{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/anaconda3/envs/dassl/lib/python3.8/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "from scipy import stats\n",
    "import numpy as np\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import monai\n",
    "import torch"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_dices(preds, gt):\n",
    "    intersection = np.sum(gt * preds,axis=(-1,-2))\n",
    "    union = np.sum(gt, axis=(-1,-2)) + np.sum(preds, axis=(-1,-2))\n",
    "    dice = ((2. * intersection + 1)/(union + 1))\n",
    "    return dice\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def t_test(dices_a, dices_b):\n",
    "    differences = dices_a - dices_b\n",
    "    t_statistic, p_value = stats.ttest_rel(dices_a, dices_b)\n",
    "    print(p_value)\n",
    "    alpha = 0.05\n",
    "    if p_value < alpha:\n",
    "        print(\"There is significant difference\")\n",
    "    return p_value "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def t_test_folder(pred_folder1, pred_folder2):\n",
    "    dices_a = []\n",
    "    dices_b = []\n",
    "    try:\n",
    "        for c in os.listdir(pred_folder1):\n",
    "            for i in os.listdir(os.path.join(pred_folder1,c,'rescaled_preds')):\n",
    "                p1 = plt.imread(os.path.join(pred_folder1, c, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "                # plt.imshow(p1, cmap='gray')\n",
    "                # plt.show()\n",
    "                p2 = plt.imread(os.path.join(pred_folder2, c, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "                gt = plt.imread(os.path.join(pred_folder2, c, 'rescaled_gt', i))[80:400,150:500,0]\n",
    "                # print(p1.shape)\n",
    "                # 1/0\n",
    "                dice_a = compute_dices(p1, gt)\n",
    "                dice_b = compute_dices(p2, gt)\n",
    "                dices_a.append(dice_a)\n",
    "                dices_b.append(dice_b)\n",
    "    except:\n",
    "        1/0\n",
    "    return dices_a, dices_b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def t_test_lits2(pred_folder1, pred_folder2):\n",
    "    dices_a = []\n",
    "    dices_b = []\n",
    "    try:\n",
    "        for i in os.listdir(os.path.join(pred_folder1,'rescaled_preds')):\n",
    "            p1 = plt.imread(os.path.join(pred_folder1, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            # plt.imshow(p1, cmap='gray')\n",
    "            # plt.show()\n",
    "            p2 = plt.imread(os.path.join(pred_folder2, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            gt = plt.imread(os.path.join(pred_folder2, 'rescaled_gt', i))[80:400,150:500,0]\n",
    "            # print(p1.shape)\n",
    "            # 1/0\n",
    "            dice_a = compute_dices(p1, gt)\n",
    "            dice_b = compute_dices(p2, gt)\n",
    "            dices_a.append(dice_a)\n",
    "            dices_b.append(dice_b)\n",
    "    except:\n",
    "        1/0\n",
    "    return dices_a, dices_b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4.5704910462821235e-26\n",
      "There is significant difference\n",
      "4.5704910462821235e-26\n"
     ]
    }
   ],
   "source": [
    "#CholecSeg8k\n",
    "folder1 = './cholec8k/svd_shiftscale_cholec_tal_focal075_1e-4'\n",
    "folder2 = './cholec8k/lora_cholec_tmp/'\n",
    "dices_a, dices_b = t_test_folder(folder1, folder2)\n",
    "print(t_test(np.array(dices_a), np.array(dices_b)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.117079926030917e-06\n",
      "There is significant difference\n",
      "2.117079926030917e-06\n"
     ]
    }
   ],
   "source": [
    "#Ultrasound\n",
    "folder1 = '/media/ubuntu/New Volume/jay/ultrasound/abdominal_US/abdominal_US/RUS/SVDSAM'\n",
    "folder2 = '/media/ubuntu/New Volume/jay/ultrasound/abdominal_US/abdominal_US/RUS/Lora_results'\n",
    "dices_a, dices_b = t_test_folder(folder1, folder2)\n",
    "print(t_test(np.array(dices_a), np.array(dices_b)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "5.985398380396225e-14\n",
      "There is significant difference\n",
      "5.985398380396225e-14\n"
     ]
    }
   ],
   "source": [
    "#ChestXDet\n",
    "folder1 = './chestXDet/SVDSAM'\n",
    "folder2 = './chestXDet/lora_chestxdet_tal_focal075_1e-3'\n",
    "dices_a, dices_b = t_test_folder(folder1, folder2)\n",
    "print(t_test(np.array(dices_a), np.array(dices_b)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.00015471901272947407\n",
      "There is significant difference\n",
      "0.00015471901272947407\n"
     ]
    }
   ],
   "source": [
    "#LITS2\n",
    "folder1 = './lits2/svdsam'\n",
    "folder2 = './lits2/lora_lits2/'\n",
    "dices_a, dices_b = t_test_lits2(folder1, folder2)\n",
    "print(t_test(np.array(dices_a), np.array(dices_b)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2.5465986918502266e-23\n",
      "There is significant difference\n",
      "2.5465986918502266e-23\n"
     ]
    }
   ],
   "source": [
    "#GLAS\n",
    "folder1 = './glas/svdshiftscale_glas_tal_CE_1e-4'\n",
    "folder2 = './glas/lorasam'\n",
    "dices_a, dices_b = t_test_lits2(folder1, folder2)\n",
    "print(t_test(np.array(dices_a), np.array(dices_b)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_surface_dice(preds, gt, threshold=[0]):\n",
    "    preds = torch.Tensor(preds).unsqueeze(0).unsqueeze(0)\n",
    "    gt = torch.Tensor(gt).unsqueeze(0).unsqueeze(0)\n",
    "    cd = monai.metrics.compute_surface_dice(preds, gt, threshold, include_background=True)\n",
    "    return cd.item()\n",
    "\n",
    "def compute_hd95(preds, gt):\n",
    "    preds = torch.Tensor(preds).unsqueeze(0).unsqueeze(0)\n",
    "    gt = torch.Tensor(gt).unsqueeze(0).unsqueeze(0)\n",
    "    cd = monai.metrics.compute_hausdorff_distance(preds, gt, include_background=True, percentile=95)\n",
    "    return cd.item()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def surface_dice_lits(pred_folder1, pred_folder2):\n",
    "    dices_a = []\n",
    "    dices_b = []\n",
    "    try:\n",
    "        for i in os.listdir(os.path.join(pred_folder1,'rescaled_preds')):\n",
    "            p1 = plt.imread(os.path.join(pred_folder1, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            # plt.imshow(p1, cmap='gray')\n",
    "            # plt.show()\n",
    "            p2 = plt.imread(os.path.join(pred_folder2, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            gt = plt.imread(os.path.join(pred_folder2, 'rescaled_gt', i))[80:400,150:500,0]\n",
    "            # print(p1.shape)\n",
    "            # 1/0\n",
    "            dice_a = compute_surface_dice(p1, gt)\n",
    "            dice_b = compute_surface_dice(p2, gt)\n",
    "            dices_a.append(dice_a)\n",
    "            dices_b.append(dice_b)\n",
    "    except:\n",
    "        1/0\n",
    "    return np.mean(dices_a), np.mean(dices_b)\n",
    "\n",
    "def hd95_lits(pred_folder1, pred_folder2):\n",
    "    dices_a = []\n",
    "    dices_b = []\n",
    "    try:\n",
    "        for i in os.listdir(os.path.join(pred_folder1,'rescaled_preds')):\n",
    "            p1 = plt.imread(os.path.join(pred_folder1, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            # plt.imshow(p1, cmap='gray')\n",
    "            # plt.show()\n",
    "            p2 = plt.imread(os.path.join(pred_folder2, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            gt = plt.imread(os.path.join(pred_folder2, 'rescaled_gt', i))[80:400,150:500,0]\n",
    "            # print(p1.shape)\n",
    "            # 1/0\n",
    "            dice_a = compute_hd95(p1, gt)\n",
    "            dice_b = compute_hd95(p2, gt)\n",
    "            dices_a.append(dice_a)\n",
    "            dices_b.append(dice_b)\n",
    "    except:\n",
    "        1/0\n",
    "    return np.mean(dices_a), np.mean(dices_b)\n",
    "\n",
    "\n",
    "def surface_dice(pred_folder1, pred_folder2):\n",
    "    dices_a = []\n",
    "    dices_b = []\n",
    "    try:\n",
    "        for c in os.listdir(pred_folder1):\n",
    "            for i in os.listdir(os.path.join(pred_folder1,c,'rescaled_preds')):\n",
    "                p1 = plt.imread(os.path.join(pred_folder1, c, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "                # plt.imshow(p1, cmap='gray')\n",
    "                # plt.show()\n",
    "                p2 = plt.imread(os.path.join(pred_folder2, c, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "                gt = plt.imread(os.path.join(pred_folder2, c, 'rescaled_gt', i))[80:400,150:500,0]\n",
    "                if not gt.any():\n",
    "                    continue\n",
    "                # print(p1.shape)\n",
    "                # 1/0\n",
    "                dice_a = compute_surface_dice(p1, gt)\n",
    "                dice_b = compute_surface_dice(p2, gt)\n",
    "                if not np.isnan(dice_a):\n",
    "                    dices_a.append(dice_a)\n",
    "                if not np.isnan(dice_b):\n",
    "                    dices_b.append(dice_b)\n",
    "    except:\n",
    "        1/0\n",
    "    return np.mean(dices_a), np.mean(dices_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Surface distance of S-SAM:  0.1306660174857825\n",
      "Surface distance of LoRA:  0.19686604533344507\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/anaconda3/envs/dassl/lib/python3.8/site-packages/monai/metrics/utils.py:338: UserWarning: the prediction of class 0 is all 0, this may result in nan/inf distance.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Surface distance of adaptivesam:  0.21469646031036974\n",
      "Surface distance of sam point zs:  0.026642101479228585\n"
     ]
    }
   ],
   "source": [
    "#GLAS\n",
    "folder1 = './glas/svdshiftscale_glas_tal_CE_1e-4'\n",
    "folder2 = './glas/lorasam'\n",
    "folder3 = './glas/adaptivesam'\n",
    "folder4 = './glas/sam_point_glas'\n",
    "dices_a, dices_b = surface_dice_lits(folder1, folder2)\n",
    "print('Surface distance of S-SAM: ', dices_a)\n",
    "print('Surface distance of LoRA: ', dices_b)\n",
    "dices_c, dices_d = surface_dice_lits(folder3, folder4)\n",
    "print('Surface distance of adaptivesam: ', dices_c)\n",
    "print('Surface distance of sam point zs: ', dices_d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of S-SAM:  49.37816700935364\n",
      "HD95 of LoRA:  44.13817956447601\n"
     ]
    }
   ],
   "source": [
    "#GLAS\n",
    "folder1 = './glas/svdshiftscale_glas_tal_CE_1e-4'\n",
    "folder2 = './glas/lorasam'\n",
    "dices_a, dices_b = hd95_lits(folder1, folder2)\n",
    "print('HD95 of S-SAM: ', dices_a)\n",
    "print('HD95 of LoRA: ', dices_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/anaconda3/envs/dassl/lib/python3.8/site-packages/monai/metrics/utils.py:333: UserWarning: the ground truth of class 0 is all 0, this may result in nan/inf distance.\n",
      "  warnings.warn(\n",
      "/home/ubuntu/anaconda3/envs/dassl/lib/python3.8/site-packages/monai/metrics/utils.py:338: UserWarning: the prediction of class 0 is all 0, this may result in nan/inf distance.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Surface distance of S-SAM:  0.1387163305927508\n",
      "Surface distance of LoRA:  0.11434185548909355\n"
     ]
    }
   ],
   "source": [
    "#CholecSeg8k\n",
    "folder1 = './cholec8k/svd_shiftscale_cholec_tal_focal075_1e-4'\n",
    "folder2 = './cholec8k/lora_cholec_tmp/'\n",
    "dices_a, dices_b = surface_dice(folder1, folder2)\n",
    "print('Surface distance of S-SAM: ', dices_a)\n",
    "print('Surface distance of LoRA: ', dices_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "def hd95(folder, trim=True):\n",
    "    hd_a = []\n",
    "    for i in os.listdir(os.path.join(folder,'rescaled_preds')):\n",
    "        if trim:\n",
    "            p1 = plt.imread(os.path.join(folder, 'rescaled_preds', i))[80:400,150:500,0]\n",
    "            gt = plt.imread(os.path.join(folder, 'rescaled_gt', i))[80:400,150:500,0]\n",
    "        else:\n",
    "            p1 = plt.imread(os.path.join(folder, 'rescaled_preds', i))\n",
    "            gt = plt.imread(os.path.join(folder, 'rescaled_gt', i))\n",
    "        hd = compute_hd95(p1, gt)\n",
    "        if np.isnan(hd):\n",
    "            continue\n",
    "        hd_a.append(hd)\n",
    "\n",
    "    return np.mean(hd_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/ubuntu/anaconda3/envs/dassl/lib/python3.8/site-packages/monai/metrics/utils.py:333: UserWarning: the ground truth of class 0 is all 0, this may result in nan/inf distance.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of ISIC lora point:  30.952387651606422\n"
     ]
    }
   ],
   "source": [
    "folder1 = './isic2018/point_lora'\n",
    "hds_a = hd95(folder1)\n",
    "print('HD95 of ISIC lora point: ', hds_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of ISIC VP:  70.9658404827118\n"
     ]
    }
   ],
   "source": [
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/vp'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC VP: ', hds_b)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of Kvasir VP:  99.31089088042577\n"
     ]
    }
   ],
   "source": [
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/vp'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir VP: ', hds_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of refuge VP:  201.66749757289887\n"
     ]
    }
   ],
   "source": [
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/refuge/vp'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of refuge VP: ', hds_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of ev17 VP:  83.76758135225008\n"
     ]
    }
   ],
   "source": [
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/endovis17/vp'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of ev17 VP: ', hds_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of ISIC SAMZS:  85.98172161102295\n",
      "HD95 of Kvasir SAMZS:  99.28112110296885\n",
      "HD95 of refuge SAMZS:  214.98097531795503\n",
      "HD95 of ev17 SAMZS:  89.458785050192\n"
     ]
    }
   ],
   "source": [
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/sam_zs'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC SAMZS: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/sam_zs'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir SAMZS: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/refuge/sam_zs'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of refuge SAMZS: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/endovis17/samzs'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of ev17 SAMZS: ', hds_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of ISIC blackbox:  70.33514058589935\n",
      "HD95 of Kvasir blackbox:  83.55525796731312\n",
      "HD95 of refuge blackbox:  176.66371806144716\n",
      "HD95 of ev17 blackbox:  81.5668385080266\n"
     ]
    }
   ],
   "source": [
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_besttr'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC blackbox: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/model26'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir blackbox: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/refuge/model1'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of refuge blackbox: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/endovis17/model2_tr'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of ev17 blackbox: ', hds_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of ev17 blackbox:  85.75525301522042\n"
     ]
    }
   ],
   "source": [
    "#BlackVIP\n",
    "# folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model2_bestval'\n",
    "# hds_b = hd95(folder2, trim=False)\n",
    "# print('HD95 of ISIC blackbox: ', hds_b)\n",
    "\n",
    "# folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/model23'\n",
    "# hds_b = hd95(folder3, trim=False)\n",
    "# print('HD95 of Kvasir blackbox: ', hds_b)\n",
    "\n",
    "# folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/refuge/model1_tr2'\n",
    "# hds_b = hd95(folder3, trim=False)\n",
    "# print('HD95 of refuge blackbox: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/endovis17/model2_val'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of ev17 blackbox: ', hds_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "ename": "FileNotFoundError",
     "evalue": "[Errno 2] No such file or directory: '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_medsam_tr/rescaled_preds'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[10], line 3\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;66;03m#BlackMedSAM\u001b[39;00m\n\u001b[1;32m      2\u001b[0m folder2 \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_medsam_tr\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[0;32m----> 3\u001b[0m hds_b \u001b[38;5;241m=\u001b[39m \u001b[43mhd95\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfolder2\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mtrim\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m)\u001b[49m\n\u001b[1;32m      4\u001b[0m \u001b[38;5;28mprint\u001b[39m(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mHD95 of ISIC model2: \u001b[39m\u001b[38;5;124m'\u001b[39m, hds_b)\n\u001b[1;32m      6\u001b[0m folder2 \u001b[38;5;241m=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_medsam_tr\u001b[39m\u001b[38;5;124m'\u001b[39m\n",
      "Cell \u001b[0;32mIn[2], line 3\u001b[0m, in \u001b[0;36mhd95\u001b[0;34m(folder, trim)\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mhd95\u001b[39m(folder, trim\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m):\n\u001b[1;32m      2\u001b[0m     hd_a \u001b[38;5;241m=\u001b[39m []\n\u001b[0;32m----> 3\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[43mos\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mlistdir\u001b[49m\u001b[43m(\u001b[49m\u001b[43mos\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpath\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mjoin\u001b[49m\u001b[43m(\u001b[49m\u001b[43mfolder\u001b[49m\u001b[43m,\u001b[49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mrescaled_preds\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m)\u001b[49m:\n\u001b[1;32m      4\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m trim:\n\u001b[1;32m      5\u001b[0m             p1 \u001b[38;5;241m=\u001b[39m plt\u001b[38;5;241m.\u001b[39mimread(os\u001b[38;5;241m.\u001b[39mpath\u001b[38;5;241m.\u001b[39mjoin(folder, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mrescaled_preds\u001b[39m\u001b[38;5;124m'\u001b[39m, i))[\u001b[38;5;241m80\u001b[39m:\u001b[38;5;241m400\u001b[39m,\u001b[38;5;241m150\u001b[39m:\u001b[38;5;241m500\u001b[39m,\u001b[38;5;241m0\u001b[39m]\n",
      "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_medsam_tr/rescaled_preds'"
     ]
    }
   ],
   "source": [
    "#BlackMedSAM\n",
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_medsam_tr'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC model2: ', hds_b)\n",
    "\n",
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/model1_medsam_tr'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC model1: ', hds_b)\n",
    "\n",
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/vp_medsam'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC vp: ', hds_b)\n",
    "\n",
    "folder2 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/isic/medsam_zs'\n",
    "hds_b = hd95(folder2, trim=False)\n",
    "print('HD95 of ISIC medsam-zs: ', hds_b)\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HD95 of Kvasir blackbox medsam model1:  97.05688610394796\n",
      "HD95 of Kvasir blackbox medsam model2:  80.01432559490203\n",
      "HD95 of Kvasir blackbox:  90.36405935287476\n",
      "HD95 of Kvasir blackbox medsam zs:  98.39987009366354\n"
     ]
    }
   ],
   "source": [
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/model1_medsam_besttr'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir blackbox medsam model1: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/model2_medsam_besttr'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir blackbox medsam model2: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/vp_medsam'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir blackbox: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/Kvasir-Seg/medsam_zs'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of Kvasir blackbox medsam zs: ', hds_b)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/refuge/model1_tr2'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of refuge blackbox: ', hds_b)\n",
    "\n",
    "folder3 = '/home/ubuntu/Desktop/Domain_Adaptation_Project/repos/Blackbox/eval/endovis17/model2_val'\n",
    "hds_b = hd95(folder3, trim=False)\n",
    "print('HD95 of ev17 blackbox: ', hds_b)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.16 ('dassl')",
   "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": "3.8.16"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "5b0d24c0401191df5ff06ef3cb04a21077c1fd7ca08d243336ea8a8a1206ff02"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}