Skip to content

Instantly share code, notes, and snippets.

@hkristen

hkristen/102a_coco_retinanet.ipynb

Created December 11, 2018 14:12
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save hkristen/e7e49d2a278074d585bf1b2a535d16b1 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save hkristen/e7e49d2a278074d585bf1b2a535d16b1 to your computer and use it in GitHub Desktop.
Download ZIP
Trying to implement Retinanet with fastai
Display the source blob
Display the rendered blob
Raw
102a_coco_retinanet.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"%reload_ext autoreload\n",
"%autoreload 2"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"from fastai import *\n",
"from fastai.vision import *"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Data"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"PATH = Path('/mnt/backup/Workspace_Storagebox_Festmeter/harald/retinanet_fastai/coco_sample')\n",
"ANNOT_PATH = 'annotations'"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"scrolled": true
},
"outputs": [],
"source": [
"images, lbl_bbox = get_annotations(PATH / ANNOT_PATH / 'train_sample.json')"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"img2bbox = dict(zip(images, lbl_bbox))\n",
"get_y_func = lambda o:img2bbox[o.name]"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"data = (ObjectItemList.from_folder(PATH / 'train_sample')\n",
" #Where are the images?\n",
" .random_split_by_pct() \n",
" #How to split in train/valid? -> randomly with the default 20% in valid\n",
" .label_from_func(get_y_func)\n",
" #How to find the labels? -> use get_y_func\n",
" .transform(get_transforms(), tfm_y=True, padding_mode='zeros', do_crop=False, size=128,)\n",
" #Data augmentation? -> Standard transforms with tfm_y=True\n",
" .databunch(bs=16, collate_fn=bb_pad_collate)) \n",
" #Finally we convert to a DataBunch and we use bb_pad_collate"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"scrolled": false
},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x432 with 4 Axes>"
]
},
"metadata": {
"needs_background": "light"
},
"output_type": "display_data"
}
],
"source": [
"data.show_batch(rows=2, ds_type=DatasetType.Valid, figsize=(6,6))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Model"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def _get_sfs_idxs(sizes:Sizes) -> List[int]:\n",
" \"Get the indexes of the layers where the size of the activation changes.\"\n",
" feature_szs = [size[-1] for size in sizes]\n",
" sfs_idxs = list(np.where(np.array(feature_szs[:-1]) != np.array(feature_szs[1:]))[0])\n",
" if feature_szs[0] != feature_szs[1]: sfs_idxs = [0] + sfs_idxs\n",
" return sfs_idxs"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"scrolled": false
},
"outputs": [],
"source": [
"import torchvision.models as tvm\n",
"encoder = create_body(tvm.resnet50(True), -2)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"class LateralUpsampleMerge(nn.Module):\n",
" \n",
" def __init__(self, ch, ch_lat, hook):\n",
" super().__init__()\n",
" self.hook = hook\n",
" self.conv_lat = conv2d(ch_lat, ch, ks=1, bias=True)\n",
" \n",
" def forward(self, x):\n",
" return self.conv_lat(self.hook.stored) + F.interpolate(x, scale_factor=2)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"from fastai.callbacks.hooks import * "
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"class RetinaNet(nn.Module):\n",
" \"Implements RetinaNet from https://arxiv.org/abs/1708.02002 \"\n",
" def __init__(self, encoder:nn.Module, n_classes, final_bias=0., chs=256, n_anchors=9, flatten=True):\n",
" super().__init__()\n",
" self.n_classes,self.flatten = n_classes,flatten\n",
" imsize = (256,256)\n",
" sfs_szs = model_sizes(encoder, size=imsize)\n",
" hooks = hook_outputs(encoder)\n",
" sfs_idxs = _get_sfs_idxs(sfs_szs)\n",
" self.encoder = encoder\n",
" self.c5top5 = conv2d(sfs_szs[-1][1], chs, ks=1, bias=True)\n",
" self.c5top6 = conv2d(sfs_szs[-1][1], chs, stride=2, bias=True)\n",
" self.p6top7 = nn.Sequential(nn.ReLU(), conv2d(chs, chs, stride=2, bias=True))\n",
" self.merges = nn.ModuleList([LateralUpsampleMerge(chs, szs[1], hook) \n",
" for szs,hook in zip(sfs_szs[-2:-4:-1], hooks[-2:-4:-1])])\n",
" self.smoothers = nn.ModuleList([conv2d(chs, chs, 3, bias=True) for _ in range(3)])\n",
" self.classifier = self._head_subnet(n_classes, n_anchors, final_bias, chs=chs)\n",
" self.box_regressor = self._head_subnet(4, n_anchors, 0., chs=chs)\n",
" \n",
" def _head_subnet(self, n_classes, n_anchors, final_bias=0., n_conv=4, chs=256):\n",
" layers = [conv2d_relu(chs, chs, bias=True) for _ in range(n_conv)]\n",
" layers += [conv2d(chs, n_classes * n_anchors, bias=True)]\n",
" layers[-1].bias.data.zero_().add_(final_bias)\n",
" layers[-1].weight.data.fill_(0)\n",
" return nn.Sequential(*layers)\n",
" \n",
" def _apply_transpose(self, func, p_states, n_classes):\n",
" if not self.flatten: \n",
" sizes = [[p.size(0), p.size(2), p.size(3)] for p in p_states]\n",
" return [func(p).permute(0,2,3,1).view(*sz,-1,n_classes) for p,sz in zip(p_states,sizes)]\n",
" else:\n",
" return torch.cat([func(p).permute(0,2,3,1).contiguous().view(p.size(0),-1,n_classes) for p in p_states],1)\n",
" \n",
" def forward(self, x):\n",
" c5 = self.encoder(x)\n",
" p_states = [self.c5top5(c5.clone()), self.c5top6(c5)]\n",
" p_states.append(self.p6top7(p_states[-1]))\n",
" for merge in self.merges: p_states = [merge(p_states[0])] + p_states\n",
" for i, smooth in enumerate(self.smoothers[:3]):\n",
" p_states[i] = smooth(p_states[i])\n",
" return [self._apply_transpose(self.classifier, p_states, self.n_classes), \n",
" self._apply_transpose(self.box_regressor, p_states, 4),\n",
" [[p.size(2), p.size(3)] for p in p_states]]"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'conv2d_relu' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-16-e218b43fbe02>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0mencoder\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcreate_body\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtvm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mresnet50\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m-\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mRetinaNet\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mencoder\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m6\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m-\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;32m<ipython-input-15-cd4be8595d0e>\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, encoder, n_classes, final_bias, chs, n_anchors, flatten)\u001b[0m\n\u001b[1;32m 16\u001b[0m for szs,hook in zip(sfs_szs[-2:-4:-1], hooks[-2:-4:-1])])\n\u001b[1;32m 17\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msmoothers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mnn\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mModuleList\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mconv2d\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m3\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m3\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---> 18\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclassifier\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_head_subnet\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_classes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_anchors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal_bias\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 19\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbox_regressor\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_head_subnet\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_anchors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m0.\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m<ipython-input-15-cd4be8595d0e>\u001b[0m in \u001b[0;36m_head_subnet\u001b[0;34m(self, n_classes, n_anchors, final_bias, n_conv, chs)\u001b[0m\n\u001b[1;32m 20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 21\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_head_subnet\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_classes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_anchors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal_bias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_conv\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m256\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 22\u001b[0;31m \u001b[0mlayers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mconv2d_relu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_conv\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 23\u001b[0m \u001b[0mlayers\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mconv2d\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_classes\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mn_anchors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 24\u001b[0m \u001b[0mlayers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbias\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfinal_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m<ipython-input-15-cd4be8595d0e>\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 20\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 21\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0m_head_subnet\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_classes\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_anchors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfinal_bias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0.\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_conv\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m4\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m256\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 22\u001b[0;31m \u001b[0mlayers\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mconv2d_relu\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0m_\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mn_conv\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 23\u001b[0m \u001b[0mlayers\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mconv2d\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mchs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_classes\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mn_anchors\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mbias\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 24\u001b[0m \u001b[0mlayers\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbias\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfinal_bias\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mNameError\u001b[0m: name 'conv2d_relu' is not defined"
]
}
],
"source": [
"encoder = create_body(tvm.resnet50(True), -2)\n",
"model = RetinaNet(encoder, 6, -4) "
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"model.eval()\n",
"x = torch.randn(2,3,256,256)\n",
"output = model(x)"
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"ename": "NameError",
"evalue": "name 'output' is not defined",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mNameError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-33-1e83ed2a14fc>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0;34m[\u001b[0m\u001b[0my\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0my\u001b[0m \u001b[0;32min\u001b[0m \u001b[0moutput\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moutput\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;31mNameError\u001b[0m: name 'output' is not defined"
]
}
],
"source": [
"[y.size() for y in output[:2]], output[2]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Anchors"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We need to create the corresponding anchors in this order:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"torch.arange(1,17).long().view(4,4)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def create_grid(size):\n",
" \"Create a grid of a given `size`.\"\n",
" H, W = size if is_tuple(size) else (size,size)\n",
" grid = FloatTensor(H, W, 2)\n",
" linear_points = torch.linspace(-1+1/W, 1-1/W, W) if W > 1 else tensor([0.])\n",
" grid[:, :, 1] = torch.ger(torch.ones(H), linear_points).expand_as(grid[:, :, 0])\n",
" linear_points = torch.linspace(-1+1/H, 1-1/H, H) if H > 1 else tensor([0.])\n",
" grid[:, :, 0] = torch.ger(linear_points, torch.ones(W)).expand_as(grid[:, :, 1])\n",
" return grid.view(-1,2)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Convention (-1.,-1.) to (1.,1.), first is y, second is x (like for the bboxes). -1 is left/top, 1 is right/bottom."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def show_anchors(ancs, size):\n",
" _,ax = plt.subplots(1,1, figsize=(5,5))\n",
" ax.set_xticks(np.linspace(-1,1, size[1]+1))\n",
" ax.set_yticks(np.linspace(-1,1, size[0]+1))\n",
" ax.grid()\n",
" ax.scatter(ancs[:,1], ancs[:,0]) #y is first\n",
" ax.set_yticklabels([])\n",
" ax.set_xticklabels([])\n",
" ax.set_xlim(-1,1)\n",
" ax.set_ylim(1,-1) #-1 is top, 1 is bottom\n",
" for i, (x, y) in enumerate(zip(ancs[:, 1], ancs[:, 0])): ax.annotate(i, xy = (x,y))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"size = (4,4)\n",
"show_anchors(create_grid(size), size)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def create_anchors(sizes, ratios, scales, flatten=True):\n",
" \"Create anchor of `sizes`, `ratios` and `scales`.\"\n",
" aspects = [[[s*math.sqrt(r), s*math.sqrt(1/r)] for s in scales] for r in ratios]\n",
" aspects = torch.tensor(aspects).view(-1,2)\n",
" anchors = []\n",
" for h,w in sizes:\n",
" #4 here to have the anchors overlap.\n",
" sized_aspects = 4 * (aspects * torch.tensor([2/h,2/w])).unsqueeze(0)\n",
" base_grid = create_grid((h,w)).unsqueeze(1)\n",
" n,a = base_grid.size(0),aspects.size(0)\n",
" ancs = torch.cat([base_grid.expand(n,a,2), sized_aspects.expand(n,a,2)], 2)\n",
" anchors.append(ancs.view(h,w,a,4))\n",
" return torch.cat([anc.view(-1,4) for anc in anchors],0) if flatten else anchors"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"ratios = [1/2,1,2]\n",
"#scales = [1,2**(-1/3), 2**(-2/3)]\n",
"scales = [1,2**(1/3), 2**(2/3)]\n",
"sizes = [(2**i,2**i) for i in range(5)]\n",
"sizes.reverse()\n",
"anchors = create_anchors(sizes, ratios, scales)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"anchors.size()\n",
"#[anc.size() for anc in anchors]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.cm as cmx\n",
"import matplotlib.colors as mcolors\n",
"from cycler import cycler\n",
"\n",
"def get_cmap(N):\n",
" color_norm = mcolors.Normalize(vmin=0, vmax=N-1)\n",
" return cmx.ScalarMappable(norm=color_norm, cmap='Set3').to_rgba\n",
"\n",
"num_color = 12\n",
"cmap = get_cmap(num_color)\n",
"color_list = [cmap(float(x)) for x in range(num_color)]\n",
"\n",
"def draw_outline(o, lw):\n",
" o.set_path_effects([patheffects.Stroke(\n",
" linewidth=lw, foreground='black'), patheffects.Normal()])\n",
"\n",
"def draw_rect(ax, b, color='white'):\n",
" patch = ax.add_patch(patches.Rectangle(b[:2], *b[-2:], fill=False, edgecolor=color, lw=2))\n",
" draw_outline(patch, 4)\n",
"\n",
"def draw_text(ax, xy, txt, sz=14, color='white'):\n",
" text = ax.text(*xy, txt,\n",
" verticalalignment='top', color=color, fontsize=sz, weight='bold')\n",
" draw_outline(text, 1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def show_boxes(boxes):\n",
" \"Show the `boxes` (size by 4)\"\n",
" _, ax = plt.subplots(1,1, figsize=(5,5))\n",
" ax.set_xlim(-1,1)\n",
" ax.set_ylim(1,-1)\n",
" for i, bbox in enumerate(boxes):\n",
" bb = bbox.numpy()\n",
" rect = [bb[1]-bb[3]/2, bb[0]-bb[2]/2, bb[3], bb[2]]\n",
" draw_rect(ax, rect, color=color_list[i%num_color])\n",
" draw_text(ax, [bb[1]-bb[3]/2,bb[0]-bb[2]/2], str(i), color=color_list[i%num_color])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"show_boxes(anchors[-9:])"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def activ_to_bbox(acts, anchors, flatten=True):\n",
" \"Extrapolate bounding boxes on anchors from the model activations.\"\n",
" if flatten:\n",
" acts.mul_(acts.new_tensor([[0.1, 0.1, 0.2, 0.2]]))\n",
" centers = anchors[...,2:] * acts[...,:2] + anchors[...,:2]\n",
" sizes = anchors[...,2:] * torch.exp(acts[...,:2])\n",
" return torch.cat([centers, sizes], -1)\n",
" else: return [activ_to_bbox(act,anc) for act,anc in zip(acts, anchors)]\n",
" return res"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"size=(3,4)\n",
"anchors = create_grid(size)\n",
"anchors = torch.cat([anchors, torch.tensor([2/size[0],2/size[1]]).expand_as(anchors)], 1)\n",
"activations = 0.1 * torch.randn(size[0]*size[1], 4)\n",
"bboxes = activ_to_bbox(activations, anchors)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"show_boxes(bboxes)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def cthw2tlbr(boxes):\n",
" \"Convert center/size format `boxes` to top/left bottom/right corners.\"\n",
" top_left = boxes[:,:2] - boxes[:,2:]/2\n",
" bot_right = boxes[:,:2] + boxes[:,2:]/2\n",
" return torch.cat([top_left, bot_right], 1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def intersection(anchors, targets):\n",
" \"Compute the sizes of the intersections of `anchors` by `targets`.\"\n",
" ancs, tgts = cthw2tlbr(anchors), cthw2tlbr(targets)\n",
" a, t = ancs.size(0), tgts.size(0)\n",
" ancs, tgts = ancs.unsqueeze(1).expand(a,t,4), tgts.unsqueeze(0).expand(a,t,4)\n",
" top_left_i = torch.max(ancs[...,:2], tgts[...,:2])\n",
" bot_right_i = torch.min(ancs[...,2:], tgts[...,2:])\n",
" sizes = torch.clamp(bot_right_i - top_left_i, min=0) \n",
" return sizes[...,0] * sizes[...,1]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"show_boxes(anchors)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"targets = torch.tensor([[0.,0.,2.,2.], [-0.5,-0.5,1.,1.], [1/3,0.5,0.5,0.5]])\n",
"show_boxes(targets)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"intersection(anchors, targets)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def IoU_values(anchors, targets):\n",
" \"Compute the IoU values of `anchors` by `targets`.\"\n",
" inter = intersection(anchors, targets)\n",
" anc_sz, tgt_sz = anchors[:,2] * anchors[:,3], targets[:,2] * targets[:,3]\n",
" union = anc_sz.unsqueeze(1) + tgt_sz.unsqueeze(0) - inter\n",
" return inter/(union+1e-8)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"IoU_values(anchors, targets)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Manually checked that those are right."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def match_anchors(anchors, targets, match_thr=0.5, bkg_thr=0.4):\n",
" \"Match `anchors` to targets. -1 is match to background, -2 is ignore.\"\n",
" ious = IoU_values(anchors, targets)\n",
" matches = anchors.new(anchors.size(0)).zero_().long() - 2\n",
" vals,idxs = torch.max(ious,1)\n",
" matches[vals < bkg_thr] = -1\n",
" matches[vals > match_thr] = idxs[vals > match_thr]\n",
" #Overwrite matches with each target getting the anchor that has the max IoU.\n",
" #vals,idxs = torch.max(ious,0)\n",
" #If idxs contains repetition, this doesn't bug and only the last is considered.\n",
" #matches[idxs] = targets.new_tensor(list(range(targets.size(0)))).long()\n",
" return matches"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Last example"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"match_anchors(anchors, targets)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"With anchors very close to the targets."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"size=(3,4)\n",
"anchors = create_grid(size)\n",
"anchors = torch.cat([anchors, torch.tensor([2/size[0],2/size[1]]).expand_as(anchors)], 1)\n",
"activations = 0.1 * torch.randn(size[0]*size[1], 4)\n",
"bboxes = activ_to_bbox(activations, anchors)\n",
"match_anchors(anchors,bboxes)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"With anchors in the grey area."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"anchors = create_grid((2,2))\n",
"anchors = torch.cat([anchors, torch.tensor([1.,1.]).expand_as(anchors)], 1)\n",
"targets = anchors.clone()\n",
"anchors = torch.cat([anchors, torch.tensor([[-0.5,0.,1.,1.8]])], 0)\n",
"match_anchors(anchors,targets)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def tlbr2cthw(boxes):\n",
" \"Convert top/left bottom/right format `boxes` to center/size corners.\"\n",
" center = (boxes[:,:2] + boxes[:,2:])/2\n",
" sizes = boxes[:,2:] - boxes[:,:2]\n",
" return torch.cat([center, sizes], 1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def bbox_to_activ(bboxes, anchors, flatten=True):\n",
" \"Return the target of the model on `anchors` for the `bboxes`.\"\n",
" if flatten:\n",
" t_centers = (bboxes[...,:2] - anchors[...,:2]) / anchors[...,2:] \n",
" t_sizes = torch.log(bboxes[...,2:] / anchors[...,2:] + 1e-8) \n",
" return torch.cat([t_centers, t_sizes], -1).div_(bboxes.new_tensor([[0.1, 0.1, 0.2, 0.2]]))\n",
" else: return [activ_to_bbox(act,anc) for act,anc in zip(acts, anchors)]\n",
" return res"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def encode_class(idxs, n_classes):\n",
" target = idxs.new_zeros(len(idxs), n_classes).float()\n",
" mask = idxs != 0\n",
" i1s = LongTensor(list(range(len(idxs))))\n",
" target[i1s[mask],idxs[mask]-1] = 1\n",
" return target"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"encode_class(LongTensor([1,2,0,1,3]),3)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"class RetinaNetFocalLoss(nn.Module):\n",
" \n",
" def __init__(self, gamma:float=2., alpha:float=0.25, pad_idx:int=0, scales:Collection[float]=None, \n",
" ratios:Collection[float]=None, reg_loss:LossFunction=F.smooth_l1_loss):\n",
" super().__init__()\n",
" self.gamma,self.alpha,self.pad_idx,self.reg_loss = gamma,alpha,pad_idx,reg_loss\n",
" self.scales = ifnone(scales, [1,2**(-1/3), 2**(-2/3)])\n",
" self.ratios = ifnone(ratios, [1/2,1,2])\n",
" \n",
" def _change_anchors(self, sizes:Sizes) -> bool:\n",
" if not hasattr(self, 'sizes'): return True\n",
" for sz1, sz2 in zip(self.sizes, sizes):\n",
" if sz1[0] != sz2[0] or sz1[1] != sz2[1]: return True\n",
" return False\n",
" \n",
" def _create_anchors(self, sizes:Sizes, device:torch.device):\n",
" self.sizes = sizes\n",
" self.anchors = create_anchors(sizes, self.ratios, self.scales).to(device)\n",
" \n",
" def _unpad(self, bbox_tgt, clas_tgt):\n",
" i = torch.min(torch.nonzero(clas_tgt-self.pad_idx))\n",
" return tlbr2cthw(bbox_tgt[i:]), clas_tgt[i:]-1+self.pad_idx\n",
" \n",
" def _focal_loss(self, clas_pred, clas_tgt):\n",
" encoded_tgt = encode_class(clas_tgt, clas_pred.size(1))\n",
" ps = torch.sigmoid(clas_pred)\n",
" weights = encoded_tgt * (1-ps) + (1-encoded_tgt) * ps\n",
" alphas = (1-encoded_tgt) * self.alpha + encoded_tgt * (1-self.alpha)\n",
" weights.pow_(self.gamma).mul_(alphas)\n",
" clas_loss = F.binary_cross_entropy_with_logits(clas_pred, encoded_tgt, weights, reduction='sum')\n",
" return clas_loss\n",
" \n",
" def _one_loss(self, clas_pred, bbox_pred, clas_tgt, bbox_tgt):\n",
" bbox_tgt, clas_tgt = self._unpad(bbox_tgt, clas_tgt)\n",
" matches = match_anchors(self.anchors, bbox_tgt)\n",
" bbox_mask = matches>=0\n",
" if bbox_mask.sum() != 0:\n",
" bbox_pred = bbox_pred[bbox_mask]\n",
" bbox_tgt = bbox_tgt[matches[bbox_mask]]\n",
" bb_loss = self.reg_loss(bbox_pred, bbox_to_activ(bbox_tgt, self.anchors[bbox_mask]))\n",
" else: bb_loss = 0.\n",
" matches.add_(1)\n",
" clas_tgt = clas_tgt + 1\n",
" clas_mask = matches>=0\n",
" clas_pred = clas_pred[clas_mask]\n",
" clas_tgt = torch.cat([clas_tgt.new_zeros(1).long(), clas_tgt])\n",
" clas_tgt = clas_tgt[matches[clas_mask]]\n",
" return bb_loss + self._focal_loss(clas_pred, clas_tgt)/torch.clamp(bbox_mask.sum(), min=1.)\n",
" \n",
" def forward(self, output, bbox_tgts, clas_tgts):\n",
" clas_preds, bbox_preds, sizes = output\n",
" if self._change_anchors(sizes): self._create_anchors(sizes, clas_preds.device)\n",
" n_classes = clas_preds.size(2)\n",
" return sum([self._one_loss(cp, bp, ct, bt)\n",
" for (cp, bp, ct, bt) in zip(clas_preds, bbox_preds, clas_tgts, bbox_tgts)])/clas_tgts.size(0)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Alternative to the L1 smooth loss used in online implementations"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"class SigmaL1SmoothLoss(nn.Module):\n",
"\n",
" def forward(self, output, target):\n",
" reg_diff = torch.abs(target - output)\n",
" reg_loss = torch.where(torch.le(reg_diff, 1/9), 4.5 * torch.pow(reg_diff, 2), reg_diff - 1/18)\n",
" return reg_loss.mean()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Sketch to test the loss"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"LongTensor([[[0,0,64,128,0], [32,64,128,128,1]], [[128,96,256,192,2], [96,192,128,256,3]]]).float().cuda()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"tgt_clas = LongTensor([[1,2], [3,4]])\n",
"tgt_bbox = FloatTensor([[[0,0,128,64], [64,32,128,128]], [[96,128,192,256], [192,96,256,128]]])\n",
"tgt_bbox = tgt_bbox / 128 - 1.\n",
"y = [tgt_bbox.cuda(), tgt_clas.cuda()]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"clas = torch.load(PATH/'models'/'tst_clas.pth')\n",
"regr = torch.load(PATH/'models'/'tst_regr.pth')\n",
"sizes = [[32, 32], [16, 16], [8, 8], [4, 4], [2, 2]]\n",
"output = [logit(clas), regr, sizes]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"crit(output, *y)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Checking the output"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def unpad(tgt_bbox, tgt_clas, pad_idx=0):\n",
" i = torch.min(torch.nonzero(tgt_clas-pad_idx))\n",
" return tlbr2cthw(tgt_bbox[i:]), tgt_clas[i:]-1+pad_idx"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"idx = 0\n",
"clas_pred,bbox_pred,sizes = output[0][idx].cpu(), output[1][idx].cpu(), output[2]\n",
"bbox_tgt, clas_tgt = y[0][idx].cpu(),y[1][idx].cpu()\n",
"bbox_tgt, clas_tgt = unpad(bbox_tgt, clas_tgt)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"bbox_tgt"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"anchors = create_anchors(sizes, ratios, scales)\n",
"ious = IoU_values(anchors, bbox_tgt)\n",
"matches = match_anchors(anchors, bbox_tgt)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"ious[-9:]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"(matches==-2).sum(), (matches==-1).sum(), (matches>=0).sum()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"bbox_mask = matches>=0\n",
"bbox_pred = bbox_pred[bbox_mask]\n",
"bbox_tgt = bbox_tgt[matches[bbox_mask]]\n",
"bb_loss = F.smooth_l1_loss(bbox_pred, bbox_to_activ(bbox_tgt, anchors[bbox_mask]))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"F.smooth_l1_loss(bbox_pred, bbox_to_activ(bbox_tgt, anchors[bbox_mask]))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"tst_loss = SigmaL1SmoothLoss()\n",
"tst_loss(bbox_pred, bbox_to_activ(bbox_tgt, anchors[bbox_mask]))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"crit.reg_loss(bbox_pred, bbox_to_activ(bbox_tgt, anchors[bbox_mask]))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"matches.add_(1)\n",
"clas_tgt += 1\n",
"clas_mask = matches>=0\n",
"clas_pred = clas_pred[clas_mask]\n",
"clas_tgt = torch.cat([clas_tgt.new_zeros(1).long(), clas_tgt])\n",
"clas_tgt = clas_tgt[matches[clas_mask]]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Focal loss"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"alpha, gamma, n_classes = 0.25, 2., 6\n",
"encoded_tgt = encode_class(clas_tgt, n_classes)\n",
"ps = torch.sigmoid(clas_pred)\n",
"weights = encoded_tgt * (1-ps) + (1-encoded_tgt) * ps\n",
"alphas = encoded_tgt * alpha + (1-encoded_tgt) * (1-alpha)\n",
"weights.pow_(gamma).mul_(alphas)\n",
"clas_loss = F.binary_cross_entropy_with_logits(clas_pred, encoded_tgt, weights, reduction='sum') / bbox_mask.sum()\n",
"clas_loss"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's look at the objects missclassified."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"clas_pred[clas_tgt.nonzero().squeeze()]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"F.binary_cross_entropy_with_logits(clas_pred[clas_tgt.nonzero().squeeze()], encoded_tgt[clas_tgt.nonzero().squeeze()], weights[clas_tgt.nonzero().squeeze()], reduction='sum') / bbox_mask.sum()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"They account for half the loss!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Training"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"n_classes = 6\n",
"encoder = create_body(tvm.resnet50(True), -2)\n",
"model = RetinaNet(encoder, n_classes,final_bias=-4) \n",
"crit = RetinaNetFocalLoss(scales=scales, ratios=ratios)\n",
"learn = Learner(data, model, loss_fn=crit)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.split([model.encoder[6], model.c5top5])\n",
"learn.freeze()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.lr_find()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.recorder.plot()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.fit_one_cycle(1, 1e-4)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.save('sample')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Inference"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"learn.load('sample')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"img,target = next(iter(data.valid_dl))\n",
"with torch.no_grad():\n",
" output = model(img)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"torch.save(img, PATH/'models'/'tst_input.pth')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def _draw_outline(o:Patch, lw:int):\n",
" \"Outline bounding box onto image `Patch`.\"\n",
" o.set_path_effects([patheffects.Stroke(\n",
" linewidth=lw, foreground='black'), patheffects.Normal()])\n",
"\n",
"def draw_rect(ax:plt.Axes, b:Collection[int], color:str='white', text=None, text_size=14):\n",
" \"Draw bounding box on `ax`.\"\n",
" patch = ax.add_patch(patches.Rectangle(b[:2], *b[-2:], fill=False, edgecolor=color, lw=2))\n",
" _draw_outline(patch, 4)\n",
" if text is not None:\n",
" patch = ax.text(*b[:2], text, verticalalignment='top', color=color, fontsize=text_size, weight='bold')\n",
" _draw_outline(patch,1)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def show_preds(img, output, idx, detect_thresh=0.3, classes=None):\n",
" clas_pred,bbox_pred,sizes = output[0][idx].cpu(), output[1][idx].cpu(), output[2]\n",
" anchors = create_anchors(sizes, ratios, scales)\n",
" bbox_pred = activ_to_bbox(bbox_pred, anchors)\n",
" clas_pred = torch.sigmoid(clas_pred)\n",
" detect_mask = clas_pred.max(1)[0] > detect_thresh\n",
" bbox_pred, clas_pred = bbox_pred[detect_mask], clas_pred[detect_mask]\n",
" t_sz = torch.Tensor([*img.size])[None].float()\n",
" bbox_pred[:,:2] = bbox_pred[:,:2] - bbox_pred[:,2:]/2\n",
" bbox_pred[:,:2] = (bbox_pred[:,:2] + 1) * t_sz/2\n",
" bbox_pred[:,2:] = bbox_pred[:,2:] * t_sz\n",
" bbox_pred = bbox_pred.long()\n",
" _, ax = plt.subplots(1,1)\n",
" for bbox, c in zip(bbox_pred, clas_pred.argmax(1)):\n",
" img.show(ax=ax)\n",
" txt = str(c.item()) if classes is None else classes[c.item()+1]\n",
" draw_rect(ax, [bbox[1],bbox[0],bbox[3],bbox[2]], text=txt)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"idx = 0\n",
"img = data.valid_ds[idx][0]\n",
"classes = data.train_ds.classes\n",
"show_preds(img, output, idx, detect_thresh=0.2, classes=classes)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def nms(boxes, scores, thresh=0.5):\n",
" idx_sort = scores.argsort(descending=True)\n",
" boxes, scores = boxes[idx_sort], scores[idx_sort]\n",
" to_keep, indexes = [], torch.LongTensor(range_of(scores))\n",
" while len(scores) > 0:\n",
" #pdb.set_trace()\n",
" to_keep.append(idx_sort[indexes[0]])\n",
" iou_vals = IoU_values(boxes, boxes[:1]).squeeze()\n",
" mask_keep = iou_vals <= thresh\n",
" if len(mask_keep.nonzero()) == 0: break\n",
" idx_first = mask_keep.nonzero().min().item()\n",
" boxes, scores, indexes = boxes[mask_keep], scores[mask_keep], indexes[mask_keep]\n",
" return LongTensor(to_keep)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def process_output(output, i, detect_thresh=0.25):\n",
" clas_pred,bbox_pred,sizes = output[0][i], output[1][i], output[2]\n",
" anchors = create_anchors(sizes, ratios, scales).to(clas_pred.device)\n",
" bbox_pred = activ_to_bbox(bbox_pred, anchors)\n",
" clas_pred = torch.sigmoid(clas_pred)\n",
" detect_mask = clas_pred.max(1)[0] > detect_thresh\n",
" bbox_pred, clas_pred = bbox_pred[detect_mask], clas_pred[detect_mask]\n",
" bbox_pred = tlbr2cthw(torch.clamp(cthw2tlbr(bbox_pred), min=-1, max=1)) \n",
" scores, preds = clas_pred.max(1)\n",
" return bbox_pred, scores, preds"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def show_preds(img, output, idx, detect_thresh=0.25, classes=None):\n",
" bbox_pred, scores, preds = process_output(output, idx, detect_thresh)\n",
" to_keep = nms(bbox_pred, scores)\n",
" bbox_pred, preds, scores = bbox_pred[to_keep].cpu(), preds[to_keep].cpu(), scores[to_keep].cpu()\n",
" t_sz = torch.Tensor([*img.size])[None].float()\n",
" bbox_pred[:,:2] = bbox_pred[:,:2] - bbox_pred[:,2:]/2\n",
" bbox_pred[:,:2] = (bbox_pred[:,:2] + 1) * t_sz/2\n",
" bbox_pred[:,2:] = bbox_pred[:,2:] * t_sz\n",
" bbox_pred = bbox_pred.long()\n",
" _, ax = plt.subplots(1,1)\n",
" for bbox, c, scr in zip(bbox_pred, preds, scores):\n",
" img.show(ax=ax)\n",
" txt = str(c.item()) if classes is None else classes[c.item()+1]\n",
" draw_rect(ax, [bbox[1],bbox[0],bbox[3],bbox[2]], text=f'{txt} {scr:.2f}')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"idx = 0\n",
"img = data.valid_ds[idx][0]\n",
"show_preds(img, output, idx, detect_thresh=0.2, classes=data.classes)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def get_predictions(output, idx, detect_thresh=0.05):\n",
" bbox_pred, scores, preds = process_output(output, idx, detect_thresh)\n",
" to_keep = nms(bbox_pred, scores)\n",
" return bbox_pred[to_keep], preds[to_keep], scores[to_keep]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"get_predictions(output, 0)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## mAP"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def compute_ap(precision, recall):\n",
" \"Compute the average precision for `precision` and `recall` curve.\"\n",
" recall = np.concatenate(([0.], list(recall), [1.]))\n",
" precision = np.concatenate(([0.], list(precision), [0.]))\n",
" for i in range(len(precision) - 1, 0, -1):\n",
" precision[i - 1] = np.maximum(precision[i - 1], precision[i])\n",
" idx = np.where(recall[1:] != recall[:-1])[0]\n",
" ap = np.sum((recall[idx + 1] - recall[idx]) * precision[idx + 1])\n",
" return ap"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"#export\n",
"def compute_class_AP(model, dl, n_classes, iou_thresh=0.5, detect_thresh=0.05, num_keep=100):\n",
" tps, clas, p_scores = [], [], []\n",
" classes, n_gts = LongTensor(range(n_classes)),torch.zeros(n_classes).long()\n",
" with torch.no_grad():\n",
" for input,target in progress_bar(dl):\n",
" output = model(input)\n",
" for i in range(target[0].size(0)):\n",
" bbox_pred, preds, scores = get_predictions(output, i, detect_thresh)\n",
" tgt_bbox, tgt_clas = unpad(target[0][i], target[1][i])\n",
" ious = IoU_values(bbox_pred, tgt_bbox)\n",
" max_iou, matches = ious.max(1)\n",
" detected = []\n",
" for i in range_of(preds):\n",
" if max_iou[i] >= iou_thresh and matches[i] not in detected and tgt_clas[matches[i]] == preds[i]:\n",
" detected.append(matches[i])\n",
" tps.append(1)\n",
" else: tps.append(0)\n",
" clas.append(preds.cpu())\n",
" p_scores.append(scores.cpu())\n",
" n_gts += (tgt_clas.cpu()[:,None] == classes[None,:]).sum(0)\n",
" tps, p_scores, clas = torch.tensor(tps), torch.cat(p_scores,0), torch.cat(clas,0)\n",
" fps = 1-tps\n",
" idx = p_scores.argsort(descending=True)\n",
" tps, fps, clas = tps[idx], fps[idx], clas[idx]\n",
" aps = []\n",
" #return tps, clas\n",
" for cls in range(n_classes):\n",
" tps_cls, fps_cls = tps[clas==cls].float().cumsum(0), fps[clas==cls].float().cumsum(0)\n",
" if tps_cls[-1] != 0:\n",
" precision = tps_cls / (tps_cls + fps_cls + 1e-8)\n",
" recall = tps_cls / (n_gts[cls] + 1e-8)\n",
" aps.append(compute_ap(precision, recall))\n",
" else: aps.append(0.)\n",
" return aps"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"L = compute_class_AP(learn.model, tst_dl, 6)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"L[0]"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Environment (conda_fastai)",
"language": "python",
"name": "conda_fastai"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.0"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment