02_train_cross_entropy_loss-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.vision.all import *\n",
    "from moving_mnist.models.conv_rnn import *\n",
    "from moving_mnist.data import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Quadro RTX 8000\n"
     ]
    }
   ],
   "source": [
    "if torch.cuda.is_available():\n",
    "    torch.cuda.set_device(0)\n",
    "    print(torch.cuda.get_device_name())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Train Example:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We wil predict:\n",
    "- `n_in`: 5 images\n",
    "- `n_out`: 5 images  \n",
    "- `n_obj`: up to 3 objects"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Path('/home/tcapelle/SteadySun/moving_mnist')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Path.cwd()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_PATH = Path.cwd()/'data'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds = MovingMNIST(DATA_PATH, n_in=5, n_out=5, n_obj=[1,2], th=0.7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_tl = TfmdLists(range(7500), ImageTupleTransform(ds, cl_type=TensorMask))\n",
    "valid_tl = TfmdLists(range(100), ImageTupleTransform(ds, cl_type=TensorMask))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "dls = DataLoaders.from_dsets(train_tl, valid_tl, bs=64,\n",
    "                             after_batch=[Normalize.from_stats(imagenet_stats[0][0], \n",
    "                                                               imagenet_stats[1][0])]).cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "loss_func = StackLoss(CrossEntropyLossFlat(axis=1))\n",
    "metrics = [MultiImageDice(axis=1)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Left: Input, Right: Target"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x777.6 with 18 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dls.show_batch()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Baseline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`StackUnstack` takes cares of stacking the list of images into a fat tensor, and unstacking them at the end, we will need to modify our loss function to take a list of tensors as input and target."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = StackUnstack(SimpleModel(n_in=1, n_out=2, act=partial(nn.LeakyReLU, 0.2, inplace=True))).cuda()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As the `ImageSeq` is a `tuple` of images, we will need to stack them to compute loss."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = Learner(dls, model, loss_func=loss_func, metrics=metrics, cbs=[]).to_fp16()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I have a weird bug that if I use `nn.LeakyReLU` after doing `learn.lr_find()` the model does not train (the loss get stucked)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# learn.lr_find()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>multi_image_dice</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.344523</td>\n",
       "      <td>00:13</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-14-9ee627eeb2b5>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit_one_cycle\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1e-3\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/Apps/fastcore/fastcore/utils.py\u001b[0m in \u001b[0;36m_f\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    470\u001b[0m         \u001b[0minit_args\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlog\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    471\u001b[0m         \u001b[0msetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minst\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'init_args'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minit_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 472\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0minst\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mto_return\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    473\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0m_f\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    474\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/callback/schedule.py\u001b[0m in \u001b[0;36mfit_one_cycle\u001b[0;34m(self, n_epoch, lr_max, div, div_final, pct_start, wd, moms, cbs, reset_opt)\u001b[0m\n\u001b[1;32m    111\u001b[0m     scheds = {'lr': combined_cos(pct_start, lr_max/div, lr_max, lr_max/div_final),\n\u001b[1;32m    112\u001b[0m               'mom': combined_cos(pct_start, *(self.moms if moms is None else moms))}\n\u001b[0;32m--> 113\u001b[0;31m     \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfit\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcbs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mParamScheduler\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mscheds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0mL\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcbs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreset_opt\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mreset_opt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwd\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mwd\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    114\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    115\u001b[0m \u001b[0;31m# Cell\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/utils.py\u001b[0m in \u001b[0;36m_f\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    470\u001b[0m         \u001b[0minit_args\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlog\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    471\u001b[0m         \u001b[0msetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minst\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'init_args'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minit_args\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 472\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0minst\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mto_return\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    473\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0m_f\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    474\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36mfit\u001b[0;34m(self, n_epoch, lr, wd, cbs, reset_opt)\u001b[0m\n\u001b[1;32m    205\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_hypers\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlr\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlr\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlr\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mlr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    206\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mn_epoch\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m0.\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 207\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_fit\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'fit'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelFitException\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_end_cleanup\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    208\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    209\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_end_cleanup\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mxb\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0myb\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpred\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    156\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_do_fit\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    195\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_epoch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    196\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mepoch\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 197\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'epoch'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelEpochException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    198\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    199\u001b[0m     \u001b[0;34m@\u001b[0m\u001b[0mlog_args\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbut\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'cbs'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    156\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_do_epoch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    189\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    190\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 191\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    192\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_epoch_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    193\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_do_epoch_train\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    181\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    182\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdls\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtrain\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 183\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mall_batches\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'train'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelTrainException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    184\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    185\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mds_idx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdl\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    156\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36mall_batches\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    159\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mall_batches\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    160\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mn_iter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 161\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mo\u001b[0m \u001b[0;32min\u001b[0m \u001b[0menumerate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdl\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mone_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    162\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    163\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_one_batch\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36mone_batch\u001b[0;34m(self, i, b)\u001b[0m\n\u001b[1;32m    177\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0miter\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    178\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_split\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 179\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_do_one_batch\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'batch'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mCancelBatchException\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    180\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    181\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_do_epoch_train\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_with_events\u001b[0;34m(self, f, event_type, ex, final)\u001b[0m\n\u001b[1;32m    153\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_with_events\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m         \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m       \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'before_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m       \u001b[0;34m;\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    156\u001b[0m         \u001b[0;32mexcept\u001b[0m \u001b[0mex\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_cancel_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m         \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m   \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf'after_{event_type}'\u001b[0m\u001b[0;34m)\u001b[0m        \u001b[0;34m;\u001b[0m\u001b[0mfinal\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_do_one_batch\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    169\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'before_backward'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    170\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_backward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 171\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'after_backward'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    172\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_step\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    173\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'after_step'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, event_name)\u001b[0m\n\u001b[1;32m    131\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mordered_cbs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mcb\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcb\u001b[0m \u001b[0;32min\u001b[0m \u001b[0msort_by_run\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcbs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    132\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 133\u001b[0;31m     \u001b[0;32mdef\u001b[0m \u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mL\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_call_one\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    134\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    135\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_call_one\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/foundation.py\u001b[0m in \u001b[0;36mmap\u001b[0;34m(self, f, *args, **kwargs)\u001b[0m\n\u001b[1;32m    381\u001b[0m              \u001b[0;32melse\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mstr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    382\u001b[0m              else f.__getitem__)\n\u001b[0;32m--> 383\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_new\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    384\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    385\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mfilter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnegate\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mFalse\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/foundation.py\u001b[0m in \u001b[0;36m_new\u001b[0;34m(self, items, *args, **kwargs)\u001b[0m\n\u001b[1;32m    331\u001b[0m     \u001b[0;34m@\u001b[0m\u001b[0mproperty\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    332\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_xtra\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 333\u001b[0;31m     \u001b[0;32mdef\u001b[0m \u001b[0m_new\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mitems\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0muse_list\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    334\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m__getitem__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mis_indexer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mL\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0muse_list\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    335\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_new\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcopy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/foundation.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(cls, x, *args, **kwargs)\u001b[0m\n\u001b[1;32m     45\u001b[0m             \u001b[0;32mreturn\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 47\u001b[0;31m         \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msuper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__call__\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     48\u001b[0m         \u001b[0mres\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_newchk\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     49\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mres\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/foundation.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, items, use_list, match, *rest)\u001b[0m\n\u001b[1;32m    322\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mitems\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mitems\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    323\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0muse_list\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0m_is_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 324\u001b[0;31m             \u001b[0mitems\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0muse_list\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0m_listify\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mitems\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    325\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mmatch\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    326\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mis_coll\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mmatch\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmatch\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/foundation.py\u001b[0m in \u001b[0;36m_listify\u001b[0;34m(o)\u001b[0m\n\u001b[1;32m    235\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mo\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    236\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0m_is_array\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 237\u001b[0;31m     \u001b[0;32mif\u001b[0m \u001b[0mis_iter\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    238\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mo\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    239\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastcore/fastcore/foundation.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    298\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0m_Arg\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mk\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpop\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    299\u001b[0m         \u001b[0mfargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0m_Arg\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32melse\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mx\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpargs\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmaxi\u001b[0m\u001b[0;34m+\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 300\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mfargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    301\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    302\u001b[0m \u001b[0;31m# Cell\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m_call_one\u001b[0;34m(self, event_name)\u001b[0m\n\u001b[1;32m    135\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_call_one\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    136\u001b[0m         \u001b[0;32massert\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 137\u001b[0;31m         \u001b[0;34m[\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcb\u001b[0m \u001b[0;32min\u001b[0m \u001b[0msort_by_run\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcbs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    138\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    139\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_bn_bias_state\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwith_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mnorm_bias_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwith_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/learner.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m    135\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_call_one\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    136\u001b[0m         \u001b[0;32massert\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 137\u001b[0;31m         \u001b[0;34m[\u001b[0m\u001b[0mcb\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mevent_name\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mcb\u001b[0m \u001b[0;32min\u001b[0m \u001b[0msort_by_run\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcbs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    138\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    139\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_bn_bias_state\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwith_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mnorm_bias_params\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mwith_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mopt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstate\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/callback/core.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, event_name)\u001b[0m\n\u001b[1;32m     42\u001b[0m                (self.run_valid and not getattr(self, 'training', False)))\n\u001b[1;32m     43\u001b[0m         \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0m_run\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mres\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgetattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnoop\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     45\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mevent_name\u001b[0m\u001b[0;34m==\u001b[0m\u001b[0;34m'after_fit'\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrun\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m \u001b[0;31m#Reset self.run to True at each end of fit\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mres\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/callback/fp16.py\u001b[0m in \u001b[0;36mafter_backward\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m     99\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlearn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss\u001b[0m \u001b[0;34m/=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_scale\u001b[0m \u001b[0;31m#To record the real loss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    100\u001b[0m         \u001b[0;31m#First, check for an overflow\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 101\u001b[0;31m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdynamic\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mgrad_overflow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel_pgs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    102\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mloss_scale\u001b[0m \u001b[0;34m/=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdiv_factor\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    103\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/callback/fp16.py\u001b[0m in \u001b[0;36mgrad_overflow\u001b[0;34m(pgs)\u001b[0m\n\u001b[1;32m     49\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mpg\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpgs\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     50\u001b[0m         \u001b[0;32mfor\u001b[0m \u001b[0mp\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mpg\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 51\u001b[0;31m             \u001b[0;32mif\u001b[0m \u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgrad\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mtest_overflow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mgrad\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     52\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0;32mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     53\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/Apps/fastai/fastai/callback/fp16.py\u001b[0m in \u001b[0;36mtest_overflow\u001b[0;34m(x)\u001b[0m\n\u001b[1;32m     42\u001b[0m \u001b[0;31m# Cell\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     43\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mtest_overflow\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 44\u001b[0;31m     \u001b[0ms\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     45\u001b[0m     \u001b[0;32mreturn\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0ms\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'inf'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0ms\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'-inf'\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0ms\u001b[0m \u001b[0;34m!=\u001b[0m \u001b[0ms\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "learn.fit_one_cycle(10, 1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p,t = learn.get_preds()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, the results is a list of 5 tensors with 100 samples each."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, torch.Size([100, 2, 64, 64]))"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(p), p[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_res(t, idx, argmax=False):\n",
    "    if argmax:\n",
    "        im_seq = ImageSeq.create([t[i][idx].argmax(0).unsqueeze(0) for i in range(5)], TensorMask)\n",
    "    else:\n",
    "        im_seq = ImageSeq.create([t[i][idx] for i in range(5)])\n",
    "    im_seq.show(figsize=(8,4));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "k = random.randint(0,99)\n",
    "show_res(t,k)\n",
    "show_res(p,k, True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = StackUnstack(SimpleModel(n_in=1, n_out=2, act=partial(nn.LeakyReLU, 0.2, inplace=True), norm=NormType.Batch))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn_zero = Learner(dls, model, loss_func=loss_func, metrics=metrics, cbs=[]).to_fp16()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>multi_image_dice</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.254579</td>\n",
       "      <td>0.141652</td>\n",
       "      <td>0.344572</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.099406</td>\n",
       "      <td>0.082443</td>\n",
       "      <td>0.261798</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.073535</td>\n",
       "      <td>0.071964</td>\n",
       "      <td>0.399151</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.065572</td>\n",
       "      <td>0.067531</td>\n",
       "      <td>0.289359</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.059867</td>\n",
       "      <td>0.061780</td>\n",
       "      <td>0.327050</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.055963</td>\n",
       "      <td>0.054625</td>\n",
       "      <td>0.444001</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.053440</td>\n",
       "      <td>0.052219</td>\n",
       "      <td>0.559101</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.050892</td>\n",
       "      <td>0.051416</td>\n",
       "      <td>0.534606</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.047976</td>\n",
       "      <td>0.049504</td>\n",
       "      <td>0.620260</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.046920</td>\n",
       "      <td>0.047337</td>\n",
       "      <td>0.611564</td>\n",
       "      <td>00:33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn_zero.fit_one_cycle(10, 1e-2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p,t = learn_zero.get_preds()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "k = random.randint(0,100)\n",
    "show_res(t,k)\n",
    "show_res(p,k, True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## A bigger Decoder"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will pass:\n",
    "- `blur`: to use blur on the upsampling path (this is done by using and a poolling layer and a replication)\n",
    "- `attn`: to include a self attention layer on the decoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_attn = StackUnstack(SimpleModel(n_in=1, n_out=2, act=partial(nn.LeakyReLU, 0.2, inplace=True),blur=True, attn=True))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have to reduce batch size as the self attention layer is heavy."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "dls = DataLoaders.from_dsets(train_tl, valid_tl, bs=32,\n",
    "                             after_batch=[Normalize.from_stats(imagenet_stats[0][0], \n",
    "                                                               imagenet_stats[1][0])]).cuda()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn2 = Learner(dls, model_attn, loss_func=loss_func, metrics=metrics, cbs=[]).to_fp16()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "SuggestedLRs(lr_min=0.00831763744354248, lr_steep=0.0691830962896347)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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": [
    "learn2.lr_find()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>multi_image_dice</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.104985</td>\n",
       "      <td>0.087807</td>\n",
       "      <td>0.085597</td>\n",
       "      <td>02:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.071374</td>\n",
       "      <td>0.068986</td>\n",
       "      <td>0.230056</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.061551</td>\n",
       "      <td>0.061798</td>\n",
       "      <td>0.463436</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.055956</td>\n",
       "      <td>0.056584</td>\n",
       "      <td>0.538007</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.051868</td>\n",
       "      <td>0.046795</td>\n",
       "      <td>0.570374</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.048773</td>\n",
       "      <td>0.043458</td>\n",
       "      <td>0.594460</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.045248</td>\n",
       "      <td>0.049654</td>\n",
       "      <td>0.600914</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.042721</td>\n",
       "      <td>0.042592</td>\n",
       "      <td>0.674483</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.040948</td>\n",
       "      <td>0.041948</td>\n",
       "      <td>0.677052</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.040500</td>\n",
       "      <td>0.039499</td>\n",
       "      <td>0.687206</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn2.fit_one_cycle(10, 1e-3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p,t = learn2.get_preds()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, the results is a list of 5 tensors with 100 samples each."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, torch.Size([100, 2, 64, 64]))"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(p), p[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_res(t, idx, argmax=False):\n",
    "    if argmax:\n",
    "        im_seq = ImageSeq.create([t[i][idx].argmax(0).unsqueeze(0) for i in range(5)], TensorMask)\n",
    "    else:\n",
    "        im_seq = ImageSeq.create([t[i][idx] for i in range(5)])\n",
    "    im_seq.show(figsize=(8,4));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "k = random.randint(0,99)\n",
    "show_res(t,k)\n",
    "show_res(p,k, True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## CoordConv"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Adds a CoordConv at the beggining of the model to propagate (x,y) position."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = StackUnstack(SimpleModel(n_in=1, n_out=2, act=partial(nn.LeakyReLU, 0.2, inplace=True),blur=True, attn=True, coord_conv=True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "learn = Learner(dls, model, loss_func=loss_func, metrics=metrics, cbs=[]).to_fp16()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "SuggestedLRs(lr_min=0.02089296132326126, lr_steep=0.17378008365631104)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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.lr_find()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>multi_image_dice</th>\n",
       "      <th>time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>0.104526</td>\n",
       "      <td>0.089156</td>\n",
       "      <td>0.097315</td>\n",
       "      <td>02:08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>0.070884</td>\n",
       "      <td>0.069416</td>\n",
       "      <td>0.299960</td>\n",
       "      <td>02:09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>0.062324</td>\n",
       "      <td>0.064788</td>\n",
       "      <td>0.477132</td>\n",
       "      <td>02:10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0.055580</td>\n",
       "      <td>0.053487</td>\n",
       "      <td>0.557042</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>0.051007</td>\n",
       "      <td>0.048482</td>\n",
       "      <td>0.562408</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>0.047630</td>\n",
       "      <td>0.050244</td>\n",
       "      <td>0.628599</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>0.044153</td>\n",
       "      <td>0.042525</td>\n",
       "      <td>0.626944</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>0.041914</td>\n",
       "      <td>0.042671</td>\n",
       "      <td>0.670981</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>0.040074</td>\n",
       "      <td>0.040222</td>\n",
       "      <td>0.677560</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>0.039878</td>\n",
       "      <td>0.036492</td>\n",
       "      <td>0.700991</td>\n",
       "      <td>02:11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "learn.fit_one_cycle(10, 5e-4)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "p,t = learn.get_preds()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As you can see, the results is a list of 5 tensors with 100 samples each."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5, torch.Size([100, 2, 64, 64]))"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(p), p[0].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "def show_res(t, idx, argmax=False):\n",
    "    if argmax:\n",
    "        im_seq = ImageSeq.create([t[i][idx].argmax(0).unsqueeze(0) for i in range(5)], TensorMask)\n",
    "    else:\n",
    "        im_seq = ImageSeq.create([t[i][idx] for i in range(5)])\n",
    "    im_seq.show(figsize=(8,4));"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAcwAAABnCAYAAACNd3IfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAEJklEQVR4nO3d3XHaQBQGUJxJIWkhdaQMN5E3D29pwmWkDreQTsiDh4HBSHySdiWxe84jPwZWYj/du0h+OZ1OBwBg3Let3wAAPAOBCQABgQkAAYEJAAGBCQABgQkAge9jd/74/fdtrTcCAHvw78+v473bVZgAEBCYABAQmAAQEJgAEBCYABAQmAAQWCUw315/Ht5ef67xUhyMN0ANKkwACKwamKoeAJ5V9cC8Dsnj+0ftl+uegxKAOrRkASAgMAEgMHrx9SW0Btd3HnOtb4DyVqkwj+8fm03ivQW3sASoQ0sWAAJVAnNvVV3rJ/K3/vnglv2dLVRbwzwr2SK0RjfOuKzLKVPQFy1ZAAgUD8ylrZKh9uL1bXNfo8U2zpLPdB7rOX9DG/jTlj9oA9a12wqz1mTc8iS/xWdreTyHOFjYznnsHaSsy/7+qfoa5thAX+/0t4+7/UIc3z8mbbReN/DcyWTJJGRtmTX0+p1mP3ZbYQLAnmwamCXW36ZqvQpKP5+jdSBhrrgo3pK91zpNJvHz88Yee77v9gdAt88Z2sAthuWePlMPa0u3+1aNdrTTVb7S9mcPtGQBIFDlRz9zjwKX/uik9N/tTXoUr0Wznh6qdvZJp+Or3VSYU9Ykbbz5rsf5fA5h2tJ+pMftkpyH6TSUcpaOpe3AErsJTADYs+rnYT4yt+wfeuycHxz1YspYD7Vn7x2h9zTGPX3WVmlzP6Yde9/mgVna1Asc9GZo5x8aN+vD88295CAXta/4Zf9liuYCk2GPJod795vA5xk7+CjRSemV8Vhfst/2UpFawwSAwOYVZumjERVRWfcuFnF9O9PdrqH1vi48ptQ4mBcyJZYRWm53bx6YtbS4sXgej9bSTeDlzL2yGDnj9UlLFgACzVaYsLX0RxJTnsfF0BgmLW/uu16CWboftnj6jsBksta+BFsYWhumDAcky5QYoxbHWWDChm7XOlucZPbE+C7X80GeNUwACKgwgaeUXh5z7LFkplSVLY91cxVm8t8jmKbnFgzPx/e/LGF50VxgAkANApOHWj9q3JLqvQ6dpnKM44U1TCbx5anDuNZjbJd7dOWqXsZYYBLp5QtBG1Tu5bmutJYsAEQEJgCx64qyp+rycNCShU31NuHQhl73WxUmAAQEJgAEtGSB5vTaMqQuFSYAsZ5P2RGYABAQmADEem53C0wACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACAhMAAgITAAICEwACLycTqet3wMA7J4KEwACAhMAAgITAAICEwACAhMAAgITAAL/AWgoc6dGCH7aAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "k = random.randint(0,99)\n",
    "show_res(t,k)\n",
    "show_res(p,k, True)"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}