examples/swav_iwang_128-new.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.vision.all import *\n",
    "from self_supervised.augmentations import *\n",
    "from self_supervised.layers import *\n",
    "from self_supervised.swav import *\n",
    "from fastai.callback.wandb import WandbCallback\n",
    "import wandb"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Pretraining"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sqrmom=0.99\n",
    "mom=0.95\n",
    "beta=0.\n",
    "eps=1e-4\n",
    "opt_func = partial(ranger, mom=mom, sqr_mom=sqrmom, eps=eps, beta=beta)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dls(size, bs, workers=None):\n",
    "    path = URLs.IMAGEWANG_160 if size <= 160 else URLs.IMAGEWANG\n",
    "    source = untar_data(path)\n",
    "    \n",
    "    files = get_image_files(source)\n",
    "    tfms = [[PILImage.create, ToTensor, RandomResizedCrop(size, min_scale=0.9)], \n",
    "            [parent_label, Categorize()]]\n",
    "    \n",
    "    dsets = Datasets(files, tfms=tfms, splits=RandomSplitter(valid_pct=0.1)(files))\n",
    "    \n",
    "    batch_tfms = [IntToFloatTensor]\n",
    "    dls = dsets.dataloaders(bs=bs, num_workers=workers, after_batch=batch_tfms)\n",
    "    return dls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "bs=128\n",
    "resize, size = 160, 128"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "arch = \"xresnet34\"\n",
    "encoder = create_encoder(arch, pretrained=False, n_in=3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "xtra_config = {\"Arch\":arch, \"Resize\":resize, \"Size\":size, \"Algorithm\":\"SWAV\"}\n",
    "wandb.init(project=\"self-supervised-imagewang\", config=xtra_config)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n",
      "Pipeline: RandomResizedCrop -> RandomHorizontalFlip -> ColorJitter -> RandomGrayscale -> Rotate -- {'size': None, 'mode': 'bilinear', 'pad_mode': 'reflection', 'mode_mask': 'nearest', 'align_corners': True, 'p': 1.0} -> Normalize -- {'mean': tensor([[[[0.4850]],\n",
      "\n",
      "         [[0.4560]],\n",
      "\n",
      "         [[0.4060]]]], device='cuda:0'), 'std': tensor([[[[0.2290]],\n",
      "\n",
      "         [[0.2240]],\n",
      "\n",
      "         [[0.2250]]]], device='cuda:0'), 'axes': (0, 2, 3)}\n"
     ]
    }
   ],
   "source": [
    "dls = get_dls(resize, bs)\n",
    "model = create_swav_model(encoder, n_in=3)\n",
    "learn = Learner(dls, model, SWAVLoss(),\n",
    "                cbs=[SWAV(aug_func=get_batch_augs,\n",
    "                          crop_sizes=[size,96], \n",
    "                          num_crops=[2,6],\n",
    "                          min_scales=[0.25,0.2],\n",
    "                          max_scales=[1.0,0.35],                \n",
    "                          rotate=True,\n",
    "                          rotate_deg=10,\n",
    "                          jitter=True,\n",
    "                          bw=True,\n",
    "                          blur=False\n",
    "                          ),\n",
    "                     TerminateOnNaNCallback(),\n",
    "                     WandbCallback(log_preds=False,log_model=False)\n",
    "                     ])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1728x216 with 8 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "b = dls.one_batch()\n",
    "learn._split(b)\n",
    "learn('before_batch')\n",
    "learn.swav.show_one()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn.to_fp16();"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# learn.lr_find()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "lr=1e-2\n",
    "wd=1e-2\n",
    "epochs=100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>7.562223</td>\n",
       "      <td>7.556176</td>\n",
       "      <td>01:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>7.256415</td>\n",
       "      <td>7.216019</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>7.002712</td>\n",
       "      <td>7.094245</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>6.810123</td>\n",
       "      <td>6.977897</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>6.670819</td>\n",
       "      <td>6.838521</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>6.498086</td>\n",
       "      <td>6.639339</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>6.344620</td>\n",
       "      <td>6.492225</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>6.203028</td>\n",
       "      <td>6.306142</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>6.122333</td>\n",
       "      <td>6.169743</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>6.027206</td>\n",
       "      <td>6.147151</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>5.983262</td>\n",
       "      <td>6.153629</td>\n",
       "      <td>01:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>5.901104</td>\n",
       "      <td>6.120547</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>5.864655</td>\n",
       "      <td>6.068148</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>5.855183</td>\n",
       "      <td>6.004277</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>5.787126</td>\n",
       "      <td>6.037642</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>5.732891</td>\n",
       "      <td>5.946775</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>5.717632</td>\n",
       "      <td>5.945467</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>5.681522</td>\n",
       "      <td>5.823079</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>5.669751</td>\n",
       "      <td>5.853506</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>5.618251</td>\n",
       "      <td>5.794825</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>5.597249</td>\n",
       "      <td>5.812010</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>5.582451</td>\n",
       "      <td>5.749178</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>5.558737</td>\n",
       "      <td>5.671377</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>5.545779</td>\n",
       "      <td>5.727174</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>5.531798</td>\n",
       "      <td>5.747025</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>5.495986</td>\n",
       "      <td>5.690720</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>5.472614</td>\n",
       "      <td>5.718486</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>5.475045</td>\n",
       "      <td>5.669025</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>5.438272</td>\n",
       "      <td>5.693813</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>5.470737</td>\n",
       "      <td>5.655054</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>5.452989</td>\n",
       "      <td>5.645219</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>5.432491</td>\n",
       "      <td>5.648370</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>5.406768</td>\n",
       "      <td>5.583323</td>\n",
       "      <td>01:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>5.393102</td>\n",
       "      <td>5.637488</td>\n",
       "      <td>01:23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>5.402245</td>\n",
       "      <td>5.573010</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>5.398229</td>\n",
       "      <td>5.525831</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>36</td>\n",
       "      <td>5.368255</td>\n",
       "      <td>5.508889</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>5.374063</td>\n",
       "      <td>5.608894</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>5.360675</td>\n",
       "      <td>5.648408</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>5.343800</td>\n",
       "      <td>5.577901</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>5.350117</td>\n",
       "      <td>5.540042</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>5.324234</td>\n",
       "      <td>5.561817</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>42</td>\n",
       "      <td>5.300221</td>\n",
       "      <td>5.501461</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>43</td>\n",
       "      <td>5.308800</td>\n",
       "      <td>5.493710</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>44</td>\n",
       "      <td>5.281006</td>\n",
       "      <td>5.538874</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45</td>\n",
       "      <td>5.291098</td>\n",
       "      <td>5.535818</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>46</td>\n",
       "      <td>5.296422</td>\n",
       "      <td>5.622439</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>47</td>\n",
       "      <td>5.283659</td>\n",
       "      <td>5.525898</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>48</td>\n",
       "      <td>5.280865</td>\n",
       "      <td>5.514113</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>49</td>\n",
       "      <td>5.238877</td>\n",
       "      <td>5.518435</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>5.227177</td>\n",
       "      <td>5.582239</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>51</td>\n",
       "      <td>5.253059</td>\n",
       "      <td>5.576575</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52</td>\n",
       "      <td>5.271305</td>\n",
       "      <td>5.472447</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>53</td>\n",
       "      <td>5.257043</td>\n",
       "      <td>5.462628</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>54</td>\n",
       "      <td>5.239423</td>\n",
       "      <td>5.509766</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>55</td>\n",
       "      <td>5.222016</td>\n",
       "      <td>5.558569</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>56</td>\n",
       "      <td>5.213808</td>\n",
       "      <td>5.415207</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>57</td>\n",
       "      <td>5.189425</td>\n",
       "      <td>5.523403</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>58</td>\n",
       "      <td>5.184707</td>\n",
       "      <td>5.405925</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>59</td>\n",
       "      <td>5.178570</td>\n",
       "      <td>5.528091</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>5.172749</td>\n",
       "      <td>5.449420</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61</td>\n",
       "      <td>5.131134</td>\n",
       "      <td>5.353100</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>62</td>\n",
       "      <td>5.164092</td>\n",
       "      <td>5.420002</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>63</td>\n",
       "      <td>5.126656</td>\n",
       "      <td>5.428546</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>64</td>\n",
       "      <td>5.110462</td>\n",
       "      <td>5.408085</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>65</td>\n",
       "      <td>5.097446</td>\n",
       "      <td>5.393282</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>66</td>\n",
       "      <td>5.111385</td>\n",
       "      <td>5.430235</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>67</td>\n",
       "      <td>5.081972</td>\n",
       "      <td>5.425187</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>68</td>\n",
       "      <td>5.090536</td>\n",
       "      <td>5.380910</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>69</td>\n",
       "      <td>5.076035</td>\n",
       "      <td>5.412733</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>70</td>\n",
       "      <td>5.087616</td>\n",
       "      <td>5.440988</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>71</td>\n",
       "      <td>5.031078</td>\n",
       "      <td>5.324538</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>72</td>\n",
       "      <td>5.032365</td>\n",
       "      <td>5.323254</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>73</td>\n",
       "      <td>5.029209</td>\n",
       "      <td>5.269480</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>74</td>\n",
       "      <td>5.003679</td>\n",
       "      <td>5.341503</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>4.980991</td>\n",
       "      <td>5.318098</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>76</td>\n",
       "      <td>4.972736</td>\n",
       "      <td>5.348806</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77</td>\n",
       "      <td>4.938922</td>\n",
       "      <td>5.367326</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>4.921985</td>\n",
       "      <td>5.278971</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>79</td>\n",
       "      <td>4.921616</td>\n",
       "      <td>5.319994</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>80</td>\n",
       "      <td>4.905596</td>\n",
       "      <td>5.275771</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>81</td>\n",
       "      <td>4.870277</td>\n",
       "      <td>5.301620</td>\n",
       "      <td>01:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>82</td>\n",
       "      <td>4.865699</td>\n",
       "      <td>5.279308</td>\n",
       "      <td>01:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>83</td>\n",
       "      <td>4.863782</td>\n",
       "      <td>5.234282</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84</td>\n",
       "      <td>4.841960</td>\n",
       "      <td>5.267246</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>85</td>\n",
       "      <td>4.835383</td>\n",
       "      <td>5.289878</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>86</td>\n",
       "      <td>4.819319</td>\n",
       "      <td>5.246760</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>87</td>\n",
       "      <td>4.811168</td>\n",
       "      <td>5.240457</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>88</td>\n",
       "      <td>4.780126</td>\n",
       "      <td>5.318847</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>89</td>\n",
       "      <td>4.808592</td>\n",
       "      <td>5.245132</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90</td>\n",
       "      <td>4.798241</td>\n",
       "      <td>5.216918</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>91</td>\n",
       "      <td>4.763688</td>\n",
       "      <td>5.198220</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>92</td>\n",
       "      <td>4.759356</td>\n",
       "      <td>5.221266</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93</td>\n",
       "      <td>4.728040</td>\n",
       "      <td>5.233225</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>94</td>\n",
       "      <td>4.748582</td>\n",
       "      <td>5.192582</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>95</td>\n",
       "      <td>4.739158</td>\n",
       "      <td>5.257190</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96</td>\n",
       "      <td>4.732437</td>\n",
       "      <td>5.159089</td>\n",
       "      <td>01:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>97</td>\n",
       "      <td>4.740786</td>\n",
       "      <td>5.217196</td>\n",
       "      <td>01:24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>98</td>\n",
       "      <td>4.728659</td>\n",
       "      <td>5.217085</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>99</td>\n",
       "      <td>4.738997</td>\n",
       "      <td>5.216799</td>\n",
       "      <td>01:26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.unfreeze()\n",
    "learn.fit_flat_cos(epochs, lr, wd=wd, pct_start=0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<br/>Waiting for W&B process to finish, PID 4089442<br/>Program ended successfully."
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "384b0bb44b0d41079128d864211474e3",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "VBox(children=(Label(value=' 0.02MB of 0.02MB uploaded (0.00MB deduped)\\r'), FloatProgress(value=1.0, max=1.0)…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Find user logs for this run at: <code>/home/code-base/extra_space/self_supervised/examples/wandb/run-20210207_203318-219s8lgo/logs/debug.log</code>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Find internal logs for this run at: <code>/home/code-base/extra_space/self_supervised/examples/wandb/run-20210207_203318-219s8lgo/logs/debug-internal.log</code>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h3>Run summary:</h3><br/><style>\n",
       "    table.wandb td:nth-child(1) { padding: 0 10px; text-align: right }\n",
       "    </style><table class=\"wandb\">\n",
       "<tr><td>epoch</td><td>100</td></tr><tr><td>train_loss</td><td>4.739</td></tr><tr><td>raw_loss</td><td>4.62989</td></tr><tr><td>wd_0</td><td>0.01</td></tr><tr><td>sqr_mom_0</td><td>0.99</td></tr><tr><td>lr_0</td><td>0.0</td></tr><tr><td>mom_0</td><td>0.9</td></tr><tr><td>eps_0</td><td>1e-05</td></tr><tr><td>_step</td><td>18499</td></tr><tr><td>_runtime</td><td>8546</td></tr><tr><td>_timestamp</td><td>1612738544</td></tr><tr><td>valid_loss</td><td>5.2168</td></tr></table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<h3>Run history:</h3><br/><style>\n",
       "    table.wandb td:nth-child(1) { padding: 0 10px; text-align: right }\n",
       "    </style><table class=\"wandb\">\n",
       "<tr><td>epoch</td><td>▁▁▁▁▂▂▂▂▂▃▃▃▃▃▃▄▄▄▄▄▅▅▅▅▅▅▆▆▆▆▆▇▇▇▇▇▇███</td></tr><tr><td>train_loss</td><td>█▇▆▅▄▄▄▃▃▃▃▃▃▃▃▃▂▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▁▁▁▁▁▁▁▁</td></tr><tr><td>raw_loss</td><td>██▆▅▄▄▄▄▃▄▃▃▃▃▃▃▃▃▃▄▃▃▃▃▃▂▂▃▃▂▂▂▂▂▂▁▁▁▁▁</td></tr><tr><td>wd_0</td><td>▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁</td></tr><tr><td>sqr_mom_0</td><td>▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁</td></tr><tr><td>lr_0</td><td>████████████████████████▇▇▆▆▅▅▄▄▃▃▂▂▂▁▁▁</td></tr><tr><td>mom_0</td><td>▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁</td></tr><tr><td>eps_0</td><td>▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁</td></tr><tr><td>_step</td><td>▁▁▁▁▂▂▂▂▂▃▃▃▃▃▃▄▄▄▄▄▅▅▅▅▅▅▆▆▆▆▆▇▇▇▇▇▇███</td></tr><tr><td>_runtime</td><td>▁▁▁▁▂▂▂▂▂▃▃▃▃▃▃▄▄▄▄▄▅▅▅▅▅▅▆▆▆▆▆▇▇▇▇▇▇███</td></tr><tr><td>_timestamp</td><td>▁▁▁▁▂▂▂▂▂▃▃▃▃▃▃▄▄▄▄▄▅▅▅▅▅▅▆▆▆▆▆▇▇▇▇▇▇███</td></tr><tr><td>valid_loss</td><td>█▇▅▄▄▄▃▃▃▂▃▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▁▁▂▁▁▁▁▁▁▁▁▁</td></tr></table><br/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Synced 5 W&B file(s), 1 media file(s), 0 artifact file(s) and 0 other file(s)"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "\n",
       "                    <br/>Synced <strong style=\"color:#cdcd00\">earnest-water-2</strong>: <a href=\"https://wandb.ai/keremturgutlu/self-supervised-imagewang/runs/219s8lgo\" target=\"_blank\">https://wandb.ai/keremturgutlu/self-supervised-imagewang/runs/219s8lgo</a><br/>\n",
       "                "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "wandb.finish()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "save_name = f'swav_iwang_sz{size}_epc{epochs}'\n",
    "learn.save(save_name)\n",
    "torch.save(learn.model.encoder.state_dict(), learn.path/learn.model_dir/f'{save_name}_encoder.pth')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.recorder.plot_loss()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Downstream Task"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(128, 128)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bs, size"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_dls(size, bs, workers=None):\n",
    "    path = URLs.IMAGEWANG_160 if size <= 160 else URLs.IMAGEWANG\n",
    "    source = untar_data(path)\n",
    "    files = get_image_files(source, folders=['train', 'val'])\n",
    "    splits = GrandparentSplitter(valid_name='val')(files)\n",
    "    \n",
    "    item_aug = [RandomResizedCrop(size, min_scale=0.35), FlipItem(0.5)]\n",
    "    tfms = [[PILImage.create, ToTensor, *item_aug], \n",
    "            [parent_label, Categorize()]]\n",
    "    \n",
    "    dsets = Datasets(files, tfms=tfms, splits=splits)\n",
    "    \n",
    "    batch_tfms = [IntToFloatTensor, Normalize.from_stats(*imagenet_stats)]\n",
    "    dls = dsets.dataloaders(bs=bs, num_workers=workers, after_batch=batch_tfms)\n",
    "    return dls"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "def split_func(m): return L(m[0], m[1]).map(params)\n",
    "\n",
    "def create_learner(size=size, arch='xresnet34', encoder_path=\"models/swav_iwang_sz128_epc100_encoder.pth\"):\n",
    "    \n",
    "    dls = get_dls(size, bs=bs//2)\n",
    "    pretrained_encoder = torch.load(encoder_path)\n",
    "    encoder = create_encoder(arch, pretrained=False, n_in=3)\n",
    "    encoder.load_state_dict(pretrained_encoder)\n",
    "    nf = encoder(torch.randn(2,3,224,224)).size(-1)\n",
    "    classifier = create_cls_module(nf, dls.c)\n",
    "    model = nn.Sequential(encoder, classifier)\n",
    "    learn = Learner(dls, model, opt_func=opt_func, splitter=split_func,\n",
    "                metrics=[accuracy,top_k_accuracy], loss_func=LabelSmoothingCrossEntropy())\n",
    "    return learn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def finetune(size, epochs, arch, encoder_path, lr=1e-2, wd=1e-2):\n",
    "    learn = create_learner(size, arch, encoder_path)\n",
    "    learn.unfreeze()\n",
    "    learn.fit_flat_cos(epochs, lr, wd=wd)\n",
    "    final_acc = learn.recorder.values[-1][-2]\n",
    "    return final_acc"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 5 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.004153</td>\n",
       "      <td>1.726009</td>\n",
       "      <td>0.634258</td>\n",
       "      <td>0.927208</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.952761</td>\n",
       "      <td>1.601531</td>\n",
       "      <td>0.666073</td>\n",
       "      <td>0.914228</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.902247</td>\n",
       "      <td>1.522821</td>\n",
       "      <td>0.684398</td>\n",
       "      <td>0.932553</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.852205</td>\n",
       "      <td>1.578140</td>\n",
       "      <td>0.676763</td>\n",
       "      <td>0.917282</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.776530</td>\n",
       "      <td>1.387038</td>\n",
       "      <td>0.733265</td>\n",
       "      <td>0.944261</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.992634</td>\n",
       "      <td>1.793369</td>\n",
       "      <td>0.635022</td>\n",
       "      <td>0.903283</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.953077</td>\n",
       "      <td>1.641586</td>\n",
       "      <td>0.659710</td>\n",
       "      <td>0.912700</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.913724</td>\n",
       "      <td>1.780718</td>\n",
       "      <td>0.600407</td>\n",
       "      <td>0.878850</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.853015</td>\n",
       "      <td>1.528081</td>\n",
       "      <td>0.688979</td>\n",
       "      <td>0.929244</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.782257</td>\n",
       "      <td>1.390942</td>\n",
       "      <td>0.722576</td>\n",
       "      <td>0.938152</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.987617</td>\n",
       "      <td>1.695586</td>\n",
       "      <td>0.664546</td>\n",
       "      <td>0.916264</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.936876</td>\n",
       "      <td>1.656744</td>\n",
       "      <td>0.659201</td>\n",
       "      <td>0.907610</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.896482</td>\n",
       "      <td>1.630533</td>\n",
       "      <td>0.655383</td>\n",
       "      <td>0.913464</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.862476</td>\n",
       "      <td>1.531032</td>\n",
       "      <td>0.687198</td>\n",
       "      <td>0.930008</td>\n",
       "      <td>00:29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.771756</td>\n",
       "      <td>1.360692</td>\n",
       "      <td>0.735302</td>\n",
       "      <td>0.942988</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.992702</td>\n",
       "      <td>1.851998</td>\n",
       "      <td>0.611861</td>\n",
       "      <td>0.900484</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.956672</td>\n",
       "      <td>1.551878</td>\n",
       "      <td>0.671672</td>\n",
       "      <td>0.928226</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.899147</td>\n",
       "      <td>1.645386</td>\n",
       "      <td>0.654620</td>\n",
       "      <td>0.901756</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.859878</td>\n",
       "      <td>1.540646</td>\n",
       "      <td>0.667091</td>\n",
       "      <td>0.930771</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.772086</td>\n",
       "      <td>1.391685</td>\n",
       "      <td>0.727921</td>\n",
       "      <td>0.939679</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.006662</td>\n",
       "      <td>1.856413</td>\n",
       "      <td>0.620005</td>\n",
       "      <td>0.894884</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.948674</td>\n",
       "      <td>1.674155</td>\n",
       "      <td>0.643675</td>\n",
       "      <td>0.910155</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.911278</td>\n",
       "      <td>1.632882</td>\n",
       "      <td>0.635531</td>\n",
       "      <td>0.899466</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.865876</td>\n",
       "      <td>1.699242</td>\n",
       "      <td>0.632985</td>\n",
       "      <td>0.892848</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.777478</td>\n",
       "      <td>1.380007</td>\n",
       "      <td>0.728175</td>\n",
       "      <td>0.941715</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "acc = []\n",
    "runs = 5\n",
    "for i in range(runs): acc += [finetune(size, epochs=5, arch='xresnet34', encoder_path=f'models/swav_iwang_sz{size}_epc100_encoder.pth')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7294476985931396"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 20 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.001759</td>\n",
       "      <td>1.818392</td>\n",
       "      <td>0.609061</td>\n",
       "      <td>0.895139</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.956160</td>\n",
       "      <td>1.561072</td>\n",
       "      <td>0.682362</td>\n",
       "      <td>0.930771</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.909854</td>\n",
       "      <td>1.544676</td>\n",
       "      <td>0.671163</td>\n",
       "      <td>0.920336</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.866563</td>\n",
       "      <td>1.680852</td>\n",
       "      <td>0.638585</td>\n",
       "      <td>0.900993</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.818689</td>\n",
       "      <td>1.481085</td>\n",
       "      <td>0.704505</td>\n",
       "      <td>0.918809</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.807902</td>\n",
       "      <td>1.541274</td>\n",
       "      <td>0.681853</td>\n",
       "      <td>0.917027</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.794057</td>\n",
       "      <td>1.541773</td>\n",
       "      <td>0.684653</td>\n",
       "      <td>0.925681</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.782262</td>\n",
       "      <td>1.500459</td>\n",
       "      <td>0.700687</td>\n",
       "      <td>0.929753</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.756448</td>\n",
       "      <td>1.586472</td>\n",
       "      <td>0.665055</td>\n",
       "      <td>0.913973</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.756565</td>\n",
       "      <td>1.730361</td>\n",
       "      <td>0.640366</td>\n",
       "      <td>0.894884</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.741618</td>\n",
       "      <td>1.538961</td>\n",
       "      <td>0.688216</td>\n",
       "      <td>0.920845</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.733222</td>\n",
       "      <td>1.649867</td>\n",
       "      <td>0.667091</td>\n",
       "      <td>0.911937</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0.736973</td>\n",
       "      <td>1.688843</td>\n",
       "      <td>0.646475</td>\n",
       "      <td>0.910664</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0.721703</td>\n",
       "      <td>1.677604</td>\n",
       "      <td>0.663019</td>\n",
       "      <td>0.912191</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0.730637</td>\n",
       "      <td>1.734889</td>\n",
       "      <td>0.638839</td>\n",
       "      <td>0.894630</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>0.707327</td>\n",
       "      <td>1.533894</td>\n",
       "      <td>0.698906</td>\n",
       "      <td>0.921863</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0.695737</td>\n",
       "      <td>1.611027</td>\n",
       "      <td>0.674981</td>\n",
       "      <td>0.911173</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>0.672729</td>\n",
       "      <td>1.487459</td>\n",
       "      <td>0.708323</td>\n",
       "      <td>0.927717</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0.654346</td>\n",
       "      <td>1.434375</td>\n",
       "      <td>0.721558</td>\n",
       "      <td>0.929499</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>0.642734</td>\n",
       "      <td>1.449689</td>\n",
       "      <td>0.720794</td>\n",
       "      <td>0.931026</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.018685</td>\n",
       "      <td>1.887084</td>\n",
       "      <td>0.604479</td>\n",
       "      <td>0.894121</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.948562</td>\n",
       "      <td>1.585955</td>\n",
       "      <td>0.656401</td>\n",
       "      <td>0.917536</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.913015</td>\n",
       "      <td>1.728321</td>\n",
       "      <td>0.628659</td>\n",
       "      <td>0.886740</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.856341</td>\n",
       "      <td>1.555931</td>\n",
       "      <td>0.677017</td>\n",
       "      <td>0.913719</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.831206</td>\n",
       "      <td>1.489864</td>\n",
       "      <td>0.703487</td>\n",
       "      <td>0.924663</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.805444</td>\n",
       "      <td>1.458251</td>\n",
       "      <td>0.705014</td>\n",
       "      <td>0.930262</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.782979</td>\n",
       "      <td>1.536513</td>\n",
       "      <td>0.697633</td>\n",
       "      <td>0.926699</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.788585</td>\n",
       "      <td>1.460763</td>\n",
       "      <td>0.704759</td>\n",
       "      <td>0.932807</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.757668</td>\n",
       "      <td>1.569479</td>\n",
       "      <td>0.680326</td>\n",
       "      <td>0.930262</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.755905</td>\n",
       "      <td>1.658669</td>\n",
       "      <td>0.652074</td>\n",
       "      <td>0.911682</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.742875</td>\n",
       "      <td>1.556949</td>\n",
       "      <td>0.679562</td>\n",
       "      <td>0.909392</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.736641</td>\n",
       "      <td>1.448801</td>\n",
       "      <td>0.720031</td>\n",
       "      <td>0.931026</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0.737124</td>\n",
       "      <td>1.529505</td>\n",
       "      <td>0.685416</td>\n",
       "      <td>0.917282</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0.733549</td>\n",
       "      <td>1.585351</td>\n",
       "      <td>0.688470</td>\n",
       "      <td>0.919572</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0.719247</td>\n",
       "      <td>1.491304</td>\n",
       "      <td>0.695851</td>\n",
       "      <td>0.926444</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>0.710253</td>\n",
       "      <td>1.623216</td>\n",
       "      <td>0.672436</td>\n",
       "      <td>0.918300</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0.698034</td>\n",
       "      <td>1.490301</td>\n",
       "      <td>0.703487</td>\n",
       "      <td>0.928990</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>0.668760</td>\n",
       "      <td>1.458111</td>\n",
       "      <td>0.715449</td>\n",
       "      <td>0.930517</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0.652986</td>\n",
       "      <td>1.436692</td>\n",
       "      <td>0.724866</td>\n",
       "      <td>0.929244</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>0.648907</td>\n",
       "      <td>1.447760</td>\n",
       "      <td>0.719012</td>\n",
       "      <td>0.928226</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.009940</td>\n",
       "      <td>1.745073</td>\n",
       "      <td>0.649020</td>\n",
       "      <td>0.916264</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.942719</td>\n",
       "      <td>1.670455</td>\n",
       "      <td>0.660473</td>\n",
       "      <td>0.914482</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.903281</td>\n",
       "      <td>1.609442</td>\n",
       "      <td>0.665818</td>\n",
       "      <td>0.902011</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.865467</td>\n",
       "      <td>1.554827</td>\n",
       "      <td>0.675999</td>\n",
       "      <td>0.924154</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.827116</td>\n",
       "      <td>1.676045</td>\n",
       "      <td>0.650293</td>\n",
       "      <td>0.919063</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.807941</td>\n",
       "      <td>1.534003</td>\n",
       "      <td>0.687198</td>\n",
       "      <td>0.905065</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.782891</td>\n",
       "      <td>1.458077</td>\n",
       "      <td>0.696869</td>\n",
       "      <td>0.926699</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.779451</td>\n",
       "      <td>1.580788</td>\n",
       "      <td>0.665564</td>\n",
       "      <td>0.913719</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.763879</td>\n",
       "      <td>1.575968</td>\n",
       "      <td>0.666073</td>\n",
       "      <td>0.917282</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.759454</td>\n",
       "      <td>1.603148</td>\n",
       "      <td>0.667854</td>\n",
       "      <td>0.924154</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.738754</td>\n",
       "      <td>1.628122</td>\n",
       "      <td>0.661237</td>\n",
       "      <td>0.900993</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.740608</td>\n",
       "      <td>1.481813</td>\n",
       "      <td>0.706287</td>\n",
       "      <td>0.930771</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0.721303</td>\n",
       "      <td>1.461213</td>\n",
       "      <td>0.712650</td>\n",
       "      <td>0.935607</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0.731538</td>\n",
       "      <td>1.589221</td>\n",
       "      <td>0.675744</td>\n",
       "      <td>0.918809</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0.718515</td>\n",
       "      <td>1.625018</td>\n",
       "      <td>0.668872</td>\n",
       "      <td>0.918554</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>0.720691</td>\n",
       "      <td>1.592125</td>\n",
       "      <td>0.679308</td>\n",
       "      <td>0.909901</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0.701939</td>\n",
       "      <td>1.628794</td>\n",
       "      <td>0.663782</td>\n",
       "      <td>0.911937</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>0.672540</td>\n",
       "      <td>1.429808</td>\n",
       "      <td>0.723594</td>\n",
       "      <td>0.931789</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0.651598</td>\n",
       "      <td>1.398880</td>\n",
       "      <td>0.728939</td>\n",
       "      <td>0.935353</td>\n",
       "      <td>00:30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>0.646542</td>\n",
       "      <td>1.405867</td>\n",
       "      <td>0.725630</td>\n",
       "      <td>0.931789</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.7218121687571207"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acc = []\n",
    "runs = 3\n",
    "for i in range(runs): acc += [finetune(size, epochs=20, arch='xresnet34', encoder_path=f'models/swav_iwang_sz{size}_epc100_encoder.pth')]\n",
    "np.mean(acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 80 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.987533</td>\n",
       "      <td>1.751684</td>\n",
       "      <td>0.633495</td>\n",
       "      <td>0.920590</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.942716</td>\n",
       "      <td>1.630277</td>\n",
       "      <td>0.659201</td>\n",
       "      <td>0.914228</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.903545</td>\n",
       "      <td>1.450592</td>\n",
       "      <td>0.710613</td>\n",
       "      <td>0.935353</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.857153</td>\n",
       "      <td>1.608183</td>\n",
       "      <td>0.657165</td>\n",
       "      <td>0.924154</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.833454</td>\n",
       "      <td>1.570727</td>\n",
       "      <td>0.681089</td>\n",
       "      <td>0.915755</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.806094</td>\n",
       "      <td>1.519949</td>\n",
       "      <td>0.688470</td>\n",
       "      <td>0.923899</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.796067</td>\n",
       "      <td>1.490939</td>\n",
       "      <td>0.705523</td>\n",
       "      <td>0.929753</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.776064</td>\n",
       "      <td>1.544555</td>\n",
       "      <td>0.684907</td>\n",
       "      <td>0.927717</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.764636</td>\n",
       "      <td>1.525187</td>\n",
       "      <td>0.691270</td>\n",
       "      <td>0.912700</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.755392</td>\n",
       "      <td>1.620285</td>\n",
       "      <td>0.679817</td>\n",
       "      <td>0.924408</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.739467</td>\n",
       "      <td>1.663396</td>\n",
       "      <td>0.667091</td>\n",
       "      <td>0.914737</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.748029</td>\n",
       "      <td>1.471423</td>\n",
       "      <td>0.701705</td>\n",
       "      <td>0.924663</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0.732373</td>\n",
       "      <td>1.538525</td>\n",
       "      <td>0.695342</td>\n",
       "      <td>0.914228</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0.729599</td>\n",
       "      <td>1.599101</td>\n",
       "      <td>0.671163</td>\n",
       "      <td>0.923390</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0.716881</td>\n",
       "      <td>1.568462</td>\n",
       "      <td>0.686434</td>\n",
       "      <td>0.913464</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>0.721436</td>\n",
       "      <td>1.546810</td>\n",
       "      <td>0.688216</td>\n",
       "      <td>0.920081</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0.712635</td>\n",
       "      <td>1.693091</td>\n",
       "      <td>0.662510</td>\n",
       "      <td>0.897429</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>0.724263</td>\n",
       "      <td>1.557871</td>\n",
       "      <td>0.684653</td>\n",
       "      <td>0.906083</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0.720896</td>\n",
       "      <td>1.595457</td>\n",
       "      <td>0.667091</td>\n",
       "      <td>0.914228</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>0.712949</td>\n",
       "      <td>1.660906</td>\n",
       "      <td>0.665309</td>\n",
       "      <td>0.900229</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>0.704695</td>\n",
       "      <td>1.572611</td>\n",
       "      <td>0.679562</td>\n",
       "      <td>0.909392</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>0.696917</td>\n",
       "      <td>1.626726</td>\n",
       "      <td>0.675235</td>\n",
       "      <td>0.902265</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>0.697966</td>\n",
       "      <td>1.819760</td>\n",
       "      <td>0.619751</td>\n",
       "      <td>0.887503</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>0.708806</td>\n",
       "      <td>1.693701</td>\n",
       "      <td>0.644948</td>\n",
       "      <td>0.899975</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>0.707584</td>\n",
       "      <td>1.604366</td>\n",
       "      <td>0.677272</td>\n",
       "      <td>0.910919</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>0.697424</td>\n",
       "      <td>1.705523</td>\n",
       "      <td>0.644693</td>\n",
       "      <td>0.907101</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>0.702011</td>\n",
       "      <td>1.594821</td>\n",
       "      <td>0.672945</td>\n",
       "      <td>0.912700</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>0.698147</td>\n",
       "      <td>1.682556</td>\n",
       "      <td>0.661746</td>\n",
       "      <td>0.901502</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>0.696324</td>\n",
       "      <td>1.776768</td>\n",
       "      <td>0.643675</td>\n",
       "      <td>0.875795</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>0.694808</td>\n",
       "      <td>1.709992</td>\n",
       "      <td>0.656910</td>\n",
       "      <td>0.902774</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>0.697197</td>\n",
       "      <td>1.731780</td>\n",
       "      <td>0.650038</td>\n",
       "      <td>0.885467</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>0.699653</td>\n",
       "      <td>1.790499</td>\n",
       "      <td>0.638585</td>\n",
       "      <td>0.884449</td>\n",
       "      <td>00:29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>0.696338</td>\n",
       "      <td>1.686947</td>\n",
       "      <td>0.645711</td>\n",
       "      <td>0.886740</td>\n",
       "      <td>00:25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>0.710225</td>\n",
       "      <td>1.741390</td>\n",
       "      <td>0.629168</td>\n",
       "      <td>0.877322</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>0.683217</td>\n",
       "      <td>1.818089</td>\n",
       "      <td>0.625350</td>\n",
       "      <td>0.867142</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>0.690022</td>\n",
       "      <td>1.733344</td>\n",
       "      <td>0.644439</td>\n",
       "      <td>0.892084</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>36</td>\n",
       "      <td>0.698640</td>\n",
       "      <td>1.682131</td>\n",
       "      <td>0.648511</td>\n",
       "      <td>0.893612</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>0.694628</td>\n",
       "      <td>1.740354</td>\n",
       "      <td>0.631204</td>\n",
       "      <td>0.893357</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>0.685400</td>\n",
       "      <td>1.769358</td>\n",
       "      <td>0.634513</td>\n",
       "      <td>0.892339</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>0.692796</td>\n",
       "      <td>1.675133</td>\n",
       "      <td>0.653092</td>\n",
       "      <td>0.898702</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>0.686051</td>\n",
       "      <td>1.829152</td>\n",
       "      <td>0.622041</td>\n",
       "      <td>0.879868</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>0.689256</td>\n",
       "      <td>1.777129</td>\n",
       "      <td>0.634258</td>\n",
       "      <td>0.895139</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>42</td>\n",
       "      <td>0.692958</td>\n",
       "      <td>1.712580</td>\n",
       "      <td>0.643166</td>\n",
       "      <td>0.888267</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>43</td>\n",
       "      <td>0.684958</td>\n",
       "      <td>1.696333</td>\n",
       "      <td>0.648766</td>\n",
       "      <td>0.905065</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>44</td>\n",
       "      <td>0.688395</td>\n",
       "      <td>1.716708</td>\n",
       "      <td>0.646220</td>\n",
       "      <td>0.893612</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45</td>\n",
       "      <td>0.689520</td>\n",
       "      <td>1.881840</td>\n",
       "      <td>0.612879</td>\n",
       "      <td>0.885467</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>46</td>\n",
       "      <td>0.687738</td>\n",
       "      <td>1.774152</td>\n",
       "      <td>0.634767</td>\n",
       "      <td>0.874523</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>47</td>\n",
       "      <td>0.691658</td>\n",
       "      <td>1.830658</td>\n",
       "      <td>0.628404</td>\n",
       "      <td>0.872996</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>48</td>\n",
       "      <td>0.691930</td>\n",
       "      <td>1.788018</td>\n",
       "      <td>0.632476</td>\n",
       "      <td>0.883176</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>49</td>\n",
       "      <td>0.683554</td>\n",
       "      <td>1.729485</td>\n",
       "      <td>0.634513</td>\n",
       "      <td>0.902774</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>0.682784</td>\n",
       "      <td>1.809469</td>\n",
       "      <td>0.608552</td>\n",
       "      <td>0.889539</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>51</td>\n",
       "      <td>0.686973</td>\n",
       "      <td>1.968565</td>\n",
       "      <td>0.580555</td>\n",
       "      <td>0.879359</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52</td>\n",
       "      <td>0.691561</td>\n",
       "      <td>1.741192</td>\n",
       "      <td>0.652329</td>\n",
       "      <td>0.879613</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>53</td>\n",
       "      <td>0.678949</td>\n",
       "      <td>1.708044</td>\n",
       "      <td>0.650547</td>\n",
       "      <td>0.897429</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>54</td>\n",
       "      <td>0.689924</td>\n",
       "      <td>1.715727</td>\n",
       "      <td>0.650547</td>\n",
       "      <td>0.885467</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>55</td>\n",
       "      <td>0.680266</td>\n",
       "      <td>1.756083</td>\n",
       "      <td>0.655129</td>\n",
       "      <td>0.884703</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>56</td>\n",
       "      <td>0.692027</td>\n",
       "      <td>1.826158</td>\n",
       "      <td>0.627132</td>\n",
       "      <td>0.899975</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>57</td>\n",
       "      <td>0.688069</td>\n",
       "      <td>1.877550</td>\n",
       "      <td>0.594808</td>\n",
       "      <td>0.866633</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>58</td>\n",
       "      <td>0.679617</td>\n",
       "      <td>1.765335</td>\n",
       "      <td>0.635531</td>\n",
       "      <td>0.907865</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>59</td>\n",
       "      <td>0.680604</td>\n",
       "      <td>1.880010</td>\n",
       "      <td>0.606261</td>\n",
       "      <td>0.873505</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>0.693867</td>\n",
       "      <td>1.839015</td>\n",
       "      <td>0.609570</td>\n",
       "      <td>0.874268</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61</td>\n",
       "      <td>0.677583</td>\n",
       "      <td>1.797726</td>\n",
       "      <td>0.625095</td>\n",
       "      <td>0.877832</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>62</td>\n",
       "      <td>0.673051</td>\n",
       "      <td>1.871014</td>\n",
       "      <td>0.608806</td>\n",
       "      <td>0.870450</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>63</td>\n",
       "      <td>0.672462</td>\n",
       "      <td>1.836378</td>\n",
       "      <td>0.630949</td>\n",
       "      <td>0.876813</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>64</td>\n",
       "      <td>0.665918</td>\n",
       "      <td>1.953802</td>\n",
       "      <td>0.600662</td>\n",
       "      <td>0.870450</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>65</td>\n",
       "      <td>0.664020</td>\n",
       "      <td>1.765859</td>\n",
       "      <td>0.642148</td>\n",
       "      <td>0.888776</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>66</td>\n",
       "      <td>0.658647</td>\n",
       "      <td>1.720522</td>\n",
       "      <td>0.649275</td>\n",
       "      <td>0.902774</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>67</td>\n",
       "      <td>0.649441</td>\n",
       "      <td>1.675026</td>\n",
       "      <td>0.661491</td>\n",
       "      <td>0.893612</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>68</td>\n",
       "      <td>0.647576</td>\n",
       "      <td>1.695133</td>\n",
       "      <td>0.648257</td>\n",
       "      <td>0.895139</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>69</td>\n",
       "      <td>0.638568</td>\n",
       "      <td>1.701096</td>\n",
       "      <td>0.663528</td>\n",
       "      <td>0.886740</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>70</td>\n",
       "      <td>0.633097</td>\n",
       "      <td>1.662036</td>\n",
       "      <td>0.671672</td>\n",
       "      <td>0.908374</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>71</td>\n",
       "      <td>0.627722</td>\n",
       "      <td>1.653017</td>\n",
       "      <td>0.670400</td>\n",
       "      <td>0.905065</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>72</td>\n",
       "      <td>0.622113</td>\n",
       "      <td>1.622110</td>\n",
       "      <td>0.678544</td>\n",
       "      <td>0.906337</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>73</td>\n",
       "      <td>0.618769</td>\n",
       "      <td>1.648937</td>\n",
       "      <td>0.673708</td>\n",
       "      <td>0.911682</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>74</td>\n",
       "      <td>0.617869</td>\n",
       "      <td>1.629089</td>\n",
       "      <td>0.683125</td>\n",
       "      <td>0.907356</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>0.616874</td>\n",
       "      <td>1.609164</td>\n",
       "      <td>0.683125</td>\n",
       "      <td>0.913464</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>76</td>\n",
       "      <td>0.614521</td>\n",
       "      <td>1.600654</td>\n",
       "      <td>0.689997</td>\n",
       "      <td>0.912446</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77</td>\n",
       "      <td>0.611582</td>\n",
       "      <td>1.584713</td>\n",
       "      <td>0.695851</td>\n",
       "      <td>0.917536</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>0.615230</td>\n",
       "      <td>1.581205</td>\n",
       "      <td>0.692543</td>\n",
       "      <td>0.914737</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>79</td>\n",
       "      <td>0.616682</td>\n",
       "      <td>1.576768</td>\n",
       "      <td>0.695342</td>\n",
       "      <td>0.917282</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.6953423023223877"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acc = []\n",
    "runs = 1\n",
    "for i in range(runs): acc += [finetune(size, epochs=80, arch='xresnet34',encoder_path=f'models/swav_iwang_sz{size}_epc100_encoder.pth')]\n",
    "np.mean(acc)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 200 epochs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: left;\">\n",
       "      <th>epoch</th>\n",
       "      <th>train_loss</th>\n",
       "      <th>valid_loss</th>\n",
       "      <th>accuracy</th>\n",
       "      <th>top_k_accuracy</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1.013780</td>\n",
       "      <td>1.711377</td>\n",
       "      <td>0.644948</td>\n",
       "      <td>0.919827</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.957198</td>\n",
       "      <td>1.680223</td>\n",
       "      <td>0.638076</td>\n",
       "      <td>0.906847</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.900557</td>\n",
       "      <td>1.592249</td>\n",
       "      <td>0.671418</td>\n",
       "      <td>0.918045</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.863753</td>\n",
       "      <td>1.549858</td>\n",
       "      <td>0.678035</td>\n",
       "      <td>0.929753</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.829930</td>\n",
       "      <td>1.663092</td>\n",
       "      <td>0.643930</td>\n",
       "      <td>0.888267</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.816960</td>\n",
       "      <td>1.445078</td>\n",
       "      <td>0.709850</td>\n",
       "      <td>0.938152</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.773989</td>\n",
       "      <td>1.451971</td>\n",
       "      <td>0.705014</td>\n",
       "      <td>0.926699</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.778247</td>\n",
       "      <td>1.498179</td>\n",
       "      <td>0.689488</td>\n",
       "      <td>0.930771</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.759927</td>\n",
       "      <td>1.619587</td>\n",
       "      <td>0.667345</td>\n",
       "      <td>0.918554</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.752361</td>\n",
       "      <td>1.510601</td>\n",
       "      <td>0.685925</td>\n",
       "      <td>0.916773</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>0.742371</td>\n",
       "      <td>1.553762</td>\n",
       "      <td>0.688470</td>\n",
       "      <td>0.922627</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>11</td>\n",
       "      <td>0.744939</td>\n",
       "      <td>1.544788</td>\n",
       "      <td>0.683380</td>\n",
       "      <td>0.929753</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>12</td>\n",
       "      <td>0.734665</td>\n",
       "      <td>1.569067</td>\n",
       "      <td>0.672181</td>\n",
       "      <td>0.915755</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>13</td>\n",
       "      <td>0.726086</td>\n",
       "      <td>1.505438</td>\n",
       "      <td>0.705778</td>\n",
       "      <td>0.928226</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>14</td>\n",
       "      <td>0.720445</td>\n",
       "      <td>1.598977</td>\n",
       "      <td>0.672690</td>\n",
       "      <td>0.900738</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>15</td>\n",
       "      <td>0.720428</td>\n",
       "      <td>1.615459</td>\n",
       "      <td>0.669636</td>\n",
       "      <td>0.906847</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>16</td>\n",
       "      <td>0.724076</td>\n",
       "      <td>1.727278</td>\n",
       "      <td>0.648257</td>\n",
       "      <td>0.898193</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>17</td>\n",
       "      <td>0.714133</td>\n",
       "      <td>1.533591</td>\n",
       "      <td>0.693052</td>\n",
       "      <td>0.906847</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>18</td>\n",
       "      <td>0.706400</td>\n",
       "      <td>1.606998</td>\n",
       "      <td>0.660982</td>\n",
       "      <td>0.909901</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>19</td>\n",
       "      <td>0.700100</td>\n",
       "      <td>1.684677</td>\n",
       "      <td>0.653601</td>\n",
       "      <td>0.890557</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>0.706983</td>\n",
       "      <td>1.522488</td>\n",
       "      <td>0.704759</td>\n",
       "      <td>0.926953</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>21</td>\n",
       "      <td>0.703781</td>\n",
       "      <td>1.650764</td>\n",
       "      <td>0.670654</td>\n",
       "      <td>0.903538</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>22</td>\n",
       "      <td>0.704302</td>\n",
       "      <td>1.710395</td>\n",
       "      <td>0.655383</td>\n",
       "      <td>0.883176</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>23</td>\n",
       "      <td>0.698674</td>\n",
       "      <td>1.608507</td>\n",
       "      <td>0.674472</td>\n",
       "      <td>0.907610</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>24</td>\n",
       "      <td>0.705252</td>\n",
       "      <td>1.601199</td>\n",
       "      <td>0.671418</td>\n",
       "      <td>0.913209</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>0.699616</td>\n",
       "      <td>1.598010</td>\n",
       "      <td>0.678799</td>\n",
       "      <td>0.906847</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>26</td>\n",
       "      <td>0.693794</td>\n",
       "      <td>1.638368</td>\n",
       "      <td>0.668363</td>\n",
       "      <td>0.903538</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>27</td>\n",
       "      <td>0.706746</td>\n",
       "      <td>1.748625</td>\n",
       "      <td>0.632731</td>\n",
       "      <td>0.894884</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>28</td>\n",
       "      <td>0.705891</td>\n",
       "      <td>1.768553</td>\n",
       "      <td>0.632222</td>\n",
       "      <td>0.890812</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>29</td>\n",
       "      <td>0.694109</td>\n",
       "      <td>1.679884</td>\n",
       "      <td>0.651311</td>\n",
       "      <td>0.906592</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>0.694859</td>\n",
       "      <td>1.599584</td>\n",
       "      <td>0.676508</td>\n",
       "      <td>0.918809</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>31</td>\n",
       "      <td>0.692749</td>\n",
       "      <td>1.595019</td>\n",
       "      <td>0.670145</td>\n",
       "      <td>0.912191</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>32</td>\n",
       "      <td>0.693762</td>\n",
       "      <td>1.605884</td>\n",
       "      <td>0.683635</td>\n",
       "      <td>0.911937</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>33</td>\n",
       "      <td>0.690526</td>\n",
       "      <td>1.722587</td>\n",
       "      <td>0.651056</td>\n",
       "      <td>0.891830</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>34</td>\n",
       "      <td>0.687206</td>\n",
       "      <td>1.685755</td>\n",
       "      <td>0.660219</td>\n",
       "      <td>0.904047</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>35</td>\n",
       "      <td>0.691317</td>\n",
       "      <td>1.591248</td>\n",
       "      <td>0.671163</td>\n",
       "      <td>0.914482</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>36</td>\n",
       "      <td>0.685668</td>\n",
       "      <td>1.708884</td>\n",
       "      <td>0.656656</td>\n",
       "      <td>0.899975</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>37</td>\n",
       "      <td>0.689023</td>\n",
       "      <td>1.652835</td>\n",
       "      <td>0.677526</td>\n",
       "      <td>0.904556</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>38</td>\n",
       "      <td>0.690127</td>\n",
       "      <td>1.767820</td>\n",
       "      <td>0.630186</td>\n",
       "      <td>0.885976</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>39</td>\n",
       "      <td>0.702601</td>\n",
       "      <td>1.752626</td>\n",
       "      <td>0.624077</td>\n",
       "      <td>0.891575</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>0.691899</td>\n",
       "      <td>1.798810</td>\n",
       "      <td>0.631967</td>\n",
       "      <td>0.879868</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>41</td>\n",
       "      <td>0.686281</td>\n",
       "      <td>1.737651</td>\n",
       "      <td>0.643421</td>\n",
       "      <td>0.890048</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>42</td>\n",
       "      <td>0.685320</td>\n",
       "      <td>1.683734</td>\n",
       "      <td>0.654365</td>\n",
       "      <td>0.902520</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>43</td>\n",
       "      <td>0.695186</td>\n",
       "      <td>1.749459</td>\n",
       "      <td>0.649529</td>\n",
       "      <td>0.891575</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>44</td>\n",
       "      <td>0.698177</td>\n",
       "      <td>1.755764</td>\n",
       "      <td>0.638076</td>\n",
       "      <td>0.896411</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>45</td>\n",
       "      <td>0.686050</td>\n",
       "      <td>1.815737</td>\n",
       "      <td>0.629677</td>\n",
       "      <td>0.895902</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>46</td>\n",
       "      <td>0.693035</td>\n",
       "      <td>1.762005</td>\n",
       "      <td>0.645711</td>\n",
       "      <td>0.904556</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>47</td>\n",
       "      <td>0.687789</td>\n",
       "      <td>1.794043</td>\n",
       "      <td>0.616187</td>\n",
       "      <td>0.867651</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>48</td>\n",
       "      <td>0.689929</td>\n",
       "      <td>1.767023</td>\n",
       "      <td>0.629168</td>\n",
       "      <td>0.901502</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>49</td>\n",
       "      <td>0.688214</td>\n",
       "      <td>1.874626</td>\n",
       "      <td>0.617714</td>\n",
       "      <td>0.882413</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>0.684901</td>\n",
       "      <td>1.723247</td>\n",
       "      <td>0.645966</td>\n",
       "      <td>0.887249</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>51</td>\n",
       "      <td>0.674565</td>\n",
       "      <td>1.754197</td>\n",
       "      <td>0.646220</td>\n",
       "      <td>0.879359</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>52</td>\n",
       "      <td>0.686410</td>\n",
       "      <td>1.765136</td>\n",
       "      <td>0.629677</td>\n",
       "      <td>0.868414</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>53</td>\n",
       "      <td>0.686671</td>\n",
       "      <td>1.864584</td>\n",
       "      <td>0.613133</td>\n",
       "      <td>0.881649</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>54</td>\n",
       "      <td>0.684796</td>\n",
       "      <td>1.775911</td>\n",
       "      <td>0.631967</td>\n",
       "      <td>0.893612</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>55</td>\n",
       "      <td>0.682578</td>\n",
       "      <td>1.878444</td>\n",
       "      <td>0.622041</td>\n",
       "      <td>0.881395</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>56</td>\n",
       "      <td>0.690643</td>\n",
       "      <td>1.628620</td>\n",
       "      <td>0.667345</td>\n",
       "      <td>0.898193</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>57</td>\n",
       "      <td>0.681156</td>\n",
       "      <td>1.755227</td>\n",
       "      <td>0.644693</td>\n",
       "      <td>0.895902</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>58</td>\n",
       "      <td>0.690099</td>\n",
       "      <td>2.003966</td>\n",
       "      <td>0.589718</td>\n",
       "      <td>0.849326</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>59</td>\n",
       "      <td>0.689579</td>\n",
       "      <td>1.863669</td>\n",
       "      <td>0.607534</td>\n",
       "      <td>0.883940</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>0.675732</td>\n",
       "      <td>1.726355</td>\n",
       "      <td>0.649529</td>\n",
       "      <td>0.883176</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>61</td>\n",
       "      <td>0.682139</td>\n",
       "      <td>1.871088</td>\n",
       "      <td>0.620514</td>\n",
       "      <td>0.871469</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>62</td>\n",
       "      <td>0.679649</td>\n",
       "      <td>1.782858</td>\n",
       "      <td>0.625604</td>\n",
       "      <td>0.892084</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>63</td>\n",
       "      <td>0.695308</td>\n",
       "      <td>1.857691</td>\n",
       "      <td>0.616951</td>\n",
       "      <td>0.863578</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>64</td>\n",
       "      <td>0.682695</td>\n",
       "      <td>1.769210</td>\n",
       "      <td>0.647238</td>\n",
       "      <td>0.877577</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>65</td>\n",
       "      <td>0.688873</td>\n",
       "      <td>1.714079</td>\n",
       "      <td>0.648002</td>\n",
       "      <td>0.895139</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>66</td>\n",
       "      <td>0.682669</td>\n",
       "      <td>1.772077</td>\n",
       "      <td>0.635022</td>\n",
       "      <td>0.876304</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>67</td>\n",
       "      <td>0.695014</td>\n",
       "      <td>1.834658</td>\n",
       "      <td>0.611861</td>\n",
       "      <td>0.876050</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>68</td>\n",
       "      <td>0.685794</td>\n",
       "      <td>1.766010</td>\n",
       "      <td>0.641130</td>\n",
       "      <td>0.883940</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>69</td>\n",
       "      <td>0.683068</td>\n",
       "      <td>1.726910</td>\n",
       "      <td>0.643421</td>\n",
       "      <td>0.887249</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>70</td>\n",
       "      <td>0.679768</td>\n",
       "      <td>1.667348</td>\n",
       "      <td>0.656147</td>\n",
       "      <td>0.904047</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>71</td>\n",
       "      <td>0.675423</td>\n",
       "      <td>1.823259</td>\n",
       "      <td>0.616187</td>\n",
       "      <td>0.871469</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>72</td>\n",
       "      <td>0.690131</td>\n",
       "      <td>1.861368</td>\n",
       "      <td>0.610333</td>\n",
       "      <td>0.856452</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>73</td>\n",
       "      <td>0.684696</td>\n",
       "      <td>1.698371</td>\n",
       "      <td>0.664291</td>\n",
       "      <td>0.881649</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>74</td>\n",
       "      <td>0.686861</td>\n",
       "      <td>1.871728</td>\n",
       "      <td>0.592263</td>\n",
       "      <td>0.858234</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>0.677939</td>\n",
       "      <td>1.854122</td>\n",
       "      <td>0.606516</td>\n",
       "      <td>0.869941</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>76</td>\n",
       "      <td>0.687219</td>\n",
       "      <td>1.849885</td>\n",
       "      <td>0.614151</td>\n",
       "      <td>0.865106</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>77</td>\n",
       "      <td>0.684042</td>\n",
       "      <td>1.873260</td>\n",
       "      <td>0.610588</td>\n",
       "      <td>0.866633</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>78</td>\n",
       "      <td>0.684169</td>\n",
       "      <td>1.973410</td>\n",
       "      <td>0.587427</td>\n",
       "      <td>0.850598</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>79</td>\n",
       "      <td>0.692867</td>\n",
       "      <td>1.870039</td>\n",
       "      <td>0.602443</td>\n",
       "      <td>0.877068</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>80</td>\n",
       "      <td>0.679440</td>\n",
       "      <td>1.914167</td>\n",
       "      <td>0.592008</td>\n",
       "      <td>0.873505</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>81</td>\n",
       "      <td>0.683183</td>\n",
       "      <td>2.025739</td>\n",
       "      <td>0.555103</td>\n",
       "      <td>0.844744</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>82</td>\n",
       "      <td>0.676077</td>\n",
       "      <td>2.090912</td>\n",
       "      <td>0.560702</td>\n",
       "      <td>0.841435</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>83</td>\n",
       "      <td>0.680355</td>\n",
       "      <td>1.867959</td>\n",
       "      <td>0.619751</td>\n",
       "      <td>0.880377</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>84</td>\n",
       "      <td>0.678679</td>\n",
       "      <td>1.746353</td>\n",
       "      <td>0.650802</td>\n",
       "      <td>0.895648</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>85</td>\n",
       "      <td>0.678588</td>\n",
       "      <td>1.776539</td>\n",
       "      <td>0.632476</td>\n",
       "      <td>0.878086</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>86</td>\n",
       "      <td>0.687700</td>\n",
       "      <td>2.011112</td>\n",
       "      <td>0.589463</td>\n",
       "      <td>0.844235</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>87</td>\n",
       "      <td>0.678691</td>\n",
       "      <td>1.908354</td>\n",
       "      <td>0.588190</td>\n",
       "      <td>0.848562</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>88</td>\n",
       "      <td>0.672089</td>\n",
       "      <td>1.817081</td>\n",
       "      <td>0.602698</td>\n",
       "      <td>0.912446</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>89</td>\n",
       "      <td>0.677728</td>\n",
       "      <td>1.756991</td>\n",
       "      <td>0.641130</td>\n",
       "      <td>0.886231</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90</td>\n",
       "      <td>0.684030</td>\n",
       "      <td>1.864832</td>\n",
       "      <td>0.600153</td>\n",
       "      <td>0.872232</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>91</td>\n",
       "      <td>0.678269</td>\n",
       "      <td>1.984000</td>\n",
       "      <td>0.583864</td>\n",
       "      <td>0.866633</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>92</td>\n",
       "      <td>0.665357</td>\n",
       "      <td>1.873047</td>\n",
       "      <td>0.599898</td>\n",
       "      <td>0.854416</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>93</td>\n",
       "      <td>0.688492</td>\n",
       "      <td>1.828692</td>\n",
       "      <td>0.615678</td>\n",
       "      <td>0.874014</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>94</td>\n",
       "      <td>0.689954</td>\n",
       "      <td>1.827164</td>\n",
       "      <td>0.618733</td>\n",
       "      <td>0.856197</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>95</td>\n",
       "      <td>0.688379</td>\n",
       "      <td>1.856480</td>\n",
       "      <td>0.612115</td>\n",
       "      <td>0.860779</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>96</td>\n",
       "      <td>0.674610</td>\n",
       "      <td>1.849268</td>\n",
       "      <td>0.611351</td>\n",
       "      <td>0.868160</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>97</td>\n",
       "      <td>0.679317</td>\n",
       "      <td>1.809041</td>\n",
       "      <td>0.630949</td>\n",
       "      <td>0.884703</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>98</td>\n",
       "      <td>0.679455</td>\n",
       "      <td>1.825694</td>\n",
       "      <td>0.624586</td>\n",
       "      <td>0.892339</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>99</td>\n",
       "      <td>0.678143</td>\n",
       "      <td>1.890903</td>\n",
       "      <td>0.602952</td>\n",
       "      <td>0.869178</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>100</td>\n",
       "      <td>0.682432</td>\n",
       "      <td>1.908202</td>\n",
       "      <td>0.585136</td>\n",
       "      <td>0.856961</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>101</td>\n",
       "      <td>0.674763</td>\n",
       "      <td>1.903272</td>\n",
       "      <td>0.606261</td>\n",
       "      <td>0.852125</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>102</td>\n",
       "      <td>0.672050</td>\n",
       "      <td>1.886846</td>\n",
       "      <td>0.599644</td>\n",
       "      <td>0.870960</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>103</td>\n",
       "      <td>0.675401</td>\n",
       "      <td>1.897405</td>\n",
       "      <td>0.600916</td>\n",
       "      <td>0.863069</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>104</td>\n",
       "      <td>0.679270</td>\n",
       "      <td>1.829319</td>\n",
       "      <td>0.613388</td>\n",
       "      <td>0.870960</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>105</td>\n",
       "      <td>0.679043</td>\n",
       "      <td>1.862532</td>\n",
       "      <td>0.623568</td>\n",
       "      <td>0.878595</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>106</td>\n",
       "      <td>0.681118</td>\n",
       "      <td>1.911818</td>\n",
       "      <td>0.597862</td>\n",
       "      <td>0.876559</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>107</td>\n",
       "      <td>0.676471</td>\n",
       "      <td>1.833288</td>\n",
       "      <td>0.616951</td>\n",
       "      <td>0.878086</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>108</td>\n",
       "      <td>0.683125</td>\n",
       "      <td>1.918672</td>\n",
       "      <td>0.597608</td>\n",
       "      <td>0.857979</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>109</td>\n",
       "      <td>0.678780</td>\n",
       "      <td>1.800901</td>\n",
       "      <td>0.621532</td>\n",
       "      <td>0.881140</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>110</td>\n",
       "      <td>0.677672</td>\n",
       "      <td>1.901587</td>\n",
       "      <td>0.611606</td>\n",
       "      <td>0.878341</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>111</td>\n",
       "      <td>0.676436</td>\n",
       "      <td>1.861991</td>\n",
       "      <td>0.611351</td>\n",
       "      <td>0.872487</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>112</td>\n",
       "      <td>0.680930</td>\n",
       "      <td>2.170366</td>\n",
       "      <td>0.538305</td>\n",
       "      <td>0.840926</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>113</td>\n",
       "      <td>0.680661</td>\n",
       "      <td>1.910201</td>\n",
       "      <td>0.607788</td>\n",
       "      <td>0.878086</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>114</td>\n",
       "      <td>0.683962</td>\n",
       "      <td>1.937804</td>\n",
       "      <td>0.595317</td>\n",
       "      <td>0.854161</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>115</td>\n",
       "      <td>0.691988</td>\n",
       "      <td>1.845514</td>\n",
       "      <td>0.602698</td>\n",
       "      <td>0.891575</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>116</td>\n",
       "      <td>0.671276</td>\n",
       "      <td>1.858842</td>\n",
       "      <td>0.605752</td>\n",
       "      <td>0.875032</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>117</td>\n",
       "      <td>0.670315</td>\n",
       "      <td>1.950787</td>\n",
       "      <td>0.582591</td>\n",
       "      <td>0.840163</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>118</td>\n",
       "      <td>0.678910</td>\n",
       "      <td>2.031325</td>\n",
       "      <td>0.573683</td>\n",
       "      <td>0.864342</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>119</td>\n",
       "      <td>0.677343</td>\n",
       "      <td>1.935660</td>\n",
       "      <td>0.593790</td>\n",
       "      <td>0.864597</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>120</td>\n",
       "      <td>0.684564</td>\n",
       "      <td>1.991830</td>\n",
       "      <td>0.586918</td>\n",
       "      <td>0.825401</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>121</td>\n",
       "      <td>0.681706</td>\n",
       "      <td>1.941857</td>\n",
       "      <td>0.591245</td>\n",
       "      <td>0.855179</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>122</td>\n",
       "      <td>0.686967</td>\n",
       "      <td>1.852164</td>\n",
       "      <td>0.595062</td>\n",
       "      <td>0.867905</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>123</td>\n",
       "      <td>0.679594</td>\n",
       "      <td>1.973236</td>\n",
       "      <td>0.587681</td>\n",
       "      <td>0.850598</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>124</td>\n",
       "      <td>0.673277</td>\n",
       "      <td>1.783498</td>\n",
       "      <td>0.634513</td>\n",
       "      <td>0.882413</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>125</td>\n",
       "      <td>0.686214</td>\n",
       "      <td>1.843744</td>\n",
       "      <td>0.608043</td>\n",
       "      <td>0.865360</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>126</td>\n",
       "      <td>0.677882</td>\n",
       "      <td>1.974578</td>\n",
       "      <td>0.581318</td>\n",
       "      <td>0.855434</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>127</td>\n",
       "      <td>0.675345</td>\n",
       "      <td>1.896685</td>\n",
       "      <td>0.604225</td>\n",
       "      <td>0.862815</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>128</td>\n",
       "      <td>0.673369</td>\n",
       "      <td>1.912372</td>\n",
       "      <td>0.599389</td>\n",
       "      <td>0.875795</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>129</td>\n",
       "      <td>0.677751</td>\n",
       "      <td>2.056819</td>\n",
       "      <td>0.574192</td>\n",
       "      <td>0.823365</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>130</td>\n",
       "      <td>0.674222</td>\n",
       "      <td>1.968054</td>\n",
       "      <td>0.598371</td>\n",
       "      <td>0.858488</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>131</td>\n",
       "      <td>0.675789</td>\n",
       "      <td>1.947275</td>\n",
       "      <td>0.587681</td>\n",
       "      <td>0.870960</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>132</td>\n",
       "      <td>0.673033</td>\n",
       "      <td>1.923589</td>\n",
       "      <td>0.592517</td>\n",
       "      <td>0.865615</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>133</td>\n",
       "      <td>0.676452</td>\n",
       "      <td>1.874959</td>\n",
       "      <td>0.601934</td>\n",
       "      <td>0.879868</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>134</td>\n",
       "      <td>0.676785</td>\n",
       "      <td>1.830008</td>\n",
       "      <td>0.613897</td>\n",
       "      <td>0.871723</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>135</td>\n",
       "      <td>0.673829</td>\n",
       "      <td>2.031365</td>\n",
       "      <td>0.569611</td>\n",
       "      <td>0.840672</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>136</td>\n",
       "      <td>0.679018</td>\n",
       "      <td>2.084355</td>\n",
       "      <td>0.568338</td>\n",
       "      <td>0.826419</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>137</td>\n",
       "      <td>0.676917</td>\n",
       "      <td>1.906532</td>\n",
       "      <td>0.592008</td>\n",
       "      <td>0.859506</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>138</td>\n",
       "      <td>0.675266</td>\n",
       "      <td>1.955284</td>\n",
       "      <td>0.595826</td>\n",
       "      <td>0.852125</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>139</td>\n",
       "      <td>0.679116</td>\n",
       "      <td>2.008483</td>\n",
       "      <td>0.582082</td>\n",
       "      <td>0.832782</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>140</td>\n",
       "      <td>0.666446</td>\n",
       "      <td>1.895842</td>\n",
       "      <td>0.613642</td>\n",
       "      <td>0.880631</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>141</td>\n",
       "      <td>0.664661</td>\n",
       "      <td>1.925745</td>\n",
       "      <td>0.602698</td>\n",
       "      <td>0.867905</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>142</td>\n",
       "      <td>0.680145</td>\n",
       "      <td>1.972923</td>\n",
       "      <td>0.574955</td>\n",
       "      <td>0.841181</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>143</td>\n",
       "      <td>0.675662</td>\n",
       "      <td>2.129140</td>\n",
       "      <td>0.542123</td>\n",
       "      <td>0.846526</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>144</td>\n",
       "      <td>0.664610</td>\n",
       "      <td>1.979348</td>\n",
       "      <td>0.583355</td>\n",
       "      <td>0.846526</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>145</td>\n",
       "      <td>0.683881</td>\n",
       "      <td>1.938779</td>\n",
       "      <td>0.583355</td>\n",
       "      <td>0.859506</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>146</td>\n",
       "      <td>0.667828</td>\n",
       "      <td>2.033193</td>\n",
       "      <td>0.559430</td>\n",
       "      <td>0.843217</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>147</td>\n",
       "      <td>0.679226</td>\n",
       "      <td>2.028548</td>\n",
       "      <td>0.556121</td>\n",
       "      <td>0.840417</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>148</td>\n",
       "      <td>0.672032</td>\n",
       "      <td>1.915990</td>\n",
       "      <td>0.591245</td>\n",
       "      <td>0.844490</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>149</td>\n",
       "      <td>0.673209</td>\n",
       "      <td>1.837758</td>\n",
       "      <td>0.618733</td>\n",
       "      <td>0.868669</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>150</td>\n",
       "      <td>0.667547</td>\n",
       "      <td>1.913213</td>\n",
       "      <td>0.604225</td>\n",
       "      <td>0.871723</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>151</td>\n",
       "      <td>0.680440</td>\n",
       "      <td>1.873697</td>\n",
       "      <td>0.599644</td>\n",
       "      <td>0.878341</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>152</td>\n",
       "      <td>0.664546</td>\n",
       "      <td>1.933766</td>\n",
       "      <td>0.590227</td>\n",
       "      <td>0.875032</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>153</td>\n",
       "      <td>0.674186</td>\n",
       "      <td>1.929805</td>\n",
       "      <td>0.587936</td>\n",
       "      <td>0.843217</td>\n",
       "      <td>00:29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>154</td>\n",
       "      <td>0.666512</td>\n",
       "      <td>2.083012</td>\n",
       "      <td>0.559939</td>\n",
       "      <td>0.848562</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>155</td>\n",
       "      <td>0.667212</td>\n",
       "      <td>1.929512</td>\n",
       "      <td>0.593535</td>\n",
       "      <td>0.868669</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>156</td>\n",
       "      <td>0.668827</td>\n",
       "      <td>1.838702</td>\n",
       "      <td>0.614660</td>\n",
       "      <td>0.875286</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>157</td>\n",
       "      <td>0.670318</td>\n",
       "      <td>1.928795</td>\n",
       "      <td>0.599898</td>\n",
       "      <td>0.862815</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>158</td>\n",
       "      <td>0.658801</td>\n",
       "      <td>1.950971</td>\n",
       "      <td>0.586918</td>\n",
       "      <td>0.843726</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>159</td>\n",
       "      <td>0.662812</td>\n",
       "      <td>1.927983</td>\n",
       "      <td>0.578010</td>\n",
       "      <td>0.865615</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>160</td>\n",
       "      <td>0.658898</td>\n",
       "      <td>2.106874</td>\n",
       "      <td>0.562993</td>\n",
       "      <td>0.852125</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>161</td>\n",
       "      <td>0.659534</td>\n",
       "      <td>1.748844</td>\n",
       "      <td>0.644948</td>\n",
       "      <td>0.902011</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>162</td>\n",
       "      <td>0.645723</td>\n",
       "      <td>1.905426</td>\n",
       "      <td>0.602952</td>\n",
       "      <td>0.851616</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>163</td>\n",
       "      <td>0.663175</td>\n",
       "      <td>2.022668</td>\n",
       "      <td>0.567065</td>\n",
       "      <td>0.850598</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>164</td>\n",
       "      <td>0.654167</td>\n",
       "      <td>1.911211</td>\n",
       "      <td>0.599389</td>\n",
       "      <td>0.862306</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>165</td>\n",
       "      <td>0.645642</td>\n",
       "      <td>1.851497</td>\n",
       "      <td>0.620005</td>\n",
       "      <td>0.867905</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>166</td>\n",
       "      <td>0.652041</td>\n",
       "      <td>1.926049</td>\n",
       "      <td>0.583355</td>\n",
       "      <td>0.857216</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>167</td>\n",
       "      <td>0.644400</td>\n",
       "      <td>1.909670</td>\n",
       "      <td>0.605243</td>\n",
       "      <td>0.854670</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>168</td>\n",
       "      <td>0.635534</td>\n",
       "      <td>1.870974</td>\n",
       "      <td>0.616442</td>\n",
       "      <td>0.867142</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>169</td>\n",
       "      <td>0.638340</td>\n",
       "      <td>1.836811</td>\n",
       "      <td>0.623823</td>\n",
       "      <td>0.875795</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>170</td>\n",
       "      <td>0.635854</td>\n",
       "      <td>1.954821</td>\n",
       "      <td>0.598117</td>\n",
       "      <td>0.857725</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>171</td>\n",
       "      <td>0.629515</td>\n",
       "      <td>1.925138</td>\n",
       "      <td>0.608552</td>\n",
       "      <td>0.868923</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>172</td>\n",
       "      <td>0.633355</td>\n",
       "      <td>1.934244</td>\n",
       "      <td>0.604479</td>\n",
       "      <td>0.845253</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>173</td>\n",
       "      <td>0.632161</td>\n",
       "      <td>1.926269</td>\n",
       "      <td>0.602443</td>\n",
       "      <td>0.868414</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>174</td>\n",
       "      <td>0.628919</td>\n",
       "      <td>1.849430</td>\n",
       "      <td>0.634258</td>\n",
       "      <td>0.870960</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>175</td>\n",
       "      <td>0.619906</td>\n",
       "      <td>1.800258</td>\n",
       "      <td>0.643930</td>\n",
       "      <td>0.883940</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>176</td>\n",
       "      <td>0.621799</td>\n",
       "      <td>1.965994</td>\n",
       "      <td>0.600916</td>\n",
       "      <td>0.859252</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>177</td>\n",
       "      <td>0.621679</td>\n",
       "      <td>1.896823</td>\n",
       "      <td>0.615424</td>\n",
       "      <td>0.878850</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>178</td>\n",
       "      <td>0.618764</td>\n",
       "      <td>1.854991</td>\n",
       "      <td>0.625350</td>\n",
       "      <td>0.870705</td>\n",
       "      <td>00:29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>179</td>\n",
       "      <td>0.620322</td>\n",
       "      <td>1.787071</td>\n",
       "      <td>0.642148</td>\n",
       "      <td>0.881649</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>180</td>\n",
       "      <td>0.615774</td>\n",
       "      <td>1.755334</td>\n",
       "      <td>0.651820</td>\n",
       "      <td>0.887249</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>181</td>\n",
       "      <td>0.616013</td>\n",
       "      <td>1.827912</td>\n",
       "      <td>0.633749</td>\n",
       "      <td>0.869687</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>182</td>\n",
       "      <td>0.613129</td>\n",
       "      <td>1.838742</td>\n",
       "      <td>0.627641</td>\n",
       "      <td>0.876559</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>183</td>\n",
       "      <td>0.613006</td>\n",
       "      <td>1.800004</td>\n",
       "      <td>0.639348</td>\n",
       "      <td>0.878086</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>184</td>\n",
       "      <td>0.610388</td>\n",
       "      <td>1.770645</td>\n",
       "      <td>0.639094</td>\n",
       "      <td>0.885722</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>185</td>\n",
       "      <td>0.611526</td>\n",
       "      <td>1.776644</td>\n",
       "      <td>0.644948</td>\n",
       "      <td>0.882413</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>186</td>\n",
       "      <td>0.610270</td>\n",
       "      <td>1.794658</td>\n",
       "      <td>0.643421</td>\n",
       "      <td>0.892594</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>187</td>\n",
       "      <td>0.610258</td>\n",
       "      <td>1.761874</td>\n",
       "      <td>0.654620</td>\n",
       "      <td>0.894121</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>188</td>\n",
       "      <td>0.610160</td>\n",
       "      <td>1.755187</td>\n",
       "      <td>0.651056</td>\n",
       "      <td>0.893357</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>189</td>\n",
       "      <td>0.608980</td>\n",
       "      <td>1.759603</td>\n",
       "      <td>0.654874</td>\n",
       "      <td>0.889285</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>190</td>\n",
       "      <td>0.608568</td>\n",
       "      <td>1.754890</td>\n",
       "      <td>0.654874</td>\n",
       "      <td>0.890557</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>191</td>\n",
       "      <td>0.606893</td>\n",
       "      <td>1.732826</td>\n",
       "      <td>0.657674</td>\n",
       "      <td>0.889794</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>192</td>\n",
       "      <td>0.605577</td>\n",
       "      <td>1.713015</td>\n",
       "      <td>0.662510</td>\n",
       "      <td>0.900229</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>193</td>\n",
       "      <td>0.612271</td>\n",
       "      <td>1.712052</td>\n",
       "      <td>0.662255</td>\n",
       "      <td>0.893612</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>194</td>\n",
       "      <td>0.609803</td>\n",
       "      <td>1.738454</td>\n",
       "      <td>0.656147</td>\n",
       "      <td>0.894884</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>195</td>\n",
       "      <td>0.606605</td>\n",
       "      <td>1.739440</td>\n",
       "      <td>0.656910</td>\n",
       "      <td>0.891575</td>\n",
       "      <td>00:26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>196</td>\n",
       "      <td>0.608210</td>\n",
       "      <td>1.710757</td>\n",
       "      <td>0.662255</td>\n",
       "      <td>0.896666</td>\n",
       "      <td>00:28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>197</td>\n",
       "      <td>0.608975</td>\n",
       "      <td>1.723587</td>\n",
       "      <td>0.661237</td>\n",
       "      <td>0.896920</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>198</td>\n",
       "      <td>0.606991</td>\n",
       "      <td>1.724621</td>\n",
       "      <td>0.661491</td>\n",
       "      <td>0.895648</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>199</td>\n",
       "      <td>0.608533</td>\n",
       "      <td>1.726687</td>\n",
       "      <td>0.660473</td>\n",
       "      <td>0.893866</td>\n",
       "      <td>00:27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "0.6604734063148499"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "acc = []\n",
    "runs = 1\n",
    "for i in range(runs): acc += [finetune(size, epochs=200, arch='xresnet34', encoder_path=f'models/swav_iwang_sz{size}_epc100_encoder.pth')]\n",
    "np.mean(acc)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}