chore: import upstream snapshot with attribution
This commit is contained in:
wehub-resource-sync
@@ -0,0 +1,39 @@
|
||||
{
|
||||
"<h1>U-Net</h1>\n<p>This is an implementation of the U-Net model from the paper, <a href=\"https://arxiv.org/abs/1505.04597\">U-Net: Convolutional Networks for Biomedical Image Segmentation</a>.</p>\n<p>U-Net consists of a contracting path and an expansive path. The contracting path is a series of convolutional layers and pooling layers, where the resolution of the feature map gets progressively reduced. Expansive path is a series of up-sampling layers and convolutional layers where the resolution of the feature map gets progressively increased.</p>\n<p>At every step in the expansive path the corresponding feature map from the contracting path concatenated with the current feature map.</p>\n<p><span translate=no>_^_0_^_</span></p>\n<p>Here is the <a href=\"experiment.html\">training code</a> for an experiment that trains a U-Net on <a href=\"carvana.html\">Carvana dataset</a>.</p>\n": "<h1>\u30e6\u30fc\u30cd\u30c3\u30c8</h1>\n<p>\u3053\u308c\u306f\u3001\u8ad6\u6587\u300cU-Net<a href=\"https://arxiv.org/abs/1505.04597\">: \u751f\u7269\u533b\u5b66\u753b\u50cf\u30bb\u30b0\u30e1\u30f3\u30c6\u30fc\u30b7\u30e7\u30f3\u306e\u305f\u3081\u306e\u7573\u307f\u8fbc\u307f\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u300d\u306eU-Net\u30e2\u30c7\u30eb\u306e\u5b9f\u88c5\u3067\u3059</a>\u3002</p>\n<p>U-Net\u306f\u53ce\u7e2e\u7d4c\u8def\u3068\u62e1\u5f35\u7d4c\u8def\u3067\u69cb\u6210\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u53ce\u7e2e\u7d4c\u8def\u306f\u4e00\u9023\u306e\u7573\u307f\u8fbc\u307f\u5c64\u3068\u30d7\u30fc\u30ea\u30f3\u30b0\u5c64\u3067\u3042\u308a\u3001\u7279\u5fb4\u30de\u30c3\u30d7\u306e\u89e3\u50cf\u5ea6\u306f\u5f90\u3005\u306b\u4f4e\u4e0b\u3057\u307e\u3059\u3002\u30a8\u30af\u30b9\u30d1\u30f3\u30b7\u30d6\u30d1\u30b9\u3068\u306f\u3001\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u306e\u89e3\u50cf\u5ea6\u304c\u5f90\u3005\u306b\u4e0a\u304c\u3063\u3066\u3044\u304f\u4e00\u9023\u306e\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u30ec\u30a4\u30e4\u30fc\u3068\u7573\u307f\u8fbc\u307f\u30ec\u30a4\u30e4\u30fc\u306e\u3053\u3068\u3067\u3059</p>\u3002\n<p>\u62e1\u5f35\u30d1\u30b9\u306e\u5404\u30b9\u30c6\u30c3\u30d7\u3067\u3001\u7e2e\u5c0f\u30d1\u30b9\u304b\u3089\u306e\u5bfe\u5fdc\u3059\u308b\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u304c\u73fe\u5728\u306e\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u3068\u9023\u7d50\u3055\u308c\u307e\u3059\u3002</p>\n<p><span translate=no>_^_0_^_</span></p>\n<p>\u3053\u308c\u306f\u3001<a href=\"experiment.html\"><a href=\"carvana.html\">Carvana\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067U-Net\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3059\u308b\u5b9f\u9a13\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30b3\u30fc\u30c9\u3067\u3059</a></a>\u3002</p>\n",
|
||||
"<h2>U-Net</h2>\n": "<h2>\u30e6\u30fc\u30cd\u30c3\u30c8</h2>\n",
|
||||
"<h3>Crop and Concatenate the feature map</h3>\n<p>At every step in the expansive path the corresponding feature map from the contracting path concatenated with the current feature map.</p>\n": "<h3>\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u306e\u30c8\u30ea\u30df\u30f3\u30b0\u3068\u9023\u7d50</h3>\n<p>\u62e1\u5f35\u30d1\u30b9\u306e\u5404\u30b9\u30c6\u30c3\u30d7\u3067\u3001\u7e2e\u5c0f\u30d1\u30b9\u304b\u3089\u306e\u5bfe\u5fdc\u3059\u308b\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u304c\u73fe\u5728\u306e\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u3068\u9023\u7d50\u3055\u308c\u307e\u3059\u3002</p>\n",
|
||||
"<h3>Down-sample</h3>\n<p>Each step in the contracting path down-samples the feature map with a <span translate=no>_^_0_^_</span> max pooling layer.</p>\n": "<h3>\u30c0\u30a6\u30f3\u30b5\u30f3\u30d7\u30eb</h3>\n<p>\u53ce\u7e2e\u30d1\u30b9\u306e\u5404\u30b9\u30c6\u30c3\u30d7\u306f\u3001<span translate=no>_^_0_^_</span>\u6700\u5927\u30d7\u30fc\u30ea\u30f3\u30b0\u30ec\u30a4\u30e4\u30fc\u3092\u4f7f\u7528\u3057\u3066\u7279\u5fb4\u30de\u30c3\u30d7\u3092\u30c0\u30a6\u30f3\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3057\u307e\u3059\u3002</p>\n",
|
||||
"<h3>Two <span translate=no>_^_0_^_</span> Convolution Layers</h3>\n<p>Each step in the contraction path and expansive path have two <span translate=no>_^_1_^_</span> convolutional layers followed by ReLU activations.</p>\n<p>In the U-Net paper they used <span translate=no>_^_2_^_</span> padding, but we use <span translate=no>_^_3_^_</span> padding so that final feature map is not cropped.</p>\n": "<h3>2 <span translate=no>_^_0_^_</span> \u3064\u306e\u30b3\u30f3\u30dc\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u30ec\u30a4\u30e4\u30fc</h3>\n<p>\u53ce\u7e2e\u7d4c\u8def\u3068\u81a8\u5f35\u7d4c\u8def\u306e\u5404\u30b9\u30c6\u30c3\u30d7\u306b\u306f\u3001<span translate=no>_^_1_^_</span> 2\u3064\u306e\u7573\u307f\u8fbc\u307f\u5c64\u304c\u3042\u308a\u3001\u305d\u306e\u5f8c\u306bReLU\u6d3b\u6027\u5316\u304c\u7d9a\u304d\u307e\u3059\u3002</p>\n<p><span translate=no>_^_2_^_</span>U-Net\u306e\u8ad6\u6587\u3067\u306f\u30d1\u30c7\u30a3\u30f3\u30b0\u3092\u4f7f\u7528\u3057\u3066\u3044\u307e\u3057\u305f\u304c\u3001<span translate=no>_^_3_^_</span>\u6700\u7d42\u7684\u306a\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u304c\u30c8\u30ea\u30df\u30f3\u30b0\u3055\u308c\u306a\u3044\u3088\u3046\u306b\u30d1\u30c7\u30a3\u30f3\u30b0\u3092\u4f7f\u7528\u3057\u3066\u3044\u307e\u3059\u3002</p>\n",
|
||||
"<h3>Up-sample</h3>\n<p>Each step in the expansive path up-samples the feature map with a <span translate=no>_^_0_^_</span> up-convolution.</p>\n": "<h3>\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30eb</h3>\n<p>\u5e83\u5927\u306a\u7d4c\u8def\u306e\u5404\u30b9\u30c6\u30c3\u30d7\u306f\u3001\u30a2\u30c3\u30d7\u30b3\u30f3\u30dc\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u3067\u7279\u5fb4\u30de\u30c3\u30d7\u3092\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3057\u307e\u3059\u3002<span translate=no>_^_0_^_</span></p>\n",
|
||||
"<p> </p>\n": "<p></p>\n",
|
||||
"<p>Apply the two convolution layers and activations </p>\n": "<p>2 \u3064\u306e\u30b3\u30f3\u30dc\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3\u30ec\u30a4\u30e4\u30fc\u3068\u30a2\u30af\u30c6\u30a3\u30d9\u30fc\u30b7\u30e7\u30f3\u3092\u9069\u7528\u3057\u307e\u3059\u3002</p>\n",
|
||||
"<p>Collect the output </p>\n": "<p>\u30a2\u30a6\u30c8\u30d7\u30c3\u30c8\u3092\u53ce\u96c6</p>\n",
|
||||
"<p>Concatenate the feature maps </p>\n": "<p>\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u3092\u9023\u7d50\u3059\u308b</p>\n",
|
||||
"<p>Concatenate the output of the contracting path </p>\n": "<p>\u30b3\u30f3\u30c8\u30e9\u30af\u30c8\u30d1\u30b9\u306e\u51fa\u529b\u3092\u9023\u7d50\u3059\u308b</p>\n",
|
||||
"<p>Contracting path </p>\n": "<p>\u5951\u7d04\u7d4c\u8def</p>\n",
|
||||
"<p>Crop and concatenate layers for the expansive path. </p>\n": "<p>\u30ec\u30a4\u30e4\u30fc\u3092\u30c8\u30ea\u30df\u30f3\u30b0\u3057\u3066\u9023\u7d50\u3059\u308b\u3068\u3001\u5e83\u5927\u306a\u30d1\u30b9\u304c\u4f5c\u308c\u307e\u3059\u3002</p>\n",
|
||||
"<p>Crop the feature map from the contracting path to the size of the current feature map </p>\n": "<p>\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u3092\u7e2e\u5c0f\u30d1\u30b9\u304b\u3089\u73fe\u5728\u306e\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u306e\u30b5\u30a4\u30ba\u306b\u30c8\u30ea\u30df\u30f3\u30b0\u3057\u307e\u3059</p>\n",
|
||||
"<p>Double convolution layers for the contracting path. The number of features gets doubled at each step starting from <span translate=no>_^_0_^_</span>. </p>\n": "<p>\u53ce\u7e2e\u7d4c\u8def\u7528\u306e\u4e8c\u91cd\u7573\u307f\u8fbc\u307f\u5c64\u3002<span translate=no>_^_0_^_</span>\u304b\u3089\u59cb\u307e\u308b\u5404\u30b9\u30c6\u30c3\u30d7\u3067\u6a5f\u80fd\u306e\u6570\u304c 2 \u500d\u306b\u306a\u308a\u307e\u3059</p>\u3002\n",
|
||||
"<p>Double convolution layers for the expansive path. Their input is the concatenation of the current feature map and the feature map from the contracting path. Therefore, the number of input features is double the number of features from up-sampling. </p>\n": "<p>\u81a8\u5f35\u7d4c\u8def\u7528\u306e\u4e8c\u91cd\u7573\u307f\u8fbc\u307f\u5c64\u3002\u305d\u308c\u3089\u306e\u5165\u529b\u306f\u3001\u73fe\u5728\u306e\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u3068\u7e2e\u5c0f\u30d1\u30b9\u304b\u3089\u306e\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u3092\u9023\u7d50\u3057\u305f\u3082\u306e\u3067\u3059\u3002\u3057\u305f\u304c\u3063\u3066\u3001\u5165\u529b\u30d5\u30a3\u30fc\u30c1\u30e3\u306e\u6570\u306f\u3001\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u306b\u3088\u308b\u30d5\u30a3\u30fc\u30c1\u30e3\u6570\u306e2\u500d\u306b\u306a\u308a\u307e\u3059</p>\u3002\n",
|
||||
"<p>Down sampling layers for the contracting path </p>\n": "<p>\u53ce\u7e2e\u7d4c\u8def\u7528\u306e\u30c0\u30a6\u30f3\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u5c64</p>\n",
|
||||
"<p>Down-sample </p>\n": "<p>\u30c0\u30a6\u30f3\u30b5\u30f3\u30d7\u30eb</p>\n",
|
||||
"<p>Expansive path </p>\n": "<p>\u5e83\u5927\u306a\u9053</p>\n",
|
||||
"<p>Final <span translate=no>_^_0_^_</span> convolution layer </p>\n": "<p><span translate=no>_^_0_^_</span>\u6700\u7d42\u7573\u307f\u8fbc\u307f\u5c64</p>\n",
|
||||
"<p>Final <span translate=no>_^_0_^_</span> convolution layer to produce the output </p>\n": "<p><span translate=no>_^_0_^_</span>\u51fa\u529b\u3092\u751f\u6210\u3059\u308b\u6700\u5f8c\u306e\u7573\u307f\u8fbc\u307f\u5c64</p>\n",
|
||||
"<p>First <span translate=no>_^_0_^_</span> convolutional layer </p>\n": "<p><span translate=no>_^_0_^_</span>\u6700\u521d\u306e\u7573\u307f\u8fbc\u307f\u5c64</p>\n",
|
||||
"<p>Max pooling layer </p>\n": "<p>\u6700\u5927\u30d7\u30fc\u30ea\u30f3\u30b0\u5c64</p>\n",
|
||||
"<p>Second <span translate=no>_^_0_^_</span> convolutional layer </p>\n": "<p>2 <span translate=no>_^_0_^_</span> \u756a\u76ee\u306e\u7573\u307f\u8fbc\u307f\u5c64</p>\n",
|
||||
"<p>The two convolution layers at the lowest resolution (the bottom of the U). </p>\n": "<p>\u6700\u3082\u4f4e\u3044\u89e3\u50cf\u5ea6\u306e 2 \u3064\u306e\u7573\u307f\u8fbc\u307f\u5c64 (U \u306e\u4e0b\u90e8)\u3002</p>\n",
|
||||
"<p>To collect the outputs of contracting path for later concatenation with the expansive path. </p>\n": "<p>\u7e2e\u5c0f\u30d1\u30b9\u306e\u51fa\u529b\u3092\u53ce\u96c6\u3057\u3066\u3001\u5f8c\u3067\u62e1\u5f35\u30d1\u30b9\u3068\u9023\u7d50\u3067\u304d\u308b\u3088\u3046\u306b\u3057\u307e\u3059\u3002</p>\n",
|
||||
"<p>Two <span translate=no>_^_0_^_</span> convolutional layers </p>\n": "<p>2 <span translate=no>_^_0_^_</span> \u3064\u306e\u7573\u307f\u8fbc\u307f\u5c64</p>\n",
|
||||
"<p>Two <span translate=no>_^_0_^_</span> convolutional layers at the bottom of the U-Net </p>\n": "<p><span translate=no>_^_0_^_</span>U-Net\u306e\u4e0b\u90e8\u306b\u3042\u308b2\u3064\u306e\u7573\u307f\u8fbc\u307f\u5c64</p>\n",
|
||||
"<p>Up sampling layers for the expansive path. The number of features is halved with up-sampling. </p>\n": "<p>\u5e83\u5927\u306a\u7d4c\u8def\u306e\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u30ec\u30a4\u30e4\u30fc\u3092\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3002\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u3092\u884c\u3046\u3068\u3001\u6a5f\u80fd\u306e\u6570\u306f\u534a\u5206\u306b\u306a\u308a\u307e\u3059</p>\u3002\n",
|
||||
"<p>Up-convolution </p>\n": "<p>\u30a2\u30c3\u30d7\u30b3\u30f3\u30dc\u30ea\u30e5\u30fc\u30b7\u30e7\u30f3</p>\n",
|
||||
"<p>Up-sample </p>\n": "<p>\u30a2\u30c3\u30d7\u30b5\u30f3\u30d7\u30eb</p>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> current feature map in the expansive path </li>\n<li><span translate=no>_^_1_^_</span> corresponding feature map from the contracting path</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u30a8\u30af\u30b9\u30d1\u30f3\u30b7\u30d6\u30d1\u30b9\u5185\u306e\u73fe\u5728\u306e\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7</li>\n<li><span translate=no>_^_1_^_</span>\u5951\u7d04\u7d4c\u8def\u304b\u3089\u306e\u5bfe\u5fdc\u3059\u308b\u6a5f\u80fd\u30de\u30c3\u30d7</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> input image</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u5165\u529b\u753b\u50cf</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> is the number of input channels </li>\n<li><span translate=no>_^_1_^_</span> is the number of output channels</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u306f\u5165\u529b\u30c1\u30e3\u30f3\u30cd\u30eb\u6570</li>\n<li><span translate=no>_^_1_^_</span>\u306f\u51fa\u529b\u30c1\u30e3\u30f3\u30cd\u30eb\u6570</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> number of channels in the input image </li>\n<li><span translate=no>_^_1_^_</span> number of channels in the result feature map</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u5165\u529b\u753b\u50cf\u306e\u30c1\u30e3\u30f3\u30cd\u30eb\u6570</li>\n<li><span translate=no>_^_1_^_</span>\u7d50\u679c\u30d5\u30a3\u30fc\u30c1\u30e3\u30de\u30c3\u30d7\u306e\u30c1\u30e3\u30cd\u30eb\u6570</li></ul>\n",
|
||||
"PyTorch implementation and tutorial of U-Net model.": "PyTorch\u306e\u5b9f\u88c5\u3068U-Net\u30e2\u30c7\u30eb\u306e\u30c1\u30e5\u30fc\u30c8\u30ea\u30a2\u30eb\u3002",
|
||||
"U-Net": "\u30e6\u30fc\u30cd\u30c3\u30c8"
|
||||
}
|
||||
@@ -0,0 +1,39 @@
|
||||
{
|
||||
"<h1>U-Net</h1>\n<p>This is an implementation of the U-Net model from the paper, <a href=\"https://arxiv.org/abs/1505.04597\">U-Net: Convolutional Networks for Biomedical Image Segmentation</a>.</p>\n<p>U-Net consists of a contracting path and an expansive path. The contracting path is a series of convolutional layers and pooling layers, where the resolution of the feature map gets progressively reduced. Expansive path is a series of up-sampling layers and convolutional layers where the resolution of the feature map gets progressively increased.</p>\n<p>At every step in the expansive path the corresponding feature map from the contracting path concatenated with the current feature map.</p>\n<p><span translate=no>_^_0_^_</span></p>\n<p>Here is the <a href=\"experiment.html\">training code</a> for an experiment that trains a U-Net on <a href=\"carvana.html\">Carvana dataset</a>.</p>\n": "<h1>\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca</h1>\n<p>\u0db8\u0dd9\u0dba\u0d9a\u0da9\u0daf\u0dcf\u0dc3\u0dd2 \u0dc0\u0dbd\u0dd2\u0db1\u0dca U-Net \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dd2\u0dbb\u0dd3\u0db8, <a href=\"https://arxiv.org/abs/1505.04597\">U-Net: \u0da2\u0ddb\u0dc0 \u0dc0\u0ddb\u0daf\u0dca\u0dba \u0dbb\u0dd6\u0db4 \u0d9b\u0dab\u0dca\u0da9\u0db1\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0dc3\u0d82\u0dba\u0dd4\u0d9a\u0dca\u0dad \u0da2\u0dcf\u0dbd</a>. </p>\n<p>\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca\u0d9c\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca\u0d9c\u0dad \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0d9a\u0dca \u0dc3\u0dc4 \u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0d9a\u0dd2\u0db1\u0dca \u0dc3\u0db8\u0db1\u0dca\u0dc0\u0dd2\u0dad \u0dc0\u0dda. \u0dc4\u0dd0\u0d9a\u0dd2\u0dbd\u0dd3\u0db8\u0dda \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dba\u0db1\u0dd4 \u0dc3\u0d82\u0dc0\u0dc4\u0db1 \u0dc3\u0dca\u0dae\u0dbb \u0dc3\u0dc4 \u0dad\u0da7\u0dcf\u0d9a \u0dc3\u0dca\u0dae\u0dbb \u0db8\u0dcf\u0dbd\u0dcf\u0dc0\u0d9a\u0dca \u0dc0\u0db1 \u0d85\u0dad\u0dbb \u0d91\u0dc4\u0dd2\u0daf\u0dd3 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0dda \u0dc0\u0dd2\u0db7\u0dda\u0daf\u0db1\u0dba \u0d9a\u0dca\u0dbb\u0db8\u0dba\u0dd9\u0db1\u0dca \u0d85\u0da9\u0dd4 \u0dc0\u0dda. \u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dbd\u0d9a\u0dca\u0dc2\u0dab\u0dba \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0dba\u0ddd\u0da2\u0db1\u0dcf\u0dc0 \u0d9a\u0dca\u0dbb\u0db8\u0dba\u0dd9\u0db1\u0dca \u0dc0\u0dd0\u0da9\u0dd2 \u0dbd\u0d9a\u0dca\u0dc0\u0dd9\u0dba\u0dd2 \u0d91\u0dc4\u0dd2\u0daf\u0dd3 \u0daf\u0d9a\u0dca\u0dc0\u0dcf-\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd3\u0db8\u0dca \u0dc3\u0dca\u0dae\u0dbb \u0dc3\u0dc4 \u0dc3\u0d82\u0dc0\u0dbd\u0dd2\u0dad \u0dc3\u0dca\u0dae\u0dbb \u0db8\u0dcf\u0dbd\u0dcf\u0dc0\u0d9a\u0dca \u0dc0\u0dda. </p>\n<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0db8\u0dcf\u0dc0\u0dad\u0dda \u0dc3\u0dd1\u0db8 \u0db4\u0dd2\u0dba\u0dc0\u0dbb\u0d9a\u0daf\u0dd3\u0db8 \u0dc0\u0dbb\u0dca\u0dad\u0db8\u0dcf\u0db1 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0da7 \u0d85\u0db1\u0dd4\u0d9a\u0dd6\u0dbd \u0d9c\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca\u0d9c\u0dad \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dd9\u0db1\u0dca \u0d85\u0db1\u0dd4\u0dbb\u0dd6\u0db4 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8. </p>\n<p><span translate=no>_^_0_^_</span></p>\n<p><a href=\"carvana.html\">Carvana \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba \u0db8\u0dad U-Net</a> <a href=\"experiment.html\">\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0db1 \u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8\u0d9a\u0dca \u0dc3\u0db3\u0dc4\u0dcf \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dda\u0dad\u0dba</a> \u0db8\u0dd9\u0db1\u0dca\u0db1. </p>\n",
|
||||
"<h2>U-Net</h2>\n": "<h2>\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca</h2>\n",
|
||||
"<h3>Crop and Concatenate the feature map</h3>\n<p>At every step in the expansive path the corresponding feature map from the contracting path concatenated with the current feature map.</p>\n": "<h3>\u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c\u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0db6\u0ddd\u0d9c \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0dc3\u0d82\u0dba\u0dd4\u0d9a\u0dca\u0dad \u0d9a\u0dd2\u0dbb\u0dd3\u0db8</h3>\n<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0db8\u0dcf\u0dc0\u0dad\u0dda \u0dc3\u0dd1\u0db8 \u0db4\u0dd2\u0dba\u0dc0\u0dbb\u0d9a\u0daf\u0dd3\u0db8 \u0dc0\u0dbb\u0dca\u0dad\u0db8\u0dcf\u0db1 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0da7 \u0d85\u0db1\u0dd4\u0d9a\u0dd6\u0dbd \u0d9c\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca\u0d9c\u0dad \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dd9\u0db1\u0dca \u0d85\u0db1\u0dd4\u0dbb\u0dd6\u0db4 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8. </p>\n",
|
||||
"<h3>Down-sample</h3>\n<p>Each step in the contracting path down-samples the feature map with a <span translate=no>_^_0_^_</span> max pooling layer.</p>\n": "<h3>\u0db4\u0dc4\u0dc5-\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba</h3>\n<p>\u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dda \u0dc3\u0dd1\u0db8 \u0db4\u0dd2\u0dba\u0dc0\u0dbb\u0d9a\u0dca\u0db8 <span translate=no>_^_0_^_</span> \u0d8b\u0db4\u0dbb\u0dd2\u0db8 \u0dad\u0da7\u0dcf\u0d9a \u0dad\u0da7\u0dca\u0da7\u0dd4\u0dc0\u0d9a\u0dca \u0dc3\u0dc4\u0dd2\u0dad \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0db4\u0dc4\u0dc5\u0da7 \u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd \u0d9a\u0dbb\u0dba\u0dd2. </p>\n",
|
||||
"<h3>Two <span translate=no>_^_0_^_</span> Convolution Layers</h3>\n<p>Each step in the contraction path and expansive path have two <span translate=no>_^_1_^_</span> convolutional layers followed by ReLU activations.</p>\n<p>In the U-Net paper they used <span translate=no>_^_2_^_</span> padding, but we use <span translate=no>_^_3_^_</span> padding so that final feature map is not cropped.</p>\n": "<h3><span translate=no>_^_0_^_</span> \u0dc3\u0d82\u0dc0\u0dbd\u0dd2\u0dad \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0dd9\u0d9a\u0d9a\u0dca</h3>\n<p>\u0dc3\u0d82\u0d9a\u0ddd\u0da0\u0db1\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc4\u0dcf \u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0d91\u0d9a\u0dca \u0d91\u0d9a\u0dca \u0db4\u0dd2\u0dba\u0dc0\u0dbb RelU \u0dc3\u0d9a\u0dca\u0dbb\u0dd2\u0dba \u0dc0\u0dd2\u0dc3\u0dd2\u0db1\u0dca \u0d85\u0db1\u0dd4\u0d9c\u0db8\u0db1\u0dba <span translate=no>_^_1_^_</span> convolutional \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0dd9\u0d9a\u0d9a\u0dca \u0d87\u0dad. </p>\n<p>\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca\u0d9a\u0da9\u0daf\u0dcf\u0dc3\u0dd2 \u0dc0\u0dbd\u0daf\u0dd3 \u0d94\u0dc0\u0dd4\u0db1\u0dca <span translate=no>_^_2_^_</span> \u0db4\u0dd1\u0da9\u0dd2\u0db1\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dc5 \u0db1\u0db8\u0dd4\u0dad\u0dca \u0d85\u0dc0\u0dc3\u0dcf\u0db1 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0d9a\u0dd0\u0db4\u0dd6 \u0db1\u0ddc\u0dc0\u0db1 \u0db4\u0dbb\u0dd2\u0daf\u0dd2 \u0d85\u0db4\u0dd2 <span translate=no>_^_3_^_</span> \u0db4\u0dd1\u0da9\u0dd2\u0db1\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db8\u0dd4. </p>\n",
|
||||
"<h3>Up-sample</h3>\n<p>Each step in the expansive path up-samples the feature map with a <span translate=no>_^_0_^_</span> up-convolution.</p>\n": "<h3>\u0d89\u0dc4\u0dc5\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba</h3>\n<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db8\u0dcf\u0dc0\u0dad\u0dda \u0dc3\u0dd1\u0db8 \u0db4\u0dd2\u0dba\u0dc0\u0dbb\u0d9a\u0dca\u0db8 \u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd \u0db8\u0d9f\u0dd2\u0db1\u0dca \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 <span translate=no>_^_0_^_</span> \u0daf\u0d9a\u0dca\u0dc0\u0dcf-\u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dd4\u0db8\u0d9a\u0dd2\u0db1\u0dca \u0dba\u0dd4\u0d9a\u0dca\u0dad \u0dc0\u0dda. </p>\n",
|
||||
"<p> </p>\n": "<p> </p>\n",
|
||||
"<p>Apply the two convolution layers and activations </p>\n": "<p>\u0d9a\u0dd0\u0da7\u0dd2\u0d9c\u0dd0\u0dc3\u0dd4\u0dab\u0dd4 \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0dd9\u0d9a \u0dc3\u0dc4 \u0dc3\u0d9a\u0dca\u0dbb\u0dd2\u0dba \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dca \u0dba\u0ddc\u0daf\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Collect the output </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Concatenate the feature maps </p>\n": "<p>\u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c\u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0dca \u0dc3\u0d82\u0dba\u0dd4\u0d9a\u0dca\u0dad \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Concatenate the output of the contracting path </p>\n": "<p>\u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dda \u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba \u0dc3\u0d82\u0dba\u0dd4\u0d9a\u0dca\u0dad \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Contracting path </p>\n": "<p>\u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba </p>\n",
|
||||
"<p>Crop and concatenate layers for the expansive path. </p>\n": "<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0db6\u0ddd\u0d9c \u0dc4\u0dcf concatenate \u0dc3\u0dca\u0dae\u0dbb. </p>\n",
|
||||
"<p>Crop the feature map from the contracting path to the size of the current feature map </p>\n": "<p>\u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dda \u0dc3\u0dd2\u0da7 \u0dc0\u0dbb\u0dca\u0dad\u0db8\u0dcf\u0db1 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0dda \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba \u0daf\u0d9a\u0dca\u0dc0\u0dcf \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0dc0\u0d9c\u0dcf \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Double convolution layers for the contracting path. The number of features gets doubled at each step starting from <span translate=no>_^_0_^_</span>. </p>\n": "<p>\u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0daf\u0dca\u0dc0\u0dd2\u0dad\u0dca\u0dc0 \u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dd3\u0db8\u0dda \u0dc3\u0dca\u0dae\u0dbb. \u0dc3\u0dd2\u0da7 \u0d86\u0dbb\u0db8\u0dca\u0db7 \u0dc0\u0db1 \u0dc3\u0dd1\u0db8 \u0db4\u0dd2\u0dba\u0dc0\u0dbb\u0d9a\u0daf\u0dd3\u0db8 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0d9c\u0dab\u0db1 \u0daf\u0dd9\u0d9c\u0dd4\u0dab <span translate=no>_^_0_^_</span>\u0dc0\u0dda. </p>\n",
|
||||
"<p>Double convolution layers for the expansive path. Their input is the concatenation of the current feature map and the feature map from the contracting path. Therefore, the number of input features is double the number of features from up-sampling. </p>\n": "<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0daf\u0dca\u0dc0\u0dd2\u0dad\u0dca\u0dc0 \u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dd4\u0dab\u0dd4 \u0dc3\u0dca\u0dae\u0dbb. \u0d94\u0dc0\u0dd4\u0db1\u0dca\u0d9c\u0dda \u0d86\u0daf\u0dcf\u0db1\u0dba \u0dc0\u0db1\u0dca\u0db1\u0dda \u0dc0\u0dbb\u0dca\u0dad\u0db8\u0dcf\u0db1 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0dc3\u0dc4 \u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dd9\u0db1\u0dca \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 \u0dc3\u0d82\u0dba\u0dd4\u0d9a\u0dca\u0dad \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dba\u0dd2. \u0d91\u0db6\u0dd0\u0dc0\u0dd2\u0db1\u0dca \u0d86\u0daf\u0dcf\u0db1 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0d9c\u0dab\u0db1 \u0d89\u0dc4\u0dc5 \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd3\u0db8\u0dca \u0dc0\u0dbd\u0dd2\u0db1\u0dca \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0d9c\u0dab\u0db1 \u0daf\u0dd9\u0d9c\u0dd4\u0dab\u0dba\u0d9a\u0dca \u0dc0\u0dda. </p>\n",
|
||||
"<p>Down sampling layers for the contracting path </p>\n": "<p>\u0d9a\u0ddc\u0db1\u0dca\u0dad\u0dca\u0dbb\u0dcf\u0dad\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2 \u0dc3\u0dca\u0dae\u0dbb \u0db4\u0dc4\u0dc5\u0da7 </p>\n",
|
||||
"<p>Down-sample </p>\n": "<p>\u0db4\u0dc4\u0dc5-\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba </p>\n",
|
||||
"<p>Expansive path </p>\n": "<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba </p>\n",
|
||||
"<p>Final <span translate=no>_^_0_^_</span> convolution layer </p>\n": "<p>\u0d85\u0dc0\u0dc3\u0dcf\u0db1 <span translate=no>_^_0_^_</span> \u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dd4\u0dab\u0dd4 \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
|
||||
"<p>Final <span translate=no>_^_0_^_</span> convolution layer to produce the output </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dd2\u0dc2\u0dca\u0db4\u0dcf\u0daf\u0db1\u0dba <span translate=no>_^_0_^_</span> \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d85\u0dc0\u0dc3\u0dcf\u0db1 \u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dd4\u0dab\u0dd4 \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
|
||||
"<p>First <span translate=no>_^_0_^_</span> convolutional layer </p>\n": "<p>\u0db4\u0dc5\u0db8\u0dd4 <span translate=no>_^_0_^_</span> \u0dc3\u0d82\u0dc0\u0dd2\u0da0\u0dca\u0da1\u0dd4\u0dcf \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
|
||||
"<p>Max pooling layer </p>\n": "<p>\u0d8b\u0db4\u0dbb\u0dd2\u0db8\u0dad\u0da7\u0dcf\u0d9a \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
|
||||
"<p>Second <span translate=no>_^_0_^_</span> convolutional layer </p>\n": "<p>\u0daf\u0dd9\u0dc0\u0db1 <span translate=no>_^_0_^_</span> \u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dd4\u0dab\u0dd4 \u0dc3\u0dca\u0dae\u0dbb\u0dba </p>\n",
|
||||
"<p>The two convolution layers at the lowest resolution (the bottom of the U). </p>\n": "<p>\u0d85\u0dc0\u0db8\u0dc0\u0dd2\u0db7\u0dda\u0daf\u0db1\u0dba\u0dda (U \u0db4\u0dad\u0dd4\u0dbd\u0dda) \u0d9a\u0dd0\u0da7\u0dd2 \u0d9c\u0dd0\u0dc3\u0dca\u0dc0\u0dd3\u0db8\u0dda \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0dd9\u0d9a. </p>\n",
|
||||
"<p>To collect the outputs of contracting path for later concatenation with the expansive path. </p>\n": "<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc3\u0db8\u0d9f \u0db4\u0dc3\u0dd4\u0d9a\u0dcf\u0dbd\u0dd3\u0db1 \u0dc3\u0d82\u0d9a\u0ddd\u0da0\u0db1\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0dc3\u0d82\u0d9a\u0ddd\u0da0\u0db1 \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dda \u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1\u0dba\u0db1\u0dca \u0d91\u0d9a\u0dad\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8. </p>\n",
|
||||
"<p>Two <span translate=no>_^_0_^_</span> convolutional layers </p>\n": "<p><span translate=no>_^_0_^_</span> \u0dc3\u0d82\u0d9a\u0dd3\u0dbb\u0dca\u0dab \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0dd9\u0d9a\u0d9a\u0dca </p>\n",
|
||||
"<p>Two <span translate=no>_^_0_^_</span> convolutional layers at the bottom of the U-Net </p>\n": "<p>\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca\u0db4\u0dad\u0dd4\u0dbd\u0dda <span translate=no>_^_0_^_</span> \u0dc3\u0d82\u0dba\u0dd4\u0d9a\u0dca\u0dad \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0dd9\u0d9a\u0d9a\u0dca </p>\n",
|
||||
"<p>Up sampling layers for the expansive path. The number of features is halved with up-sampling. </p>\n": "<p>\u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca\u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2 \u0dc3\u0dca\u0dae\u0dbb \u0daf\u0d9a\u0dca\u0dc0\u0dcf. \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0d9c\u0dab\u0db1 \u0d89\u0dc4\u0dc5\u0da7 \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd3\u0db8\u0dd9\u0db1\u0dca \u0d85\u0da9\u0d9a\u0dd2\u0db1\u0dca \u0d85\u0da9\u0dd4 \u0dc0\u0dda. </p>\n",
|
||||
"<p>Up-convolution </p>\n": "<p>\u0daf\u0d9a\u0dca\u0dc0\u0dcf-\u0dc3\u0d82\u0dc0\u0dc4\u0db1 </p>\n",
|
||||
"<p>Up-sample </p>\n": "<p>\u0d89\u0dc4\u0dc5\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba </p>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> current feature map in the expansive path </li>\n<li><span translate=no>_^_1_^_</span> corresponding feature map from the contracting path</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0db4\u0dd4\u0dc5\u0dd4\u0dbd\u0dca \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0db8\u0dcf\u0dc0\u0dad\u0dda \u0dc0\u0dad\u0dca\u0db8\u0db1\u0dca \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8 </li>\n<li><span translate=no>_^_1_^_</span> \u0d9c\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca\u0d9c\u0dad \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dd9\u0db1\u0dca \u0d85\u0db1\u0dd4\u0dbb\u0dd6\u0db4 \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> input image</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0d86\u0daf\u0dcf\u0db1 \u0dbb\u0dd6\u0db4\u0dba</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> is the number of input channels </li>\n<li><span translate=no>_^_1_^_</span> is the number of output channels</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0d86\u0daf\u0dcf\u0db1 \u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1 </li>\n<li><span translate=no>_^_1_^_</span> \u0db4\u0dca\u0dbb\u0dad\u0dd2\u0daf\u0dcf\u0db1 \u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1 \u0dc0\u0dda</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> number of channels in the input image </li>\n<li><span translate=no>_^_1_^_</span> number of channels in the result feature map</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0d86\u0daf\u0dcf\u0db1 \u0dbb\u0dd6\u0db4\u0dba\u0dda \u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1 </li>\n<li><span translate=no>_^_1_^_</span> \u0db4\u0dca\u0dbb\u0dad\u0dd2 result \u0dbd \u0dc0\u0dd2\u0dc1\u0dda\u0dc2\u0dcf\u0d82\u0d9c \u0dc3\u0dd2\u0dad\u0dd2\u0dba\u0db8\u0dda \u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1</li></ul>\n",
|
||||
"PyTorch implementation and tutorial of U-Net model.": "PyTorch U-Net \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0db1\u0dd2\u0db6\u0db1\u0dca\u0db0\u0db1\u0dba.",
|
||||
"U-Net": "\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca"
|
||||
}
|
||||
@@ -0,0 +1,39 @@
|
||||
{
|
||||
"<h1>U-Net</h1>\n<p>This is an implementation of the U-Net model from the paper, <a href=\"https://arxiv.org/abs/1505.04597\">U-Net: Convolutional Networks for Biomedical Image Segmentation</a>.</p>\n<p>U-Net consists of a contracting path and an expansive path. The contracting path is a series of convolutional layers and pooling layers, where the resolution of the feature map gets progressively reduced. Expansive path is a series of up-sampling layers and convolutional layers where the resolution of the feature map gets progressively increased.</p>\n<p>At every step in the expansive path the corresponding feature map from the contracting path concatenated with the current feature map.</p>\n<p><span translate=no>_^_0_^_</span></p>\n<p>Here is the <a href=\"experiment.html\">training code</a> for an experiment that trains a U-Net on <a href=\"carvana.html\">Carvana dataset</a>.</p>\n": "<h1>U-Net</h1>\n<p>\u8fd9\u662f\u8bba\u6587\u300aU-N <a href=\"https://arxiv.org/abs/1505.04597\">et\uff1a\u751f\u7269\u533b\u5b66\u56fe\u50cf\u5206\u5272\u7684\u5377\u79ef\u7f51\u7edc\u300b\u4e2dU-Net\u6a21\u578b\u7684\u5b9e\u73b0</a>\u3002</p>\n<p>U-Net \u7531\u4e00\u6761\u6536\u7f29\u8def\u5f84\u548c\u4e00\u6761\u6269\u5c55\u8def\u5f84\u7ec4\u6210\u3002\u6536\u7f29\u8def\u5f84\u662f\u4e00\u7cfb\u5217\u5377\u79ef\u56fe\u5c42\u548c\u6c60\u5316\u56fe\u5c42\uff0c\u5176\u4e2d\u8981\u7d20\u5730\u56fe\u7684\u5206\u8fa8\u7387\u4f1a\u9010\u6e10\u964d\u4f4e\u3002\u6269\u5c55\u8def\u5f84\u662f\u4e00\u7cfb\u5217\u5411\u4e0a\u91c7\u6837\u56fe\u5c42\u548c\u5377\u79ef\u56fe\u5c42\uff0c\u5176\u4e2d\u8981\u7d20\u5730\u56fe\u7684\u5206\u8fa8\u7387\u4f1a\u9010\u6e10\u63d0\u9ad8\u3002</p>\n<p>\u5728\u6269\u5f20\u8def\u5f84\u7684\u6bcf\u4e00\u6b65\u4e2d\uff0c\u6536\u7f29\u8def\u5f84\u4e2d\u7684\u76f8\u5e94\u8981\u7d20\u5730\u56fe\u90fd\u4e0e\u5f53\u524d\u8981\u7d20\u5730\u56fe\u76f8\u8fde\u3002</p>\n<p><span translate=no>_^_0_^_</span></p>\n<p>\u4ee5\u4e0b\u662f\u5728 <a href=\"carvana.html\">Carvana \u6570\u636e\u96c6</a>\u4e0a<a href=\"experiment.html\">\u8bad\u7ec3 U-Net \u7684\u5b9e\u9a8c\u7684\u8bad\u7ec3\u4ee3\u7801</a>\u3002</p>\n",
|
||||
"<h2>U-Net</h2>\n": "<h2>U-Net</h2>\n",
|
||||
"<h3>Crop and Concatenate the feature map</h3>\n<p>At every step in the expansive path the corresponding feature map from the contracting path concatenated with the current feature map.</p>\n": "<h3>\u88c1\u526a\u5e76\u8fde\u63a5\u8981\u7d20\u5730\u56fe</h3>\n<p>\u5728\u6269\u5f20\u8def\u5f84\u7684\u6bcf\u4e00\u6b65\u4e2d\uff0c\u6536\u7f29\u8def\u5f84\u4e2d\u7684\u76f8\u5e94\u8981\u7d20\u5730\u56fe\u90fd\u4e0e\u5f53\u524d\u8981\u7d20\u5730\u56fe\u76f8\u8fde\u3002</p>\n",
|
||||
"<h3>Down-sample</h3>\n<p>Each step in the contracting path down-samples the feature map with a <span translate=no>_^_0_^_</span> max pooling layer.</p>\n": "<h3>\u5411\u4e0b\u91c7\u6837</h3>\n<p>\u6536\u7f29\u8def\u5f84\u4e2d\u7684\u6bcf\u4e2a\u6b65\u9aa4\u90fd\u4f1a\u4f7f\u7528<span translate=no>_^_0_^_</span>\u6700\u5927\u6c60\u5316\u56fe\u5c42\u5bf9\u8981\u7d20\u5730\u56fe\u8fdb\u884c\u7f29\u51cf\u91c7\u6837\u3002</p>\n",
|
||||
"<h3>Two <span translate=no>_^_0_^_</span> Convolution Layers</h3>\n<p>Each step in the contraction path and expansive path have two <span translate=no>_^_1_^_</span> convolutional layers followed by ReLU activations.</p>\n<p>In the U-Net paper they used <span translate=no>_^_2_^_</span> padding, but we use <span translate=no>_^_3_^_</span> padding so that final feature map is not cropped.</p>\n": "<h3>\u4e24\u4e2a<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</h3>\n<p>\u6536\u7f29\u8def\u5f84\u548c\u6269\u5f20\u8def\u5f84\u4e2d\u7684\u6bcf\u4e00\u6b65\u90fd\u6709\u4e24\u4e2a<span translate=no>_^_1_^_</span>\u5377\u79ef\u5c42\uff0c\u7136\u540e\u662f RelU \u6fc0\u6d3b\u3002</p>\n<p>\u5728 U-Net \u8bba\u6587\u4e2d\uff0c\u4ed6\u4eec\u4f7f\u7528<span translate=no>_^_2_^_</span>\u586b\u5145\uff0c\u4f46\u6211\u4eec\u4f7f\u7528<span translate=no>_^_3_^_</span>\u586b\u5145\uff0c\u8fd9\u6837\u6700\u7ec8\u7684\u7279\u5f81\u56fe\u5c31\u4e0d\u4f1a\u88ab\u88c1\u526a\u3002</p>\n",
|
||||
"<h3>Up-sample</h3>\n<p>Each step in the expansive path up-samples the feature map with a <span translate=no>_^_0_^_</span> up-convolution.</p>\n": "<h3>\u5411\u4e0a\u91c7\u6837</h3>\n<p>\u6269\u5c55\u8def\u5f84\u4e2d\u7684\u6bcf\u4e00\u6b65\u90fd\u4f7f\u7528\u5411\u4e0a<span translate=no>_^_0_^_</span>\u5377\u79ef\u5bf9\u8981\u7d20\u5730\u56fe\u8fdb\u884c\u5411\u4e0a\u91c7\u6837\u3002</p>\n",
|
||||
"<p> </p>\n": "<p></p>\n",
|
||||
"<p>Apply the two convolution layers and activations </p>\n": "<p>\u5e94\u7528\u4e24\u4e2a\u5377\u79ef\u5c42\u548c\u6fc0\u6d3b</p>\n",
|
||||
"<p>Collect the output </p>\n": "<p>\u6536\u96c6\u8f93\u51fa</p>\n",
|
||||
"<p>Concatenate the feature maps </p>\n": "<p>\u8fde\u63a5\u8981\u7d20\u6620\u5c04</p>\n",
|
||||
"<p>Concatenate the output of the contracting path </p>\n": "<p>\u8fde\u63a5\u6536\u7f29\u8def\u5f84\u7684\u8f93\u51fa</p>\n",
|
||||
"<p>Contracting path </p>\n": "<p>\u6536\u7f29\u8def\u5f84</p>\n",
|
||||
"<p>Crop and concatenate layers for the expansive path. </p>\n": "<p>\u88c1\u526a\u548c\u8fde\u63a5\u6269\u5c55\u8def\u5f84\u7684\u56fe\u5c42\u3002</p>\n",
|
||||
"<p>Crop the feature map from the contracting path to the size of the current feature map </p>\n": "<p>\u5c06\u8981\u7d20\u5730\u56fe\u4ece\u6536\u7f29\u8def\u5f84\u88c1\u526a\u4e3a\u5f53\u524d\u8981\u7d20\u5730\u56fe\u7684\u5927\u5c0f</p>\n",
|
||||
"<p>Double convolution layers for the contracting path. The number of features gets doubled at each step starting from <span translate=no>_^_0_^_</span>. </p>\n": "<p>\u6536\u7f29\u8def\u5f84\u7684\u53cc\u5377\u79ef\u5c42\u3002\u4ece\u5f00\u59cb\uff0c\u6bcf\u4e00\u6b65\u7684\u529f\u80fd\u6570\u91cf\u90fd\u4f1a\u589e\u52a0\u4e00\u500d<span translate=no>_^_0_^_</span>\u3002</p>\n",
|
||||
"<p>Double convolution layers for the expansive path. Their input is the concatenation of the current feature map and the feature map from the contracting path. Therefore, the number of input features is double the number of features from up-sampling. </p>\n": "<p>\u6269\u5c55\u8def\u5f84\u7684\u53cc\u5377\u79ef\u5c42\u3002\u5b83\u4eec\u7684\u8f93\u5165\u662f\u5f53\u524d\u8981\u7d20\u5730\u56fe\u548c\u6536\u7f29\u8def\u5f84\u4e2d\u7684\u8981\u7d20\u5730\u56fe\u7684\u4e32\u8054\u3002\u56e0\u6b64\uff0c\u8f93\u5165\u8981\u7d20\u7684\u6570\u91cf\u662f\u5411\u4e0a\u91c7\u6837\u7684\u8981\u7d20\u6570\u91cf\u7684\u4e24\u500d\u3002</p>\n",
|
||||
"<p>Down sampling layers for the contracting path </p>\n": "<p>\u5411\u4e0b\u91c7\u6837\u6536\u7f29\u8def\u5f84\u7684\u56fe\u5c42</p>\n",
|
||||
"<p>Down-sample </p>\n": "<p>\u5411\u4e0b\u91c7\u6837</p>\n",
|
||||
"<p>Expansive path </p>\n": "<p>\u5e7f\u9614\u7684\u9053\u8def</p>\n",
|
||||
"<p>Final <span translate=no>_^_0_^_</span> convolution layer </p>\n": "<p>\u6700\u7ec8<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</p>\n",
|
||||
"<p>Final <span translate=no>_^_0_^_</span> convolution layer to produce the output </p>\n": "<p>\u751f\u6210\u8f93\u51fa\u7684\u6700\u7ec8<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</p>\n",
|
||||
"<p>First <span translate=no>_^_0_^_</span> convolutional layer </p>\n": "<p>\u7b2c\u4e00\u4e2a<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</p>\n",
|
||||
"<p>Max pooling layer </p>\n": "<p>\u6700\u5927\u6c60\u5316\u5c42</p>\n",
|
||||
"<p>Second <span translate=no>_^_0_^_</span> convolutional layer </p>\n": "<p>\u7b2c\u4e8c\u4e2a<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</p>\n",
|
||||
"<p>The two convolution layers at the lowest resolution (the bottom of the U). </p>\n": "<p>\u5206\u8fa8\u7387\u6700\u4f4e\u7684\u4e24\u4e2a\u5377\u79ef\u5c42\uff08U \u7684\u5e95\u90e8\uff09\u3002</p>\n",
|
||||
"<p>To collect the outputs of contracting path for later concatenation with the expansive path. </p>\n": "<p>\u6536\u96c6\u6536\u7f29\u8def\u5f84\u7684\u8f93\u51fa\uff0c\u4ee5\u4fbf\u7a0d\u540e\u4e0e\u6269\u5c55\u8def\u5f84\u4e32\u8054\u3002</p>\n",
|
||||
"<p>Two <span translate=no>_^_0_^_</span> convolutional layers </p>\n": "<p>\u4e24\u4e2a<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</p>\n",
|
||||
"<p>Two <span translate=no>_^_0_^_</span> convolutional layers at the bottom of the U-Net </p>\n": "<p>U-Net \u5e95\u90e8\u6709\u4e24\u4e2a<span translate=no>_^_0_^_</span>\u5377\u79ef\u5c42</p>\n",
|
||||
"<p>Up sampling layers for the expansive path. The number of features is halved with up-sampling. </p>\n": "<p>\u5411\u4e0a\u91c7\u6837\u6269\u5c55\u8def\u5f84\u7684\u56fe\u5c42\u3002\u901a\u8fc7\u5411\u4e0a\u91c7\u6837\uff0c\u8981\u7d20\u6570\u91cf\u51cf\u534a\u3002</p>\n",
|
||||
"<p>Up-convolution </p>\n": "<p>\u5411\u4e0a\u5377\u79ef</p>\n",
|
||||
"<p>Up-sample </p>\n": "<p>\u5411\u4e0a\u91c7\u6837</p>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> current feature map in the expansive path </li>\n<li><span translate=no>_^_1_^_</span> corresponding feature map from the contracting path</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u6269\u5c55\u8def\u5f84\u4e2d\u7684\u5f53\u524d\u8981\u7d20\u5730\u56fe</li>\n<li><span translate=no>_^_1_^_</span>\u6536\u7f29\u8def\u5f84\u4e2d\u7684\u76f8\u5e94\u8981\u7d20\u5730\u56fe</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> input image</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u8f93\u5165\u56fe\u50cf</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> is the number of input channels </li>\n<li><span translate=no>_^_1_^_</span> is the number of output channels</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u662f\u8f93\u5165\u58f0\u9053\u7684\u6570\u91cf</li>\n<li><span translate=no>_^_1_^_</span>\u662f\u8f93\u51fa\u58f0\u9053\u7684\u6570\u91cf</li></ul>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> number of channels in the input image </li>\n<li><span translate=no>_^_1_^_</span> number of channels in the result feature map</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u8f93\u5165\u56fe\u50cf\u4e2d\u7684\u901a\u9053\u6570</li>\n<li><span translate=no>_^_1_^_</span>\u7ed3\u679c\u7279\u5f81\u56fe\u4e2d\u7684\u4fe1\u9053\u6570</li></ul>\n",
|
||||
"PyTorch implementation and tutorial of U-Net model.": "PyTorch \u5b9e\u73b0\u548c U-Net \u6a21\u578b\u6559\u7a0b\u3002",
|
||||
"U-Net": "U-Net"
|
||||
}
|
||||
@@ -0,0 +1,21 @@
|
||||
{
|
||||
"<h1>Carvana Dataset for the <a href=\"index.html\">U-Net</a> <a href=\"experiment.html\">experiment</a></h1>\n<p>You can find the download instructions <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">on Kaggle</a>.</p>\n<p>Save the training images inside <span translate=no>_^_0_^_</span> folder and the masks in <span translate=no>_^_1_^_</span> folder.</p>\n": "<h1><a href=\"index.html\"><a href=\"experiment.html\">U-Net\u5b9f\u9a13\u7528\u306e\u30ab\u30eb\u30d0\u30ca\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</a></a></h1>\n<p>\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u624b\u9806\u306f <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">Kaggle \u3067\u78ba\u8a8d\u3067\u304d\u307e\u3059</a>\u3002</p>\n<p><span translate=no>_^_0_^_</span>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u753b\u50cf\u3092\u30d5\u30a9\u30eb\u30c0\u30fc\u5185\u306b\u4fdd\u5b58\u3057\u3001<span translate=no>_^_1_^_</span>\u30de\u30b9\u30af\u3092\u30d5\u30a9\u30eb\u30c0\u30fc\u306b\u4fdd\u5b58\u3057\u307e\u3059\u3002</p>\n",
|
||||
"<h2>Carvana Dataset</h2>\n": "<h2>\u30ab\u30fc\u30d0\u30ca\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</h2>\n",
|
||||
"<h4>Get an image and its mask.</h4>\n<ul><li><span translate=no>_^_0_^_</span> is index of the image</li></ul>\n": "<h4>\u753b\u50cf\u3068\u305d\u306e\u30de\u30b9\u30af\u3092\u5165\u624b\u3057\u3066\u304f\u3060\u3055\u3044\u3002</h4>\n<ul><li><span translate=no>_^_0_^_</span>\u306f\u753b\u50cf\u306e\u30a4\u30f3\u30c7\u30c3\u30af\u30b9\u3067\u3059</li></ul>\n",
|
||||
"<h4>Size of the dataset</h4>\n": "<h4>\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306e\u30b5\u30a4\u30ba</h4>\n",
|
||||
"<p>Get a dictionary of images by id </p>\n": "<p>ID \u3067\u753b\u50cf\u306e\u8f9e\u66f8\u3092\u53d6\u5f97</p>\n",
|
||||
"<p>Get a dictionary of masks by id </p>\n": "<p>ID \u3067\u30de\u30b9\u30af\u306e\u8f9e\u66f8\u3092\u53d6\u5f97</p>\n",
|
||||
"<p>Get image id </p>\n": "<p>\u753b\u50cf ID \u3092\u53d6\u5f97</p>\n",
|
||||
"<p>Image ids list </p>\n": "<p>\u753b\u50cfID\u30ea\u30b9\u30c8</p>\n",
|
||||
"<p>Load image </p>\n": "<p>\u753b\u50cf\u3092\u8aad\u307f\u8fbc\u3080</p>\n",
|
||||
"<p>Load mask </p>\n": "<p>\u30ed\u30fc\u30c9\u30de\u30b9\u30af</p>\n",
|
||||
"<p>Return the image and the mask </p>\n": "<p>\u753b\u50cf\u3068\u30de\u30b9\u30af\u3092\u8fd4\u3059</p>\n",
|
||||
"<p>Testing code </p>\n": "<p>\u30c6\u30b9\u30c8\u30b3\u30fc\u30c9</p>\n",
|
||||
"<p>The mask values were not <span translate=no>_^_0_^_</span>, so we scale it appropriately. </p>\n": "<p>\u30de\u30b9\u30af\u5024\u306f\u306a\u304b\u3063\u305f\u306e\u3067<span translate=no>_^_0_^_</span>\u3001\u9069\u5207\u306b\u30b9\u30b1\u30fc\u30ea\u30f3\u30b0\u3057\u307e\u3057\u305f\u3002</p>\n",
|
||||
"<p>Transform image and convert it to a PyTorch tensor </p>\n": "<p>\u753b\u50cf\u3092\u5909\u63db\u3057\u3066 PyTorch \u30c6\u30f3\u30bd\u30eb\u306b\u5909\u63db\u3059\u308b</p>\n",
|
||||
"<p>Transform mask and convert it to a PyTorch tensor </p>\n": "<p>\u30de\u30b9\u30af\u3092\u5909\u63db\u3057\u3066 PyTorch \u30c6\u30f3\u30bd\u30eb\u306b\u5909\u63db\u3059\u308b</p>\n",
|
||||
"<p>Transformations </p>\n": "<p>\u30c8\u30e9\u30f3\u30b9\u30d5\u30a9\u30fc\u30e1\u30fc\u30b7\u30e7\u30f3</p>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> is the path to the images </li>\n<li><span translate=no>_^_1_^_</span> is the path to the masks</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u753b\u50cf\u3078\u306e\u30d1\u30b9\u3067\u3059</li>\n<li><span translate=no>_^_1_^_</span>\u30de\u30b9\u30af\u3078\u306e\u9053\u3067\u3059</li></ul>\n",
|
||||
"Carvana dataset for the U-Net experiment": "U-Net\u5b9f\u9a13\u7528\u306eCarvana\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8",
|
||||
"Carvana dataset for the U-Net experiment.": "U-Net\u5b9f\u9a13\u7528\u306eCarvana\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3002"
|
||||
}
|
||||
@@ -0,0 +1,21 @@
|
||||
{
|
||||
"<h1>Carvana Dataset for the <a href=\"index.html\">U-Net</a> <a href=\"experiment.html\">experiment</a></h1>\n<p>You can find the download instructions <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">on Kaggle</a>.</p>\n<p>Save the training images inside <span translate=no>_^_0_^_</span> folder and the masks in <span translate=no>_^_1_^_</span> folder.</p>\n": "<h1><a href=\"index.html\">\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca</a> <a href=\"experiment.html\">\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8</a> \u0dc3\u0db3\u0dc4\u0dcf \u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h1>\n<p>\u0d94\u0db6\u0da7\u0db6\u0dcf\u0d9c\u0dad \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d8b\u0db4\u0daf\u0dd9\u0dc3\u0dca <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">\u0dc3\u0ddc\u0dba\u0dcf\u0d9c\u0dad \u0dc4\u0dd0\u0d9a\u0dd2\u0dba Kaggle</a>. </p>\n<p><span translate=no>_^_0_^_</span> \u0dc6\u0ddd\u0dbd\u0dca\u0da9\u0dbb\u0dba \u0dad\u0dd4\u0dc5 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0db4\u0dd2\u0db1\u0dca\u0dad\u0dd6\u0dbb \u0dc3\u0dc4 <span translate=no>_^_1_^_</span> \u0dc6\u0ddd\u0dbd\u0dca\u0da9\u0dbb\u0dba\u0dda \u0dc0\u0dd9\u0dc3\u0dca \u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0dc3\u0dd4\u0dbb\u0d9a\u0dd2\u0db1\u0dca\u0db1. </p>\n",
|
||||
"<h2>Carvana Dataset</h2>\n": "<h2>\u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf\u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</h2>\n",
|
||||
"<h4>Get an image and its mask.</h4>\n<ul><li><span translate=no>_^_0_^_</span> is index of the image</li></ul>\n": "<h4>\u0dbb\u0dd6\u0db4\u0dba\u0d9a\u0dca\u0dc3\u0dc4 \u0d91\u0dc4\u0dd2 \u0dc0\u0dd9\u0dc3\u0dca \u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0d9a\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1. </h4>\n<ul><li><span translate=no>_^_0_^_</span> \u0dbb\u0dd6\u0db4\u0dba\u0dda \u0daf\u0dbb\u0dca\u0dc1\u0d9a\u0dba \u0dc0\u0dda</li></ul>\n",
|
||||
"<h4>Size of the dataset</h4>\n": "<h4>\u0daf\u0dad\u0dca\u0dad\u0dc3\u0db8\u0dd4\u0daf\u0dcf\u0dba \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba</h4>\n",
|
||||
"<p>Get a dictionary of images by id </p>\n": "<p>Id\u0db8\u0d9c\u0dd2\u0db1\u0dca \u0dbb\u0dd6\u0db4 \u0dc1\u0db6\u0dca\u0daf \u0d9a\u0ddd\u0dc2\u0dba\u0d9a\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Get a dictionary of masks by id </p>\n": "<p>\u0dc4\u0dd0\u0db3\u0dd4\u0db1\u0dd4\u0db8\u0dca\u0db4\u0dad\u0db8\u0d9c\u0dd2\u0db1\u0dca \u0dc0\u0dd9\u0dc3\u0dca \u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0dc1\u0db6\u0dca\u0daf \u0d9a\u0ddd\u0dc2\u0dba\u0d9a\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Get image id </p>\n": "<p>\u0dbb\u0dd6\u0db4\u0dc4\u0dd0\u0db3\u0dd4\u0db1\u0dd4\u0db8\u0dca\u0db4\u0dad \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Image ids list </p>\n": "<p>\u0dbb\u0dd6\u0db4IDS \u0dbd\u0dd0\u0dba\u0dd2\u0dc3\u0dca\u0dad\u0dd4\u0dc0 </p>\n",
|
||||
"<p>Load image </p>\n": "<p>\u0dbb\u0dd6\u0db4\u0dba\u0db4\u0da7\u0dc0\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Load mask </p>\n": "<p>\u0d86\u0dc0\u0dbb\u0dab\u0db4\u0da7\u0dc0\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Return the image and the mask </p>\n": "<p>\u0dbb\u0dd6\u0db4\u0dba\u0dc3\u0dc4 \u0dc0\u0dd9\u0dc3\u0dca\u0db8\u0dd4\u0dc4\u0dd4\u0dab \u0d86\u0db4\u0dc3\u0dd4 \u0dbd\u0db6\u0dcf \u0daf\u0dd9\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Testing code </p>\n": "<p>\u0db4\u0dbb\u0dd3\u0d9a\u0dca\u0dc2\u0dcf\u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d9a\u0dda\u0dad\u0dba </p>\n",
|
||||
"<p>The mask values were not <span translate=no>_^_0_^_</span>, so we scale it appropriately. </p>\n": "<p>\u0d86\u0dc0\u0dbb\u0dab\u0d85\u0d9c\u0dba\u0db1\u0dca \u0db1\u0ddc\u0dad\u0dd2\u0db6\u0dd4\u0dab\u0dd2 <span translate=no>_^_0_^_</span>, \u0d91\u0db6\u0dd0\u0dc0\u0dd2\u0db1\u0dca \u0d85\u0db4\u0dd2 \u0d91\u0dba \u0db1\u0dd2\u0dc3\u0dd2 \u0db4\u0dbb\u0dd2\u0daf\u0dd2 \u0db4\u0dbb\u0dd2\u0db8\u0dcf\u0dab\u0dba \u0d9a\u0dbb\u0db8\u0dd4. </p>\n",
|
||||
"<p>Transform image and convert it to a PyTorch tensor </p>\n": "<p>\u0dbb\u0dd6\u0db4\u0dba\u0db4\u0dbb\u0dd2\u0dc0\u0dbb\u0dca\u0dad\u0db1\u0dba \u0d9a\u0dbb \u0d91\u0dba \u0db4\u0dba\u0dd2\u0da7\u0ddd\u0da0\u0dca \u0da7\u0dd9\u0db1\u0dca\u0dc3\u0dbb\u0dba\u0d9a\u0dca \u0db6\u0dc0\u0da7 \u0db4\u0dbb\u0dd2\u0dc0\u0dbb\u0dca\u0dad\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Transform mask and convert it to a PyTorch tensor </p>\n": "<p>\u0dc0\u0dd9\u0dc3\u0dca\u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0db4\u0dbb\u0dd2\u0dc0\u0dbb\u0dca\u0dad\u0db1\u0dba \u0d9a\u0dbb \u0d91\u0dba \u0db4\u0dba\u0dd2\u0da7\u0ddd\u0da0\u0dca \u0da7\u0dd9\u0db1\u0dca\u0dc3\u0dbb\u0dba\u0d9a\u0dca \u0db6\u0dc0\u0da7 \u0db4\u0dbb\u0dd2\u0dc0\u0dbb\u0dca\u0dad\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Transformations </p>\n": "<p>\u0db4\u0dbb\u0dd2\u0dc0\u0dbb\u0dca\u0dad\u0db1\u0dba\u0db1\u0dca </p>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> is the path to the images </li>\n<li><span translate=no>_^_1_^_</span> is the path to the masks</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span> \u0dbb\u0dd6\u0db4 \u0dc3\u0db3\u0dc4\u0dcf \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba \u0dc0\u0dda </li>\n<li><span translate=no>_^_1_^_</span> \u0dc0\u0dd9\u0dc3\u0dca \u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0dc0\u0dd9\u0dad \u0dba\u0db1 \u0db8\u0dcf\u0dbb\u0dca\u0d9c\u0dba\u0dba\u0dd2</li></ul>\n",
|
||||
"Carvana dataset for the U-Net experiment": "\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca \u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba",
|
||||
"Carvana dataset for the U-Net experiment.": "\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca \u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf \u0db6\u0dd0\u0dbd\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba."
|
||||
}
|
||||
@@ -0,0 +1,21 @@
|
||||
{
|
||||
"<h1>Carvana Dataset for the <a href=\"index.html\">U-Net</a> <a href=\"experiment.html\">experiment</a></h1>\n<p>You can find the download instructions <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">on Kaggle</a>.</p>\n<p>Save the training images inside <span translate=no>_^_0_^_</span> folder and the masks in <span translate=no>_^_1_^_</span> folder.</p>\n": "<h1><a href=\"index.html\">U-Net</a> <a href=\"experiment.html\">\u5b9e\u9a8c</a>\u7684 Carvana \u6570\u636e\u96c6</h1>\n<p>\u4f60\u53ef\u4ee5<a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">\u5728 Kaggle \u4e0a</a>\u627e\u5230\u4e0b\u8f7d\u8bf4\u660e\u3002</p>\n<p>\u5c06\u8bad\u7ec3\u56fe\u50cf\u4fdd\u5b58\u5728<span translate=no>_^_0_^_</span>\u6587\u4ef6\u5939\u4e2d\uff0c\u5c06\u8499\u7248\u4fdd\u5b58\u5728<span translate=no>_^_1_^_</span>\u6587\u4ef6\u5939\u4e2d\u3002</p>\n",
|
||||
"<h2>Carvana Dataset</h2>\n": "<h2>Carvana \u6570\u636e\u96c6</h2>\n",
|
||||
"<h4>Get an image and its mask.</h4>\n<ul><li><span translate=no>_^_0_^_</span> is index of the image</li></ul>\n": "<h4>\u83b7\u53d6\u56fe\u50cf\u53ca\u5176\u8499\u7248\u3002</h4>\n<ul><li><span translate=no>_^_0_^_</span>\u662f\u56fe\u50cf\u7684\u7d22\u5f15</li></ul>\n",
|
||||
"<h4>Size of the dataset</h4>\n": "<h4>\u6570\u636e\u96c6\u7684\u5927\u5c0f</h4>\n",
|
||||
"<p>Get a dictionary of images by id </p>\n": "<p>\u6309 id \u83b7\u53d6\u56fe\u50cf\u8bcd\u5178</p>\n",
|
||||
"<p>Get a dictionary of masks by id </p>\n": "<p>\u901a\u8fc7 id \u83b7\u53d6\u53e3\u7f69\u5b57\u5178</p>\n",
|
||||
"<p>Get image id </p>\n": "<p>\u83b7\u53d6\u56fe\u7247\u7f16\u53f7</p>\n",
|
||||
"<p>Image ids list </p>\n": "<p>\u6620\u50cf ID \u5217\u8868</p>\n",
|
||||
"<p>Load image </p>\n": "<p>\u52a0\u8f7d\u56fe\u7247</p>\n",
|
||||
"<p>Load mask </p>\n": "<p>\u88c5\u8f7d\u63a9\u7801</p>\n",
|
||||
"<p>Return the image and the mask </p>\n": "<p>\u8fd4\u56de\u56fe\u50cf\u548c\u8499\u7248</p>\n",
|
||||
"<p>Testing code </p>\n": "<p>\u6d4b\u8bd5\u4ee3\u7801</p>\n",
|
||||
"<p>The mask values were not <span translate=no>_^_0_^_</span>, so we scale it appropriately. </p>\n": "<p>\u63a9\u7801\u503c\u4e0d\u662f<span translate=no>_^_0_^_</span>\uff0c\u56e0\u6b64\u6211\u4eec\u5bf9\u5176\u8fdb\u884c\u4e86\u9002\u5f53\u7684\u7f29\u653e\u3002</p>\n",
|
||||
"<p>Transform image and convert it to a PyTorch tensor </p>\n": "<p>\u53d8\u6362\u56fe\u50cf\u5e76\u5c06\u5176\u8f6c\u6362\u4e3a PyTorch \u5f20\u91cf</p>\n",
|
||||
"<p>Transform mask and convert it to a PyTorch tensor </p>\n": "<p>\u53d8\u6362\u63a9\u7801\u5e76\u5c06\u5176\u8f6c\u6362\u4e3a PyTorch \u5f20\u91cf</p>\n",
|
||||
"<p>Transformations </p>\n": "<p>\u8f6c\u6362</p>\n",
|
||||
"<ul><li><span translate=no>_^_0_^_</span> is the path to the images </li>\n<li><span translate=no>_^_1_^_</span> is the path to the masks</li></ul>\n": "<ul><li><span translate=no>_^_0_^_</span>\u662f\u56fe\u50cf\u7684\u8def\u5f84</li>\n<li><span translate=no>_^_1_^_</span>\u662f\u901a\u5f80\u53e3\u7f69\u7684\u8def\u5f84</li></ul>\n",
|
||||
"Carvana dataset for the U-Net experiment": "U-Net \u5b9e\u9a8c\u7684 Carvana \u6570\u636e\u96c6",
|
||||
"Carvana dataset for the U-Net experiment.": "U-Net \u5b9e\u9a8c\u7684 Carvana \u6570\u636e\u96c6\u3002"
|
||||
}
|
||||
@@ -0,0 +1,50 @@
|
||||
{
|
||||
"<h1>Training <a href=\"index.html\">U-Net</a></h1>\n<p>This trains a <a href=\"index.html\">U-Net</a> model on <a href=\"carvana.html\">Carvana dataset</a>. You can find the download instructions <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">on Kaggle</a>.</p>\n<p>Save the training images inside <span translate=no>_^_0_^_</span> folder and the masks in <span translate=no>_^_1_^_</span> folder.</p>\n<p>For simplicity, we do not do a training and validation split.</p>\n": "<h1>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0 <a href=\"index.html\">U-\u30cd\u30c3\u30c8</a></h1>\n<p>\u3053\u308c\u306b\u3088\u308a\u3001<a href=\"index.html\"><a href=\"carvana.html\">Carvana\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067U-Net\u30e2\u30c7\u30eb\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u307e\u3059</a></a>\u3002\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u624b\u9806\u306f <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">Kaggle \u3067\u78ba\u8a8d\u3067\u304d\u307e\u3059</a></p>\u3002\n<p><span translate=no>_^_0_^_</span>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u753b\u50cf\u3092\u30d5\u30a9\u30eb\u30c0\u30fc\u5185\u306b\u4fdd\u5b58\u3057\u3001<span translate=no>_^_1_^_</span>\u30de\u30b9\u30af\u3092\u30d5\u30a9\u30eb\u30c0\u30fc\u306b\u4fdd\u5b58\u3057\u307e\u3059\u3002</p>\n<p>\u308f\u304b\u308a\u3084\u3059\u304f\u3059\u308b\u305f\u3081\u306b\u3001\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3068\u691c\u8a3c\u306e\u5206\u5272\u306f\u884c\u3063\u3066\u3044\u307e\u305b\u3093\u3002</p>\n",
|
||||
"<h2>Configurations</h2>\n": "<h2>\u30b3\u30f3\u30d5\u30a3\u30ae\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3</h2>\n",
|
||||
"<h3>Sample images</h3>\n": "<h3>\u30b5\u30f3\u30d7\u30eb\u753b\u50cf</h3>\n",
|
||||
"<h3>Train for an epoch</h3>\n": "<h3>\u4e00\u6642\u4ee3\u3092\u62d3\u304f\u5217\u8eca</h3>\n",
|
||||
"<h3>Training loop</h3>\n": "<h3>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30eb\u30fc\u30d7</h3>\n",
|
||||
"<p> </p>\n": "<p></p>\n",
|
||||
"<p><a href=\"index.html\">U-Net</a> model </p>\n": "<p><a href=\"index.html\">U-\u30cd\u30c3\u30c8\u30e2\u30c7\u30eb</a></p>\n",
|
||||
"<p>Adam optimizer </p>\n": "<p>\u30a2\u30c0\u30e0\u30fb\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc</p>\n",
|
||||
"<p>Batch size </p>\n": "<p>\u30d0\u30c3\u30c1\u30b5\u30a4\u30ba</p>\n",
|
||||
"<p>Calculate loss </p>\n": "<p>\u640d\u5931\u306e\u8a08\u7b97</p>\n",
|
||||
"<p>Compute gradients </p>\n": "<p>\u52fe\u914d\u306e\u8a08\u7b97</p>\n",
|
||||
"<p>Create configurations </p>\n": "<p>\u69cb\u6210\u306e\u4f5c\u6210</p>\n",
|
||||
"<p>Create dataloader </p>\n": "<p>\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc\u306e\u4f5c\u6210</p>\n",
|
||||
"<p>Create experiment </p>\n": "<p>\u5b9f\u9a13\u3092\u4f5c\u6210</p>\n",
|
||||
"<p>Create optimizer </p>\n": "<p>\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc\u3092\u4f5c\u6210</p>\n",
|
||||
"<p>Crop the image to the size of the mask </p>\n": "<p>\u753b\u50cf\u3092\u30de\u30b9\u30af\u306e\u30b5\u30a4\u30ba\u306b\u30c8\u30ea\u30df\u30f3\u30b0\u3057\u307e\u3059</p>\n",
|
||||
"<p>Crop the target mask to the size of the logits. Size of the logits will be smaller if we don't use padding in convolutional layers in the U-Net. </p>\n": "<p>\u30bf\u30fc\u30b2\u30c3\u30c8\u30de\u30b9\u30af\u3092\u30ed\u30b8\u30c3\u30c8\u306e\u30b5\u30a4\u30ba\u306b\u30c8\u30ea\u30df\u30f3\u30b0\u3057\u307e\u3059\u3002U-Net\u306e\u7573\u307f\u8fbc\u307f\u5c64\u306b\u30d1\u30c7\u30a3\u30f3\u30b0\u3092\u4f7f\u308f\u306a\u3044\u3068\u3001\u30ed\u30b8\u30c3\u30c8\u306e\u30b5\u30a4\u30ba\u306f\u5c0f\u3055\u304f\u306a\u308a\u307e\u3059</p>\u3002\n",
|
||||
"<p>Dataloader </p>\n": "<p>\u30c7\u30fc\u30bf\u30ed\u30fc\u30c0\u30fc</p>\n",
|
||||
"<p>Dataset </p>\n": "<p>\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8</p>\n",
|
||||
"<p>Device to train the model on. <a href=\"https://docs.labml.ai/api/helpers.html#labml_helpers.device.DeviceConfigs\"><span translate=no>_^_0_^_</span></a> picks up an available CUDA device or defaults to CPU. </p>\n": "<p>\u30e2\u30c7\u30eb\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3059\u308b\u30c7\u30d0\u30a4\u30b9\u3002<a href=\"https://docs.labml.ai/api/helpers.html#labml_helpers.device.DeviceConfigs\"><span translate=no>_^_0_^_</span></a>\u4f7f\u7528\u53ef\u80fd\u306a CUDA \u30c7\u30d0\u30a4\u30b9\u3092\u9078\u629e\u3059\u308b\u304b\u3001\u30c7\u30d5\u30a9\u30eb\u30c8\u3067 CPU \u306b\u8a2d\u5b9a\u3057\u307e\u3059</p>\u3002\n",
|
||||
"<p>Get a random sample </p>\n": "<p>\u30e9\u30f3\u30c0\u30e0\u30b5\u30f3\u30d7\u30eb\u3092\u5165\u624b</p>\n",
|
||||
"<p>Get predicted mask </p>\n": "<p>\u4e88\u6e2c\u30de\u30b9\u30af\u3092\u53d6\u5f97</p>\n",
|
||||
"<p>Get predicted mask logits </p>\n": "<p>\u4e88\u6e2c\u3055\u308c\u305f\u30de\u30b9\u30af\u30ed\u30b8\u30c3\u30c8\u306e\u53d6\u5f97</p>\n",
|
||||
"<p>Image logging </p>\n": "<p>\u753b\u50cf\u30ed\u30ae\u30f3\u30b0</p>\n",
|
||||
"<p>Increment global step </p>\n": "<p>\u30b0\u30ed\u30fc\u30d0\u30eb\u30b9\u30c6\u30c3\u30d7\u3092\u30a4\u30f3\u30af\u30ea\u30e1\u30f3\u30c8</p>\n",
|
||||
"<p>Initialize </p>\n": "<p>[\u521d\u671f\u5316]</p>\n",
|
||||
"<p>Initialize the <a href=\"carvana.html\">Carvana dataset</a> </p>\n": "<p><a href=\"carvana.html\">Carvana</a> \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u521d\u671f\u5316\u3057\u307e\u3059</p>\n",
|
||||
"<p>Initialize the model </p>\n": "<p>\u30e2\u30c7\u30eb\u3092\u521d\u671f\u5316</p>\n",
|
||||
"<p>Iterate through the dataset. Use <a href=\"https://docs.labml.ai/api/monit.html#labml.monit.mix\"><span translate=no>_^_0_^_</span></a> to sample <span translate=no>_^_1_^_</span> times per epoch. </p>\n": "<p>\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u7e70\u308a\u8fd4\u3057\u51e6\u7406\u3057\u307e\u3059\u3002<a href=\"https://docs.labml.ai/api/monit.html#labml.monit.mix\"><span translate=no>_^_0_^_</span></a><span translate=no>_^_1_^_</span>\u30a8\u30dd\u30c3\u30af\u3042\u305f\u308a\u306e\u30b5\u30f3\u30d7\u30ea\u30f3\u30b0\u6642\u9593\u306b\u4f7f\u7528\u3057\u307e\u3059</p>\u3002\n",
|
||||
"<p>Learning rate </p>\n": "<p>\u5b66\u7fd2\u7387</p>\n",
|
||||
"<p>Log samples </p>\n": "<p>\u30ed\u30b0\u30b5\u30f3\u30d7\u30eb</p>\n",
|
||||
"<p>Loss function </p>\n": "<p>\u640d\u5931\u95a2\u6570</p>\n",
|
||||
"<p>Make the gradients zero </p>\n": "<p>\u30b0\u30e9\u30c7\u30fc\u30b7\u30e7\u30f3\u3092\u30bc\u30ed\u306b\u3059\u308b</p>\n",
|
||||
"<p>Move data to device </p>\n": "<p>\u30c7\u30fc\u30bf\u3092\u30c7\u30d0\u30a4\u30b9\u306b\u79fb\u52d5</p>\n",
|
||||
"<p>New line in the console </p>\n": "<p>\u30b3\u30f3\u30bd\u30fc\u30eb\u306e\u65b0\u3057\u3044\u884c</p>\n",
|
||||
"<p>Number of channels in the image. <span translate=no>_^_0_^_</span> for RGB. </p>\n": "<p>\u753b\u50cf\u5185\u306e\u30c1\u30e3\u30f3\u30cd\u30eb\u6570\u3002<span translate=no>_^_0_^_</span>RGB \u7528\u3067\u3059\u3002</p>\n",
|
||||
"<p>Number of channels in the output mask. <span translate=no>_^_0_^_</span> for binary mask. </p>\n": "<p>\u51fa\u529b\u30de\u30b9\u30af\u306e\u30c1\u30e3\u30f3\u30cd\u30eb\u6570\u3002<span translate=no>_^_0_^_</span>\u30d0\u30a4\u30ca\u30ea\u30de\u30b9\u30af\u7528\u3002</p>\n",
|
||||
"<p>Number of training epochs </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30a8\u30dd\u30c3\u30af\u306e\u6570</p>\n",
|
||||
"<p>Save the model </p>\n": "<p>\u30e2\u30c7\u30eb\u3092\u4fdd\u5b58\u3059\u308b</p>\n",
|
||||
"<p>Set configurations. You can override the defaults by passing the values in the dictionary. </p>\n": "<p>\u69cb\u6210\u3092\u8a2d\u5b9a\u3057\u307e\u3059\u3002\u30c7\u30a3\u30af\u30b7\u30e7\u30ca\u30ea\u306b\u5024\u3092\u6e21\u3059\u3053\u3068\u3067\u30c7\u30d5\u30a9\u30eb\u30c8\u3092\u30aa\u30fc\u30d0\u30fc\u30e9\u30a4\u30c9\u3067\u304d\u307e\u3059\u3002</p>\n",
|
||||
"<p>Set models for saving and loading </p>\n": "<p>\u4fdd\u5b58\u304a\u3088\u3073\u8aad\u307f\u8fbc\u307f\u7528\u306e\u30e2\u30c7\u30eb\u3092\u8a2d\u5b9a\u3059\u308b</p>\n",
|
||||
"<p>Sigmoid function for binary classification </p>\n": "<p>\u30d0\u30a4\u30ca\u30ea\u5206\u985e\u7528\u306e\u30b7\u30b0\u30e2\u30a4\u30c9\u95a2\u6570</p>\n",
|
||||
"<p>Start and run the training loop </p>\n": "<p>\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30eb\u30fc\u30d7\u3092\u958b\u59cb\u3057\u3066\u5b9f\u884c\u3059\u308b</p>\n",
|
||||
"<p>Take an optimization step </p>\n": "<p>\u6700\u9069\u5316\u306e\u4e00\u6b69\u3092\u8e0f\u307f\u51fa\u3059</p>\n",
|
||||
"<p>Track the loss </p>\n": "<p>\u640d\u5931\u3092\u30c8\u30e9\u30c3\u30ad\u30f3\u30b0</p>\n",
|
||||
"<p>Train the model </p>\n": "<p>\u30e2\u30c7\u30eb\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0</p>\n",
|
||||
"Code for training a U-Net model on Carvana dataset.": "Carvana\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067U-Net\u30e2\u30c7\u30eb\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3059\u308b\u305f\u3081\u306e\u30b3\u30fc\u30c9\u3002",
|
||||
"Training a U-Net on Carvana dataset": "Carvana\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067\u306eU-Net\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0"
|
||||
}
|
||||
@@ -0,0 +1,50 @@
|
||||
{
|
||||
"<h1>Training <a href=\"index.html\">U-Net</a></h1>\n<p>This trains a <a href=\"index.html\">U-Net</a> model on <a href=\"carvana.html\">Carvana dataset</a>. You can find the download instructions <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">on Kaggle</a>.</p>\n<p>Save the training images inside <span translate=no>_^_0_^_</span> folder and the masks in <span translate=no>_^_1_^_</span> folder.</p>\n<p>For simplicity, we do not do a training and validation split.</p>\n": "<h1><a href=\"index.html\">U-\u0dc1\u0dd4\u0daf\u0dca\u0db0</a>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4</h1>\n<p>\u0db8\u0dd9\u0dba <a href=\"carvana.html\">\u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba\u0dda</a> <a href=\"index.html\">\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca</a> \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0d9a\u0dca \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0dba\u0dd2. \u0d94\u0db6\u0da7 \u0db6\u0dcf\u0d9c\u0dad \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d8b\u0db4\u0daf\u0dd9\u0dc3\u0dca <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">\u0dc3\u0ddc\u0dba\u0dcf\u0d9c\u0dad \u0dc4\u0dd0\u0d9a\u0dd2\u0dba Kaggle</a>. </p>\n<p><span translate=no>_^_0_^_</span> \u0dc6\u0ddd\u0dbd\u0dca\u0da9\u0dbb\u0dba \u0dad\u0dd4\u0dc5 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0db4\u0dd2\u0db1\u0dca\u0dad\u0dd6\u0dbb \u0dc3\u0dc4 <span translate=no>_^_1_^_</span> \u0dc6\u0ddd\u0dbd\u0dca\u0da9\u0dbb\u0dba\u0dda \u0dc0\u0dd9\u0dc3\u0dca \u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0dc3\u0dd4\u0dbb\u0d9a\u0dd2\u0db1\u0dca\u0db1. </p>\n<p>\u0dc3\u0dbb\u0dbd\u0db6\u0dc0 \u0dc3\u0db3\u0dc4\u0dcf, \u0d85\u0db4\u0dd2 \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dc0\u0d9a\u0dca \u0dc3\u0dc4 \u0dc0\u0dbd\u0d82\u0d9c\u0dd4 \u0db7\u0dda\u0daf\u0dba\u0d9a\u0dca \u0db1\u0ddc\u0d9a\u0dbb\u0db8\u0dd4. </p>\n",
|
||||
"<h2>Configurations</h2>\n": "<h2>\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dca</h2>\n",
|
||||
"<h3>Sample images</h3>\n": "<h3>\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dbb\u0dd6\u0db4</h3>\n",
|
||||
"<h3>Train for an epoch</h3>\n": "<h3>\u0d9aepoch \u0dc3\u0db3\u0dc4\u0dcf \u0daf\u0dd4\u0db8\u0dca\u0dbb\u0dd2\u0dba</h3>\n",
|
||||
"<h3>Training loop</h3>\n": "<h3>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dbd\u0dd6\u0db4\u0dba</h3>\n",
|
||||
"<p> </p>\n": "<p> </p>\n",
|
||||
"<p><a href=\"index.html\">U-Net</a> model </p>\n": "<p><a href=\"index.html\">\u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca</a> \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba </p>\n",
|
||||
"<p>Adam optimizer </p>\n": "<p>\u0d86\u0daf\u0db8\u0dca\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0d9a\u0dbb\u0dab\u0dba </p>\n",
|
||||
"<p>Batch size </p>\n": "<p>\u0d9a\u0dab\u0dca\u0da9\u0dcf\u0dba\u0db8\u0dca\u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba </p>\n",
|
||||
"<p>Calculate loss </p>\n": "<p>\u0d85\u0dbd\u0dcf\u0db7\u0dba\u0d9c\u0dab\u0db1\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Compute gradients </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0d9c\u0dab\u0db1\u0dba </p>\n",
|
||||
"<p>Create configurations </p>\n": "<p>\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca\u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Create dataloader </p>\n": "<p>\u0daf\u0dad\u0dca\u0dad\u0d9a\u0dcf\u0dbb\u0d9a\u0dba \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Create experiment </p>\n": "<p>\u0d85\u0dad\u0dca\u0dc4\u0daf\u0dcf\u0db6\u0dd0\u0dbd\u0dd3\u0db8 \u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Create optimizer </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0d9a\u0dbb\u0dab\u0dba\u0dc3\u0dcf\u0daf\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Crop the image to the size of the mask </p>\n": "<p>\u0dc0\u0dd9\u0dc3\u0dca\u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0dd9\u0dc4\u0dd2 \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba\u0da7 \u0dbb\u0dd6\u0db4\u0dba \u0dc0\u0d9c\u0dcf \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Crop the target mask to the size of the logits. Size of the logits will be smaller if we don't use padding in convolutional layers in the U-Net. </p>\n": "<p>\u0d89\u0dbd\u0d9a\u0dca\u0d9a\u0d9c\u0dad\u0dc0\u0dd9\u0dc3\u0dca\u0db8\u0dd4\u0dc4\u0dd4\u0dab \u0db4\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba\u0da7 \u0dc0\u0d9c\u0dcf \u0d9a\u0dbb\u0db1\u0dca\u0db1. \u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca \u0dc4\u0dd2 \u0dc3\u0d82\u0dc0\u0dbd\u0dd2\u0dad \u0dc3\u0dca\u0dae\u0dbb \u0dc0\u0dbd \u0d85\u0db4\u0dd2 \u0db4\u0dd1\u0da9\u0dd2\u0db1\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0db1\u0ddc\u0d9a\u0dbb\u0db1\u0dca\u0db1\u0dda \u0db1\u0db8\u0dca \u0db4\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca \u0db4\u0dca\u0dbb\u0db8\u0dcf\u0dab\u0dba \u0d9a\u0dd4\u0da9\u0dcf \u0dc0\u0dda. </p>\n",
|
||||
"<p>Dataloader </p>\n": "<p>\u0daf\u0dad\u0dca\u0dad\u0d9a\u0dcf\u0dbb\u0d9a\u0dba </p>\n",
|
||||
"<p>Dataset </p>\n": "<p>\u0daf\u0dad\u0dca\u0dad\u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba </p>\n",
|
||||
"<p>Device to train the model on. <a href=\"https://docs.labml.ai/api/helpers.html#labml_helpers.device.DeviceConfigs\"><span translate=no>_^_0_^_</span></a> picks up an available CUDA device or defaults to CPU. </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dda \u0d8b\u0db4\u0d9a\u0dbb\u0dab\u0dba. <a href=\"https://docs.labml.ai/api/helpers.html#labml_helpers.device.DeviceConfigs\"><span translate=no>_^_0_^_</span></a> \u0dbd\u0db6\u0dcf \u0d9c\u0dad \u0dc4\u0dd0\u0d9a\u0dd2 CUDA \u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba\u0d9a\u0dca \u0d85\u0dc4\u0dd4\u0dbd\u0db1\u0dc0\u0dcf \u0dc4\u0ddd CPU \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0da7 \u0db4\u0dd9\u0dbb\u0db1\u0dd2\u0db8\u0dd2. </p>\n",
|
||||
"<p>Get a random sample </p>\n": "<p>\u0d85\u0dc4\u0db9\u0dd4\u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2\u0dba\u0d9a\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Get predicted mask </p>\n": "<p>\u0db4\u0dd4\u0dbb\u0ddd\u0d9a\u0dae\u0db1\u0dba\u0d9a\u0dc5 \u0dc0\u0dd9\u0dc3\u0dca \u0db8\u0dd4\u0dc4\u0dd4\u0dab \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Get predicted mask logits </p>\n": "<p>\u0db4\u0dd4\u0dbb\u0ddd\u0d9a\u0dae\u0db1\u0dba\u0d9a\u0dbb\u0db1 \u0dbd\u0daf \u0dc0\u0dd9\u0dc3\u0dca\u0db8\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0db4\u0dd2\u0dc0\u0dd2\u0dc3\u0dd4\u0db8\u0dca \u0dbd\u0db6\u0dcf \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Image logging </p>\n": "<p>\u0dbb\u0dd6\u0db4\u0dbd\u0ddc\u0d9c\u0dca \u0dc0\u0dd3\u0db8 </p>\n",
|
||||
"<p>Increment global step </p>\n": "<p>\u0d9c\u0ddd\u0dbd\u0dd3\u0dba\u0db4\u0dd2\u0dba\u0dc0\u0dbb \u0dc0\u0dd0\u0da9\u0dd2 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 </p>\n",
|
||||
"<p>Initialize </p>\n": "<p>\u0d86\u0dbb\u0db8\u0dca\u0db7\u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Initialize the <a href=\"carvana.html\">Carvana dataset</a> </p>\n": "<p><a href=\"carvana.html\">\u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba</a> \u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Initialize the model </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Iterate through the dataset. Use <a href=\"https://docs.labml.ai/api/monit.html#labml.monit.mix\"><span translate=no>_^_0_^_</span></a> to sample <span translate=no>_^_1_^_</span> times per epoch. </p>\n": "<p>\u0daf\u0dad\u0dca\u0dad\u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba \u0dc4\u0dbb\u0dc4\u0dcf \u0db1\u0dd0\u0dc0\u0dad \u0d9a\u0dbb\u0db1\u0dca\u0db1. \u0d91\u0d9a\u0dca <a href=\"https://docs.labml.ai/api/monit.html#labml.monit.mix\"><span translate=no>_^_0_^_</span></a> <span translate=no>_^_1_^_</span> \u0dba\u0dd4\u0d9c\u0dba\u0d9a\u0da7 \u0db1\u0dd2\u0dba\u0dd0\u0daf\u0dd2 \u0dc0\u0dda\u0dbd\u0dcf\u0dc0\u0db1\u0dca \u0db7\u0dcf\u0dc0\u0dd2\u0dad\u0dcf \u0d9a\u0dbb\u0db1\u0dca\u0db1. </p>\n",
|
||||
"<p>Learning rate </p>\n": "<p>\u0d89\u0d9c\u0dd9\u0db1\u0dd4\u0db8\u0dca\u0d85\u0db1\u0dd4\u0db4\u0dcf\u0dad\u0dba </p>\n",
|
||||
"<p>Log samples </p>\n": "<p>\u0dbd\u0ddc\u0d9c\u0dca\u0dc3\u0dcf\u0db8\u0dca\u0db4\u0dbd </p>\n",
|
||||
"<p>Loss function </p>\n": "<p>\u0db4\u0dcf\u0da9\u0dd4\u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
|
||||
"<p>Make the gradients zero </p>\n": "<p>\u0d85\u0db1\u0dd4\u0d9a\u0dca\u0dbb\u0db8\u0dd2\u0d9a\u0dc1\u0dd4\u0db1\u0dca\u0dba \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Move data to device </p>\n": "<p>\u0d8b\u0db4\u0dcf\u0d82\u0d9c\u0dba\u0dc0\u0dd9\u0dad \u0daf\u0dad\u0dca\u0dad \u0d9c\u0dd9\u0db1\u0dba\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>New line in the console </p>\n": "<p>\u0d9a\u0ddc\u0db1\u0dca\u0dc3\u0ddd\u0dbd\u0dba\u0dda\u0db1\u0dc0 \u0dbb\u0dda\u0d9b\u0dcf\u0dc0\u0d9a\u0dca </p>\n",
|
||||
"<p>Number of channels in the image. <span translate=no>_^_0_^_</span> for RGB. </p>\n": "<p>\u0dbb\u0dd6\u0db4\u0dba\u0dda\u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1. <span translate=no>_^_0_^_</span> RGB \u0dc3\u0db3\u0dc4\u0dcf. </p>\n",
|
||||
"<p>Number of channels in the output mask. <span translate=no>_^_0_^_</span> for binary mask. </p>\n": "<p>\u0db1\u0dd2\u0db8\u0dd0\u0dc0\u0dd4\u0db8\u0dca\u0d86\u0dc0\u0dbb\u0dab\u0dba\u0dda \u0db1\u0dcf\u0dbd\u0dd2\u0d9a\u0dcf \u0d9c\u0dab\u0db1. <span translate=no>_^_0_^_</span> \u0daf\u0dca\u0dc0\u0dd2\u0db8\u0dba \u0d86\u0dc0\u0dbb\u0dab \u0dc3\u0db3\u0dc4\u0dcf. </p>\n",
|
||||
"<p>Number of training epochs </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0d91\u0db4\u0ddc\u0da0\u0dca \u0d9c\u0dab\u0db1 </p>\n",
|
||||
"<p>Save the model </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0dc3\u0dd4\u0dbb\u0d9a\u0dd2\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Set configurations. You can override the defaults by passing the values in the dictionary. </p>\n": "<p>\u0dc0\u0dd2\u0db1\u0dca\u0dba\u0dcf\u0dc3\u0dba\u0db1\u0dca\u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1. \u0dc1\u0db6\u0dca\u0daf\u0d9a\u0ddd\u0dc2\u0dba\u0dda \u0d85\u0d9c\u0dba\u0db1\u0dca \u0dc3\u0db8\u0dca\u0db8\u0dad \u0d9a\u0dd2\u0dbb\u0dd3\u0db8\u0dd9\u0db1\u0dca \u0d94\u0db6\u0da7 \u0db4\u0dd9\u0dbb\u0db1\u0dd2\u0db8\u0dd2 \u0d85\u0db7\u0dd2\u0db6\u0dc0\u0dcf \u0dba\u0dcf \u0dc4\u0dd0\u0d9a\u0dd2\u0dba. </p>\n",
|
||||
"<p>Set models for saving and loading </p>\n": "<p>\u0d89\u0dad\u0dd2\u0dbb\u0dd2\u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0dc4 \u0db4\u0dd0\u0da7\u0dc0\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d86\u0d9a\u0dd8\u0dad\u0dd2 \u0dc3\u0d9a\u0dc3\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Sigmoid function for binary classification </p>\n": "<p>\u0daf\u0dca\u0dc0\u0dd2\u0db8\u0dba\u0dc0\u0dbb\u0dca\u0d9c\u0dd3\u0d9a\u0dbb\u0dab\u0dba \u0dc3\u0db3\u0dc4\u0dcf \u0dc3\u0dd2\u0d9c\u0dca\u0db8\u0ddd\u0dba\u0dd2\u0da9\u0dca \u0dc1\u0dca\u0dbb\u0dd2\u0dad\u0dba </p>\n",
|
||||
"<p>Start and run the training loop </p>\n": "<p>\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4\u0dbd\u0dd6\u0db4\u0dba \u0d86\u0dbb\u0db8\u0dca\u0db7 \u0d9a\u0dbb \u0d9a\u0dca\u0dbb\u0dd2\u0dba\u0dcf\u0dad\u0dca\u0db8\u0d9a \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Take an optimization step </p>\n": "<p>\u0db4\u0dca\u0dbb\u0dc1\u0dc3\u0dca\u0dad\u0dd2\u0d9a\u0dbb\u0dab\u0db4\u0dd2\u0dba\u0dc0\u0dbb\u0d9a\u0dca \u0d9c\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Track the loss </p>\n": "<p>\u0d85\u0dbd\u0dcf\u0db7\u0dba\u0dbd\u0dd4\u0dc4\u0dd4\u0db6\u0db3\u0dd2\u0db1\u0dca\u0db1 </p>\n",
|
||||
"<p>Train the model </p>\n": "<p>\u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dbb\u0db1\u0dca\u0db1 </p>\n",
|
||||
"Code for training a U-Net model on Carvana dataset.": "\u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba \u0db4\u0dd2\u0dc5\u0dd2\u0db6\u0db3 U-Net \u0d86\u0d9a\u0dd8\u0dad\u0dd2\u0dba\u0d9a\u0dca \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8 \u0dc3\u0db3\u0dc4\u0dcf \u0d9a\u0dda\u0dad\u0dba.",
|
||||
"Training a U-Net on Carvana dataset": "\u0d9a\u0dcf\u0dbb\u0dca\u0dc0\u0dcf\u0db1\u0dcf \u0daf\u0dad\u0dca\u0dad \u0d9a\u0da7\u0dca\u0da7\u0dbd\u0dba \u0db4\u0dd2\u0dc5\u0dd2\u0db6\u0db3 \u0dba\u0dd6-\u0db1\u0dd9\u0da7\u0dca \u0db4\u0dd4\u0dc4\u0dd4\u0dab\u0dd4 \u0d9a\u0dd2\u0dbb\u0dd3\u0db8"
|
||||
}
|
||||
@@ -0,0 +1,50 @@
|
||||
{
|
||||
"<h1>Training <a href=\"index.html\">U-Net</a></h1>\n<p>This trains a <a href=\"index.html\">U-Net</a> model on <a href=\"carvana.html\">Carvana dataset</a>. You can find the download instructions <a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">on Kaggle</a>.</p>\n<p>Save the training images inside <span translate=no>_^_0_^_</span> folder and the masks in <span translate=no>_^_1_^_</span> folder.</p>\n<p>For simplicity, we do not do a training and validation split.</p>\n": "<h1>\u8bad\u7ec3 <a href=\"index.html\">U-Net</a></h1>\n<p>\u8fd9\u4f1a\u5728 <a href=\"carvana.html\">Carvana \u6570\u636e\u96c6</a>\u4e0a\u8bad\u7ec3\u4e00\u4e2a <a href=\"index.html\">U-Net</a> \u6a21\u578b\u3002\u4f60\u53ef\u4ee5<a href=\"https://www.kaggle.com/competitions/carvana-image-masking-challenge/data\">\u5728 Kaggle \u4e0a</a>\u627e\u5230\u4e0b\u8f7d\u8bf4\u660e\u3002</p>\n<p>\u5c06\u8bad\u7ec3\u56fe\u50cf\u4fdd\u5b58\u5728<span translate=no>_^_0_^_</span>\u6587\u4ef6\u5939\u4e2d\uff0c\u5c06\u8499\u7248\u4fdd\u5b58\u5728<span translate=no>_^_1_^_</span>\u6587\u4ef6\u5939\u4e2d\u3002</p>\n<p>\u4e3a\u7b80\u5355\u8d77\u89c1\uff0c\u6211\u4eec\u4e0d\u8fdb\u884c\u8bad\u7ec3\u548c\u9a8c\u8bc1\u62c6\u5206\u3002</p>\n",
|
||||
"<h2>Configurations</h2>\n": "<h2>\u914d\u7f6e</h2>\n",
|
||||
"<h3>Sample images</h3>\n": "<h3>\u6837\u672c\u56fe\u7247</h3>\n",
|
||||
"<h3>Train for an epoch</h3>\n": "<h3>\u8bad\u7ec3\u4e00\u4e2a\u65f6\u4ee3</h3>\n",
|
||||
"<h3>Training loop</h3>\n": "<h3>\u8bad\u7ec3\u5faa\u73af</h3>\n",
|
||||
"<p> </p>\n": "<p></p>\n",
|
||||
"<p><a href=\"index.html\">U-Net</a> model </p>\n": "<p><a href=\"index.html\">U-Net</a> \u6a21\u578b</p>\n",
|
||||
"<p>Adam optimizer </p>\n": "<p>Adam \u4f18\u5316\u5668</p>\n",
|
||||
"<p>Batch size </p>\n": "<p>\u6279\u91cf\u5927\u5c0f</p>\n",
|
||||
"<p>Calculate loss </p>\n": "<p>\u8ba1\u7b97\u635f\u5931</p>\n",
|
||||
"<p>Compute gradients </p>\n": "<p>\u8ba1\u7b97\u68af\u5ea6</p>\n",
|
||||
"<p>Create configurations </p>\n": "<p>\u521b\u5efa\u914d\u7f6e</p>\n",
|
||||
"<p>Create dataloader </p>\n": "<p>\u521b\u5efa\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
|
||||
"<p>Create experiment </p>\n": "<p>\u521b\u5efa\u5b9e\u9a8c</p>\n",
|
||||
"<p>Create optimizer </p>\n": "<p>\u521b\u5efa\u4f18\u5316\u5668</p>\n",
|
||||
"<p>Crop the image to the size of the mask </p>\n": "<p>\u5c06\u56fe\u50cf\u88c1\u526a\u4e3a\u8499\u7248\u7684\u5927\u5c0f</p>\n",
|
||||
"<p>Crop the target mask to the size of the logits. Size of the logits will be smaller if we don't use padding in convolutional layers in the U-Net. </p>\n": "<p>\u5c06\u76ee\u6807\u8499\u7248\u88c1\u526a\u4e3a logits \u7684\u5927\u5c0f\u3002\u5982\u679c\u6211\u4eec\u4e0d\u5728 U-Net \u7684\u5377\u79ef\u5c42\u4e2d\u4f7f\u7528\u586b\u5145\uff0clogits \u7684\u5927\u5c0f\u4f1a\u53d8\u5c0f\u3002</p>\n",
|
||||
"<p>Dataloader </p>\n": "<p>\u6570\u636e\u52a0\u8f7d\u5668</p>\n",
|
||||
"<p>Dataset </p>\n": "<p>\u6570\u636e\u96c6</p>\n",
|
||||
"<p>Device to train the model on. <a href=\"https://docs.labml.ai/api/helpers.html#labml_helpers.device.DeviceConfigs\"><span translate=no>_^_0_^_</span></a> picks up an available CUDA device or defaults to CPU. </p>\n": "<p>\u7528\u4e8e\u8bad\u7ec3\u6a21\u578b\u7684\u8bbe\u5907\u3002<a href=\"https://docs.labml.ai/api/helpers.html#labml_helpers.device.DeviceConfigs\"><span translate=no>_^_0_^_</span></a>\u9009\u62e9\u53ef\u7528\u7684 CUDA \u8bbe\u5907\u6216\u9ed8\u8ba4\u4e3a CPU\u3002</p>\n",
|
||||
"<p>Get a random sample </p>\n": "<p>\u968f\u673a\u83b7\u53d6\u6837\u672c</p>\n",
|
||||
"<p>Get predicted mask </p>\n": "<p>\u83b7\u53d6\u9884\u6d4b\u7684\u53e3\u7f69</p>\n",
|
||||
"<p>Get predicted mask logits </p>\n": "<p>\u83b7\u53d6\u9884\u6d4b\u7684\u63a9\u7801\u65e5\u5fd7</p>\n",
|
||||
"<p>Image logging </p>\n": "<p>\u56fe\u50cf\u65e5\u5fd7\u8bb0\u5f55</p>\n",
|
||||
"<p>Increment global step </p>\n": "<p>\u9012\u589e\u5168\u5c40\u6b65\u957f</p>\n",
|
||||
"<p>Initialize </p>\n": "<p>\u521d\u59cb\u5316</p>\n",
|
||||
"<p>Initialize the <a href=\"carvana.html\">Carvana dataset</a> </p>\n": "<p>\u521d\u59cb\u5316 C <a href=\"carvana.html\">arvana \u6570\u636e\u96c6</a></p>\n",
|
||||
"<p>Initialize the model </p>\n": "<p>\u521d\u59cb\u5316\u6a21\u578b</p>\n",
|
||||
"<p>Iterate through the dataset. Use <a href=\"https://docs.labml.ai/api/monit.html#labml.monit.mix\"><span translate=no>_^_0_^_</span></a> to sample <span translate=no>_^_1_^_</span> times per epoch. </p>\n": "<p>\u904d\u5386\u6570\u636e\u96c6\u3002\u7528\u4e8e<a href=\"https://docs.labml.ai/api/monit.html#labml.monit.mix\"><span translate=no>_^_0_^_</span></a>\u5bf9\u6bcf\u4e2a\u7eaa\u5143\u7684\u91c7\u6837<span translate=no>_^_1_^_</span>\u6b21\u6570\u3002</p>\n",
|
||||
"<p>Learning rate </p>\n": "<p>\u5b66\u4e60\u7387</p>\n",
|
||||
"<p>Log samples </p>\n": "<p>\u65e5\u5fd7\u6837\u672c</p>\n",
|
||||
"<p>Loss function </p>\n": "<p>\u4e8f\u635f\u51fd\u6570</p>\n",
|
||||
"<p>Make the gradients zero </p>\n": "<p>\u5c06\u6e10\u53d8\u8bbe\u4e3a\u96f6</p>\n",
|
||||
"<p>Move data to device </p>\n": "<p>\u5c06\u6570\u636e\u79fb\u52a8\u5230\u8bbe\u5907</p>\n",
|
||||
"<p>New line in the console </p>\n": "<p>\u63a7\u5236\u53f0\u4e2d\u7684\u65b0\u884c</p>\n",
|
||||
"<p>Number of channels in the image. <span translate=no>_^_0_^_</span> for RGB. </p>\n": "<p>\u56fe\u50cf\u4e2d\u7684\u901a\u9053\u6570\u3002<span translate=no>_^_0_^_</span>\u5bf9\u4e8e RGB\u3002</p>\n",
|
||||
"<p>Number of channels in the output mask. <span translate=no>_^_0_^_</span> for binary mask. </p>\n": "<p>\u8f93\u51fa\u63a9\u7801\u4e2d\u7684\u58f0\u9053\u6570\u3002<span translate=no>_^_0_^_</span>\u7528\u4e8e\u4e8c\u8fdb\u5236\u63a9\u7801\u3002</p>\n",
|
||||
"<p>Number of training epochs </p>\n": "<p>\u8bad\u7ec3\u5468\u671f\u7684\u6570\u91cf</p>\n",
|
||||
"<p>Save the model </p>\n": "<p>\u4fdd\u5b58\u6a21\u578b</p>\n",
|
||||
"<p>Set configurations. You can override the defaults by passing the values in the dictionary. </p>\n": "<p>\u8bbe\u7f6e\u914d\u7f6e\u3002\u60a8\u53ef\u4ee5\u901a\u8fc7\u5728\u5b57\u5178\u4e2d\u4f20\u9012\u503c\u6765\u8986\u76d6\u9ed8\u8ba4\u503c\u3002</p>\n",
|
||||
"<p>Set models for saving and loading </p>\n": "<p>\u8bbe\u7f6e\u7528\u4e8e\u4fdd\u5b58\u548c\u52a0\u8f7d\u7684\u6a21\u578b</p>\n",
|
||||
"<p>Sigmoid function for binary classification </p>\n": "<p>\u4e8c\u8fdb\u5236\u5206\u7c7b\u7684 Sigmoid \u51fd\u6570</p>\n",
|
||||
"<p>Start and run the training loop </p>\n": "<p>\u542f\u52a8\u5e76\u8fd0\u884c\u8bad\u7ec3\u5faa\u73af</p>\n",
|
||||
"<p>Take an optimization step </p>\n": "<p>\u91c7\u53d6\u4f18\u5316\u6b65\u9aa4</p>\n",
|
||||
"<p>Track the loss </p>\n": "<p>\u8ffd\u8e2a\u635f\u5931</p>\n",
|
||||
"<p>Train the model </p>\n": "<p>\u8bad\u7ec3\u6a21\u578b</p>\n",
|
||||
"Code for training a U-Net model on Carvana dataset.": "\u7528\u4e8e\u5728 Carvana \u6570\u636e\u96c6\u4e0a\u8bad\u7ec3 U-Net \u6a21\u578b\u7684\u4ee3\u7801\u3002",
|
||||
"Training a U-Net on Carvana dataset": "\u5728 Carvana \u6570\u636e\u96c6\u4e0a\u8bad\u7ec3 U-Net"
|
||||
}
|
||||
Reference in New Issue
Block a user