",
-      "Moderate Accuracy: 16.666666666666664%\n",
-      "Severe Accuracy: 100.0%\n",
+      "Chip Counts\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
       "\n",
-      "Chip Percents\n",
-      "Accuracy: 80.76923076923077%\n",
-      "Mild Accuracy: 100.0%\n",
-      "Moderate Accuracy: 16.666666666666664%\n",
-      "Severe Accuracy: 100.0%\n",
+      "Chip Percents\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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\n",
+      "text/plain": [
+       "<Figure size 600x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
       "\n"
      ]
     }
@@ -442,62 +271,88 @@
    "source": [
     "for percent_threshold in [False, True]:\n",
     "    if percent_threshold:\n",
-    "        mild_threshold = 0.16\n",
-    "        severe_theshold = 0.215\n",
+    "        mild_threshold = 0.25\n",
+    "        severe_theshold = 0.285\n",
     "        print(\"Chip Percents\")\n",
     "    else:\n",
-    "        mild_threshold = 1050\n",
-    "        severe_theshold = 1600\n",
+    "        mild_threshold = 1000\n",
+    "        severe_theshold = 1850\n",
     "        print(\"Chip Counts\")\n",
     "\n",
     "    #Count Mild\n",
     "    correct_mild = 0\n",
-    "    if True:\n",
-    "        for chip_count in mild_image_severe_predictions:\n",
+    "    correct_moderate = 0\n",
+    "    correct_severe = 0\n",
+    "    y_pred = []\n",
+    "    y = []\n",
+    "    classes = ('Mild', 'Moderate', 'Severe')\n",
+    "    for img_label_batch, label in [\n",
+    "        (mild_image_severe_predictions, 'Mild'),\n",
+    "        (moderate_image_severe_predictions, 'Moderate'),\n",
+    "        (severe_image_severe_predictions, 'Severe')\n",
+    "    ]:\n",
+    "        for chip_count in img_label_batch:\n",
+    "            y.append(classes.index(label))\n",
+    "\n",
     "            count = chip_count[0]\n",
-    "            total = chip_count[1]\n",
-    "            percentage = count / total\n",
+    "            percentage = count / chip_count[1]\n",
+    "\n",
+    "            #Check Mild Prediction\n",
     "            if (\n",
     "                (percent_threshold and percentage < mild_threshold) or\n",
     "                (not percent_threshold and count < mild_threshold)\n",
-    "           ):\n",
-    "                correct_mild += 1\n",
-    "    else:\n",
-    "        mild_image_severe_predictions = []\n",
+    "            ):\n",
+    "                y_pred.append(classes.index('Mild'))\n",
+    "                if label == 'Mild':\n",
+    "                    correct_mild += 1\n",
     "\n",
-    "    #Count Moderate\n",
-    "    correct_moderate = 0\n",
-    "    if True:\n",
-    "        for chip_count in moderate_image_severe_predictions:\n",
-    "            count = chip_count[0]\n",
-    "            total = chip_count[1]\n",
-    "            percentage = count / total\n",
-    "            if (\n",
+    "            #Check Moderate Prediction\n",
+    "            elif (\n",
     "                (percent_threshold and percentage >= mild_threshold and percentage < severe_theshold) or\n",
     "                (not percent_threshold and count >= mild_threshold and count < severe_theshold)\n",
     "            ):\n",
-    "                correct_moderate += 1\n",
-    "    else:\n",
-    "        moderate_image_severe_predictions = []\n",
+    "                y_pred.append(classes.index('Moderate'))\n",
+    "                if label == 'Moderate':\n",
+    "                    correct_moderate += 1\n",
     "\n",
-    "    correct_severe = 0\n",
-    "    if True:\n",
-    "        for chip_count in severe_image_severe_predictions:\n",
-    "            count = chip_count[0]\n",
-    "            total = chip_count[1]\n",
-    "            percentage = count / total\n",
-    "            if (\n",
+    "            #Check Severe Prediction\n",
+    "            elif (\n",
     "                (percent_threshold and percentage >= severe_theshold) or\n",
     "                (not percent_threshold and count >= severe_theshold)\n",
     "            ):\n",
-    "                correct_severe += 1\n",
-    "    else:\n",
-    "        severe_image_severe_predictions = []\n",
+    "                y_pred.append(classes.index('Severe'))\n",
+    "                if label == 'Severe':\n",
+    "                    correct_severe += 1\n",
+    "\n",
+    "            else:\n",
+    "                raise Exception('Bad prediction')\n",
+    "\n",
+    "    #Calculate Accuracy\n",
+    "    accuracy = (\n",
+    "        (correct_mild + correct_moderate + correct_severe) / \n",
+    "        (len(mild_image_severe_predictions) + len(moderate_image_severe_predictions) + len(severe_image_severe_predictions))\n",
+    "    )\n",
     "\n",
-    "    print(f'Accuracy: {100 * ((correct_mild + correct_moderate + correct_severe) / (len(mild_image_severe_predictions) + len(moderate_image_severe_predictions) + len(severe_image_severe_predictions)))}%')\n",
-    "    print(f'Mild Accuracy: {100 * ((correct_mild) / (len(mild_image_severe_predictions)))}%')\n",
-    "    print(f'Moderate Accuracy: {100 * ((correct_moderate) / (len(moderate_image_severe_predictions)))}%')\n",
-    "    print(f'Severe Accuracy: {100 * ((correct_severe) / (len(severe_image_severe_predictions)))}%')\n",
+    "    #Plot confusion matrix\n",
+    "    conf_matrix = confusion_matrix(y, y_pred, normalize='true')\n",
+    "\n",
+    "    plt.figure(figsize=(6,4), dpi=100)\n",
+    "    sns.set(font_scale = 1.1)\n",
+    "    ax = sns.heatmap(conf_matrix, annot=True, fmt='.2%', cmap=\"Blues\")\n",
+    "    ax.set_title(f\"Eosinophil Diagnosis (Accuracy: {accuracy:.2%})\", fontsize=14, pad=20)\n",
+    "    ax.set_xlabel(\"Predicted Diagnosis\", fontsize=14, labelpad=20)\n",
+    "    ax.set_ylabel(\"Actual Diagnosis\", fontsize=14, labelpad=20)\n",
+    "    ax.xaxis.set_ticklabels(['Mild', 'Moderate', 'Severe'])\n",
+    "    ax.yaxis.set_ticklabels(['Mild', 'Moderate', 'Severe'])\n",
+    "\n",
+    "    plt.show()\n",
+    "    #plt.savefig('/home/ec2-user/confusion-matrix.jpeg')\n",
+    "    plt.close()\n",
+    "\n",
+    "    #print(f'Accuracy: {100 * ((correct_mild + correct_moderate + correct_severe) / (len(mild_image_severe_predictions) + len(moderate_image_severe_predictions) + len(severe_image_severe_predictions)))}%')\n",
+    "    #print(f'Mild Accuracy: {100 * ((correct_mild) / (len(mild_image_severe_predictions)))}%')\n",
+    "    #print(f'Moderate Accuracy: {100 * ((correct_moderate) / (len(moderate_image_severe_predictions)))}%')\n",
+    "    #print(f'Severe Accuracy: {100 * ((correct_severe) / (len(severe_image_severe_predictions)))}%')\n",
     "    print()"
    ]
   },
diff --git a/ml_training/evaluate.py b/ml_training/evaluate.py
index fa377ef..f62ed97 100644
--- a/ml_training/evaluate.py
+++ b/ml_training/evaluate.py
@@ -2,8 +2,11 @@ import os
 import json
 import boto3
 import numpy as np
+import seaborn as sns
 import matplotlib.pyplot as plt
 
+from sklearn.metrics import confusion_matrix
+
 def process_results(bucket_name, prefix, severe_threshold=0.5):
     s3_client = boto3.client('s3')
     s3_resource = boto3.resource('s3')
@@ -34,166 +37,249 @@ def process_results(bucket_name, prefix, severe_threshold=0.5):
                 print(e)
 
             os.remove(fname)
-        
-        severe_chips_results.append(np.array([severe_count, total_chips]))
+
+        if total_chips > 700:
+            severe_chips_results.append(np.array([severe_count, total_chips]))
 
     return np.array(severe_chips_results)
 
-#for severe_prob in [0.3, 0.35, 0.4, 0.45, 0.5, 0.55, 0.6, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9]:
-severe_prob = 0.45
-
-#Process Mild Results
-mild_image_severe_predictions = process_results('digpath-results-2', 'Mild', severe_threshold=severe_prob)
-#Process Severe Results
-moderate_image_severe_predictions = process_results('digpath-results-2', 'Moderate', severe_threshold=severe_prob)
-#Process Severe Results
-severe_image_severe_predictions = process_results('digpath-results-2', 'Severe', severe_threshold=severe_prob)
-
-#Plot Chip Counts
-plt.figure(figsize=(10, 5))
-plt.title('Chips with Predicted Eosinophil in Whole Slide Image')
-if len(mild_image_severe_predictions) > 0 and True:
-    plt.hist(
-        mild_image_severe_predictions[:,0],
-        bins=20,
-        color='black',
-        alpha=0.5,
-        label=f'Mild ({len(mild_image_severe_predictions)})',
-    )
-if len(moderate_image_severe_predictions) > 0 and True:
-    plt.hist(
-        moderate_image_severe_predictions[:,0],
-        bins=20,
-        color='tan',
-        alpha=0.5,
-        label=f'Moderate ({len(moderate_image_severe_predictions)})',
-    )
-if len(severe_image_severe_predictions) > 0 and True:
-    plt.hist(
-        severe_image_severe_predictions[:,0],
-        bins=35,
-        color='darkred',
-        alpha=0.5,
-        label=f'Severe ({len(severe_image_severe_predictions)})',
-    )
-
-#Plot thresholds
-mild_threshold = 1600
-severe_theshold = 1600
-
-y = np.arange(0, 4)
-threshold_line = np.ones_like(y)
-plt.plot(mild_threshold * threshold_line, y, 'k--')
-plt.plot(severe_theshold * threshold_line, y, 'k--')
-
-plt.xlabel('Eosinophil Chip Predictions')
-plt.ylabel('Count')
-plt.legend()
-plt.savefig(f'/home/ec2-user/chip_count_evaluation_{100*severe_prob}.jpeg')
-plt.close()
-
-#Plot Chip Percentage
-plt.figure(figsize=(10, 5))
-plt.xlim(-0.5, 100)
-plt.title('Percentage of Chips Predicted as Eosinophil in Whole Slide Image')
-if len(mild_image_severe_predictions) > 0 and True:
-    plt.hist(
-        100. * mild_image_severe_predictions[:,0] / mild_image_severe_predictions[:,1],
-        bins=8,
-        color='black',
-        alpha=0.5,
-        label=f'Mild ({len(mild_image_severe_predictions)})',
-    )
-if len(moderate_image_severe_predictions) > 0 and True:
-    plt.hist(
-        100. * moderate_image_severe_predictions[:,0] / moderate_image_severe_predictions[:,1],
-        bins=10,
-        color='tan',
-        alpha=0.5,
-        label=f'Moderate ({len(moderate_image_severe_predictions)})',
-    )
-if len(severe_image_severe_predictions) > 0 and True:
-    plt.hist(
-        100. * severe_image_severe_predictions[:,0] / severe_image_severe_predictions[:,1],
-        bins=20,
-        color='darkred',
-        alpha=0.5,
-        label=f'Severe ({len(severe_image_severe_predictions)})',
-    )
-
-#Plot thresholds
-mild_threshold = 0.205
-severe_theshold = 0.25
-
-y = np.arange(0, 9)
-threshold_line = np.ones_like(y)
-plt.plot(100 * mild_threshold * threshold_line, y, 'k--')
-plt.plot(100 * severe_theshold * threshold_line, y, 'k--')
-
-plt.xlabel('Percentage of Eosinophil Chip Predictions')
-plt.ylabel('Count')
-plt.legend()
-plt.savefig(f'/home/ec2-user/chip_percentage_evaluation_{100*severe_prob}.jpeg')
-plt.close()
-
-
-#Print Accuracies
-for percent_threshold in [False, True]:
-    if percent_threshold:
-        mild_threshold = 0.205
-        severe_theshold = 0.25
-        print("Chip Percents")
-    else:
-        mild_threshold = 1600
-        severe_theshold = 1600
-        print("Chip Counts")
-
-    #Count Mild
-    correct_mild = 0
-    if True:
-        for chip_count in mild_image_severe_predictions:
-            count = chip_count[0]
-            total = chip_count[1]
-            percentage = count / total
-            if (
-                (percent_threshold and percentage < mild_threshold) or
-                (not percent_threshold and count < mild_threshold)
-        ):
-                correct_mild += 1
-    else:
-        mild_image_severe_predictions = []
-
-    #Count Moderate
-    correct_moderate = 0
-    if True:
-        for chip_count in moderate_image_severe_predictions:
-            count = chip_count[0]
-            total = chip_count[1]
-            percentage = count / total
-            if (
-                (percent_threshold and percentage >= mild_threshold and percentage < severe_theshold) or
-                (not percent_threshold and count >= mild_threshold and count < severe_theshold)
-            ):
-                correct_moderate += 1
-    else:
-        moderate_image_severe_predictions = []
-
-    correct_severe = 0
-    if True:
-        for chip_count in severe_image_severe_predictions:
-            count = chip_count[0]
-            total = chip_count[1]
-            percentage = count / total
-            if (
-                (percent_threshold and percentage >= severe_theshold) or
-                (not percent_threshold and count >= severe_theshold)
-            ):
-                correct_severe += 1
+for thresholds in [
+    # {
+    #     'severe_prediction_prob': 0.65,
+    #     'chip_count_moderate': 750,
+    #     'chip_count_severe': 900,
+    #     'percentage_modearte': 0.095,
+    #     'percentage_severe': 0.01
+    # },
+    {
+        'severe_prediction_prob': 0.85,
+        'chip_count_moderate': 165,
+        'chip_count_severe': 260,
+        'percentage_modearte': 0.02,
+        'percentage_severe': 0.0245
+    },
+    # {
+    #     'severe_prediction_prob': 0.9,
+    #     'chip_count_moderate': 80,
+    #     'chip_count_severe': 180,
+    #     'percentage_modearte': 0.0085,
+    #     'percentage_severe': 0.0125
+    # },
+    # {
+    #     'severe_prediction_prob': 0.95,
+    #     'chip_count_moderate': 20,
+    #     'chip_count_severe': 60,
+    #     'percentage_modearte': 0.0045,
+    #     'percentage_severe': 0.0065
+    # },
+]:
+    severe_prob = thresholds['severe_prediction_prob']
+
+    print("Severe Prediction Probability Threshold: ", severe_prob)
+
+    #Process Mild Results
+    mild_image_severe_predictions = process_results('digpath-results-2', 'Mild', severe_threshold=severe_prob)
+    #Process Severe Results
+    moderate_image_severe_predictions = process_results('digpath-results-2', 'Moderate', severe_threshold=severe_prob)
+    #Process Severe Results
+    severe_image_severe_predictions = process_results('digpath-results-2', 'Severe', severe_threshold=severe_prob)
+
+    #Plot Chip Counts
+    xmax = 1800
+    plt.figure(figsize=(10, 5))
+    plt.xlim(-0.5, xmax)
+    plt.title('Chips with Predicted Eosinophil in Whole Slide Image')
+    if len(mild_image_severe_predictions) > 0 and True:
+        chip_counts = np.copy(mild_image_severe_predictions[:,0])
+        np.putmask(chip_counts, chip_counts >= xmax, xmax + 50)
+        plt.hist(
+            chip_counts,
+            bins=15,
+            color='black',
+            alpha=0.5,
+            label=f'Mild ({len(mild_image_severe_predictions)})',
+        )
+    if len(moderate_image_severe_predictions) > 0 and True:
+        chip_counts = np.copy(moderate_image_severe_predictions[:,0])
+        np.putmask(chip_counts, chip_counts >= xmax, xmax + 50)
+        plt.hist(
+            chip_counts,
+            bins=45,
+            color='tan',
+            alpha=0.5,
+            label=f'Moderate ({len(moderate_image_severe_predictions)})',
+        )
+    if len(severe_image_severe_predictions) > 0 and True:
+        chip_counts = np.copy(severe_image_severe_predictions[:,0])
+        np.putmask(chip_counts, chip_counts >= xmax, xmax + 50)
+        plt.hist(
+            chip_counts,
+            bins=45,
+            color='darkred',
+            alpha=0.5,
+            label=f'Severe ({len(severe_image_severe_predictions)})',
+        )
+
+    #Plot thresholds
+    mild_threshold = thresholds['chip_count_moderate']
+    severe_theshold = thresholds['chip_count_severe']
+
+    y_max = 6
+    y = np.arange(0, y_max+1)
+    plt.ylim(0, y_max)
+    threshold_line = np.ones_like(y)
+    plt.plot(mild_threshold * threshold_line, y, 'k--')
+    plt.plot(severe_theshold * threshold_line, y, 'k--')
+
+    plt.xlabel('Eosinophil Chip Predictions')
+    plt.ylabel('Count')
+    plt.legend()
+    plt.savefig(f'/home/ec2-user/chip_count_evaluation_{int(100*severe_prob)}.jpeg')
+    plt.close()
+
+
+
+
+    #Plot Chip Percentage
+    if severe_prob < 0.8:
+        xmax = 75
     else:
-        severe_image_severe_predictions = []
+        xmax = 30
+    plt.figure(figsize=(10, 5))
+    plt.xlim(-0.5, xmax)
+    plt.title('Percentage of Chips Predicted as Eosinophil in Whole Slide Image')
+    if len(mild_image_severe_predictions) > 0 and True:
+        percentage = 100 * mild_image_severe_predictions[:,0] / mild_image_severe_predictions[:,1]
+        np.putmask(percentage, percentage >= xmax, xmax - 1)
+        plt.hist(
+            percentage,
+            bins=15,
+            color='black',
+            alpha=0.5,
+            label=f'Mild ({len(mild_image_severe_predictions)})',
+        )
+    if len(moderate_image_severe_predictions) > 0 and True:
+        percentage = 100 * moderate_image_severe_predictions[:,0] / moderate_image_severe_predictions[:,1]
+        np.putmask(percentage, percentage >= xmax, xmax - 1)
+        plt.hist(
+            percentage,
+            bins=25,
+            color='tan',
+            alpha=0.5,
+            label=f'Moderate ({len(moderate_image_severe_predictions)})',
+        )
+    if len(severe_image_severe_predictions) > 0 and True:
+        percentage = 100 * severe_image_severe_predictions[:,0] / severe_image_severe_predictions[:,1]
+        np.putmask(percentage, percentage >= xmax, xmax - 1)
+        plt.hist(
+            percentage,
+            bins=25,
+            color='darkred',
+            alpha=0.5,
+            label=f'Severe ({len(severe_image_severe_predictions)})',
+        )
 
-    print(f'Accuracy: {100 * ((correct_mild + correct_moderate + correct_severe) / (len(mild_image_severe_predictions) + len(moderate_image_severe_predictions) + len(severe_image_severe_predictions)))}%')
-    print(f'Mild Accuracy: {100 * ((correct_mild) / (len(mild_image_severe_predictions)))}%')
-    print(f'Moderate Accuracy: {100 * ((correct_moderate) / (len(moderate_image_severe_predictions)))}%')
-    print(f'Severe Accuracy: {100 * ((correct_severe) / (len(severe_image_severe_predictions)))}%')
-    print()
+    #Plot thresholds
+    mild_threshold = thresholds['percentage_modearte']
+    severe_theshold = thresholds['percentage_severe']
+
+    y = np.arange(0, 10)
+    threshold_line = np.ones_like(y)
+    plt.plot(100 * mild_threshold * threshold_line, y, 'k--')
+    plt.plot(100 * severe_theshold * threshold_line, y, 'k--')
+
+    plt.xlabel('Percentage of Eosinophil Chip Predictions')
+    plt.ylabel('Count')
+    plt.legend()
+    plt.savefig(f'/home/ec2-user/chip_percentage_evaluation_{int(100*severe_prob)}.jpeg')
+    plt.close()
+
+
+
+    #Calculate Accuracy and Plot Confusion Matrix
+    for percent_threshold in [False, True]:
+        if percent_threshold:
+            mild_threshold = thresholds['percentage_modearte']
+            severe_theshold = thresholds['percentage_severe']
+            print("Chip Percents")
+        else:
+            mild_threshold = thresholds['chip_count_moderate']
+            severe_theshold = thresholds['chip_count_severe']
+            print("Chip Counts")
+
+        #Count Mild
+        correct_mild = 0
+        correct_moderate = 0
+        correct_severe = 0
+        y_pred = []
+        y = []
+        classes = ('Mild', 'Moderate', 'Severe')
+        for img_label_batch, label in [
+            (mild_image_severe_predictions, 'Mild'),
+            (moderate_image_severe_predictions, 'Moderate'),
+            (severe_image_severe_predictions, 'Severe')
+        ]:
+            for chip_count in img_label_batch:
+                y.append(classes.index(label))
+
+                count = chip_count[0]
+                percentage = count / chip_count[1]
+
+                #Check Mild Prediction
+                if (
+                    (percent_threshold and percentage < mild_threshold) or
+                    (not percent_threshold and count < mild_threshold)
+                ):
+                    y_pred.append(classes.index('Mild'))
+                    if label == 'Mild':
+                        correct_mild += 1
+
+                #Check Moderate Prediction
+                elif (
+                    (percent_threshold and percentage >= mild_threshold and percentage < severe_theshold) or
+                    (not percent_threshold and count >= mild_threshold and count < severe_theshold)
+                ):
+                    y_pred.append(classes.index('Moderate'))
+                    if label == 'Moderate':
+                        correct_moderate += 1
+
+                #Check Severe Prediction
+                elif (
+                    (percent_threshold and percentage >= severe_theshold) or
+                    (not percent_threshold and count >= severe_theshold)
+                ):
+                    y_pred.append(classes.index('Severe'))
+                    if label == 'Severe':
+                        correct_severe += 1
+
+                else:
+                    raise Exception('Bad prediction')
+
+        #Calculate Accuracy
+        accuracy = (
+            (correct_mild + correct_moderate + correct_severe) / 
+            (len(mild_image_severe_predictions) + len(moderate_image_severe_predictions) + len(severe_image_severe_predictions))
+        )
+
+        #Plot confusion matrix
+        conf_matrix = confusion_matrix(y, y_pred, normalize='true')
+
+        plt.figure(figsize=(6,4), dpi=100)
+        sns.set(font_scale = 1.1)
+        ax = sns.heatmap(conf_matrix, annot=True, fmt='.2%', cmap="Blues")
+        ax.set_title(f"Eosinophil Diagnosis (Accuracy: {accuracy:.2%})", fontsize=14, pad=20)
+        ax.set_xlabel("Predicted Diagnosis", fontsize=14, labelpad=20)
+        ax.set_ylabel("Actual Diagnosis", fontsize=14, labelpad=20)
+        ax.xaxis.set_ticklabels(['Mild', 'Moderate', 'Severe'])
+        ax.yaxis.set_ticklabels(['Mild', 'Moderate', 'Severe'])
+
+        if percent_threshold:
+            plt.savefig(f'/home/ec2-user/confusion_matrix_percent_{int(100*severe_prob)}.jpeg')
+        else:
+            plt.savefig(f'/home/ec2-user/confusion_matrix_count_{int(100*severe_prob)}.jpeg')
+
+        print(f'Accuracy: {100 * ((correct_mild + correct_moderate + correct_severe) / (len(mild_image_severe_predictions) + len(moderate_image_severe_predictions) + len(severe_image_severe_predictions)))}%')
+        print(f'Mild Accuracy: {100 * ((correct_mild) / (len(mild_image_severe_predictions)))}%')
+        print(f'Moderate Accuracy: {100 * ((correct_moderate) / (len(moderate_image_severe_predictions)))}%')
+        print(f'Severe Accuracy: {100 * ((correct_severe) / (len(severe_image_severe_predictions)))}%')
+        print()
diff --git a/ml_training/test_e2e.py b/ml_training/test_e2e.py
index 945881c..8eb8422 100644
--- a/ml_training/test_e2e.py
+++ b/ml_training/test_e2e.py
@@ -36,14 +36,17 @@ image_bucket = s3_resource.Bucket(image_s3_bucket_name)
 s3_files = list(image_bucket.objects.filter(Prefix=''))
 random.shuffle(s3_files)
 
-# results_bucket = s3_resource.Bucket('digpath-results-2')
-# results_files = list(results_bucket.objects.filter(Prefix=''))
-results_files = []
+results_bucket = s3_resource.Bucket('digpath-results-2')
+results_files = list(results_bucket.objects.filter(Prefix=''))
+#results_files = []
 
 for s3_obj in s3_files:
     if '.tif' not in s3_obj.key:
         continue
 
+    if 'Moderate' in s3_obj.key and random.random() > 0.5:
+        continue
+
     already_processed = False
     for results in results_files:
         if s3_obj.key.split('/')[-1] in results.key:
-- 
GitLab