\n \ucd5c\uadfc \uae08\uc735\uc2dc\uc7a5\uc5d0\uc11c\uc758 \ub370\uc774\ud130 \uc591\uc774 \uae30\ud558\uae09\uc218\uc801\uc73c\ub85c \uc99d\uac00\ud568\uc5d0 \ub530\ub77c,
\n \ud22c\uc790\uc790 \ubc0f \ud2b8\ub808\uc774\ub354\ub4e4\uc740 \ubcf4\ub2e4 \uc815\uad50\ud55c \uc54c\uace0\ub9ac\uc998 \ubc0f \uba38\uc2e0\ub7ec\ub2dd \uae30\ubc95\uc744
\n \ud65c\uc6a9\ud558\uc5ec \ub370\uc774\ud130\ub85c\ubd80\ud130 \ud1b5\ucc30\uc744 \uc5bb\uace0\uc790 \ud569\ub2c8\ub2e4. \ubcf8 \uac15\uc88c\uc5d0\uc11c\ub294
\n \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd\uc744 \ud65c\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uae30\ubc95 \uc911
\n CNN-TA-2D \uc2dc\uacc4\uc5f4 \uad70\uc9d1\ud654 \uae30\ubc95\uc5d0 \ub300\ud574 \uc2ec\uce35\uc801\uc73c\ub85c \uc0b4\ud3b4\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.\n <\/p>\n<\/div>\n
\n \uba38\uc2e0\ub7ec\ub2dd(Machine Learning)\uc740 \ub370\uc774\ud130\ub97c \ud1b5\ud574 \ud559\uc2b5\ud558\uc5ec \uc608\uce21 \ubc0f \uacb0\uc815\uc744
\n \ub0b4\ub9ac\ub294 \uc54c\uace0\ub9ac\uc998\uc758 \uc9d1\ud569\uc744 \uc758\ubbf8\ud569\ub2c8\ub2e4. \ub525\ub7ec\ub2dd(Deep Learning)\uc740 \uba38\uc2e0\ub7ec\ub2dd\uc758
\n \ud55c \ubd84\uc57c\ub85c, \ub2e4\uce35 \uc2e0\uacbd\ub9dd(Neural Network)\uc744 \ud65c\uc6a9\ud558\uc5ec \uace0\ucc28\uc6d0 \ub370\uc774\ud130\ub97c
\n \ucc98\ub9ac\ud558\ub294 \uae30\uc220\uc785\ub2c8\ub2e4. \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc5d0\uc11c \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc744
\n \ud65c\uc6a9\ud558\ub294 \uc774\uc720\ub294 \ub370\uc774\ud130 \ud328\ud134\uc744 \uc790\ub3d9\uc73c\ub85c \ud559\uc2b5\ud558\uc5ec \ubcf4\ub2e4 \uc815\ud655\ud55c \uc608\uce21\uc744
\n \uac00\ub2a5\ud558\uac8c \ud558\uae30 \ub54c\ubb38\uc785\ub2c8\ub2e4.\n <\/p>\n<\/div>\n
\n \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc740 \ucef4\ud4e8\ud130 \ud504\ub85c\uadf8\ub7a8\uc744 \ud1b5\ud574 \uc790\ub3d9\uc73c\ub85c \ub9e4\ub9e4\ub97c
\n \uc218\ud589\ud558\ub294 \ubc29\uc2dd\uc785\ub2c8\ub2e4. \uc774 \ubc29\uc2dd\uc740 \ud22c\uc790\uc790\uc758 \uac10\uc815\uc774\ub098 \uc8fc\uad00\uc801 \ud310\ub2e8\uc5d0
\n \uc758\uc874\ud558\uc9c0 \uc54a\uac8c \ud574 \uc8fc\uba70, \ube60\ub978 \uc758\uc0ac\uacb0\uc815\uacfc \uc2e4\ud589\uc774 \uac00\ub2a5\ud569\ub2c8\ub2e4.
\n \uc77c\ubc18\uc801\uc73c\ub85c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529 \uc804\ub7b5\uc740 \ub2e4\uc74c\uacfc \uac19\uc774 \uad6c\uc131\ub429\ub2c8\ub2e4:\n <\/p>\n
\n CNN(Convolutional Neural Networks)\uc740 \uc8fc\ub85c \uc774\ubbf8\uc9c0 \ucc98\ub9ac\uc5d0 \uc0ac\uc6a9\ub418\ub294
\n \ub525\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998\uc73c\ub85c, Time-Series Analysis(TA)\ub97c \uacb0\ud569\ud558\uc5ec
\n 2D \ud615\uc2dd\uc73c\ub85c \uc2dc\uacc4\uc5f4 \ub370\uc774\ud130\ub97c \ucc98\ub9ac\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.
\n \uad70\uc9d1\ud654\ub294 \ube44\uc2b7\ud55c \ub370\uc774\ud130 \ud3ec\uc778\ud2b8\ub97c \uadf8\ub8f9\ud654\ud558\uc5ec \ub370\uc774\ud130\ub97c \ubd84\uc11d\ud558\uace0,
\n \ud328\ud134\uc744 \ubc1c\uacac\ud558\ub294 \ub370 \ub3c4\uc6c0\uc744 \uc90d\ub2c8\ub2e4.
\n \uc774 \uae30\ubc95\uc740 \uc8fc\uac00 \uc608\uce21 \ubc0f \ucd5c\uc801\uc758 \ub9e4\ub9e4 \uc2dc\uc810\uc744 \ud0d0\uc0c9\ud558\ub294 \ub370 \ub9e4\uc6b0 \uc720\uc6a9\ud569\ub2c8\ub2e4.\n <\/p>\n<\/div>\n
\n CNN\uc758 \uae30\ubcf8 \uad6c\uc870\ub294 \uc785\ub825\uce35, \uc740\ub2c9\uce35 \ubc0f \ucd9c\ub825\uce35\uc73c\ub85c \uad6c\uc131\ub418\uba70,
\n \uc8fc\uc694 \uc694\uc18c\ub85c\ub294 \ud569\uc131\uacf1\uce35(Convolutional Layer), \ud480\ub9c1\uce35(Pooling Layer),
\n \uc804\uacb0\ud569\uce35(Fully Connected Layer) \ub4f1\uc774 \uc788\uc2b5\ub2c8\ub2e4.
\n \uac01 \uc694\uc18c\ub294 \uc785\ub825 \ub370\uc774\ud130\ub97c \ubcc0\ud658\ud558\uc5ec \ud2b9\uc9d5\uc744 \ucd94\ucd9c\ud558\ub294 \ub370 \uc0ac\uc6a9\ub429\ub2c8\ub2e4.\n <\/p>\n
\n \ud569\uc131\uacf1\uce35\uc740 \uc785\ub825 \ub370\uc774\ud130\uc5d0 \ud544\ud130\ub97c \uc801\uc6a9\ud558\uc5ec \ud2b9\uc131 \ub9f5\uc744 \uc0dd\uc131\ud569\ub2c8\ub2e4.
\n \uc774 \uacfc\uc815\uc744 \ud1b5\ud574 \uc6d0\ub798 \ub370\uc774\ud130\uc758 \uc800\ucc28\uc6d0 \ud45c\ud604\uc744 \uc5bb\uc744 \uc218 \uc788\uc2b5\ub2c8\ub2e4.
\n \uc8fc\uac00 \ub370\uc774\ud130\uc758 \uacbd\uc6b0, \ud2b9\uc815 \uc2dc\uac04 \uac04\uaca9\uc758 \uac00\uaca9 \ubcc0\ub3d9 \ud328\ud134\uc744
\n \ud6a8\uacfc\uc801\uc73c\ub85c \ucd94\ucd9c\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n
\n \ud480\ub9c1\uce35\uc740 \ud2b9\uc131 \ub9f5\uc758 \ucc28\uc6d0\uc744 \uc904\uc5ec \uacc4\uc0b0\ub7c9\uc744 \uac10\uc18c\uc2dc\ud0a4\uace0
\n \uacfc\uc801\ud569(overfitting)\uc744 \ubc29\uc9c0\ud558\ub294 \uc5ed\ud560\uc744 \ud569\ub2c8\ub2e4. \uc8fc\ub85c \ud3c9\uade0 \ud480\ub9c1(Average Pooling)
\n \ub610\ub294 \ucd5c\ub300 \ud480\ub9c1(Max Pooling) \uae30\ubc95\uc744 \uc0ac\uc6a9\ud569\ub2c8\ub2e4.\n <\/p>\n
\n \uc804\uacb0\ud569\uce35\uc740 \ucd5c\uc885 \ucd9c\ub825\uc744 \uc0dd\uc131\ud558\ub294 \ub2e8\uacc4\ub85c, \ubaa8\ub4e0 \ub178\ub4dc\uac00 \uc774\uc804 \uce35\uc758
\n \ub178\ub4dc\uc640 \uc5f0\uacb0\ub418\uc5b4 \uc788\uc2b5\ub2c8\ub2e4. \uc774 \ub2e8\uacc4\uc5d0\uc11c\ub294 \ucd94\ucd9c\ub41c \ud2b9\uc9d5\uc744 \ubc14\ud0d5\uc73c\ub85c
\n \ucd5c\uc885 \uc608\uce21\uc744 \uc218\ud589\ud569\ub2c8\ub2e4.\n <\/p>\n<\/div>\n
\n \uc2dc\uacc4\uc5f4 \ub370\uc774\ud130\ub97c CNN \ubaa8\ub378\uc5d0 \ub9de\uac8c \uc900\ube44\ud558\ub294 \uacfc\uc815\uc740 \ub9e4\uc6b0 \uc911\uc694\ud569\ub2c8\ub2e4.
\n \uc8fc\uc2dd \uac00\uaca9 \ub370\uc774\ud130, \uac70\ub798\ub7c9, \uae30\uc220\uc801 \uc9c0\ud45c \ub4f1\uc744 \ud65c\uc6a9\ud558\uc5ec \uc785\ub825
\n \ub370\uc774\ud130\uc14b\uc744 \uad6c\uc131\ud569\ub2c8\ub2e4. \uc774 \ub370\uc774\ud130\ub97c 2D \ud615\ud0dc\ub85c \uc815\ub9ac\ud558\uc5ec
\n CNN \ubaa8\ub378\uc5d0 \ub9de\ub294 \ud615\uc2dd\uc73c\ub85c \ubcc0\ud658\ud574\uc57c \ud569\ub2c8\ub2e4.\n <\/p>\n
\n \ub370\uc774\ud130 \uc218\uc9d1\uc740 API \ud639\uc740 \ub370\uc774\ud130\ubca0\uc774\uc2a4\ub97c \ud1b5\ud574 \uc774\ub8e8\uc5b4\uc9c0\uba70,
\n \uc218\uc9d1\ub41c \ub370\uc774\ud130\ub294 \uacb0\uce21\uce58 \ucc98\ub9ac, \uc815\uaddc\ud654, \ubcc0\ud658 \ub4f1\uc758 \uc804\ucc98\ub9ac \uacfc\uc815\uc744
\n \uac70\uccd0\uc57c \ud569\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \uadf9\ub300\ud654\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n<\/div>\n
\n TensorFlow \ub610\ub294 PyTorch\ub97c \uc0ac\uc6a9\ud558\uc5ec CNN \ubaa8\ub378\uc744 \uad6c\ud604\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.
\n \uc544\ub798\ub294 TensorFlow\ub97c \ud65c\uc6a9\ud55c \uac04\ub2e8\ud55c CNN \ubaa8\ub378 \uc608\uc81c\uc785\ub2c8\ub2e4:\n <\/p>\n
\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras import layers\n\nmodel = keras.Sequential([\n layers.Conv2D(32, (3, 3), activation='relu', input_shape=(height, width, channels)),\n layers.MaxPooling2D((2, 2)),\n layers.Conv2D(64, (3, 3), activation='relu'),\n layers.MaxPooling2D((2, 2)),\n layers.Flatten(),\n layers.Dense(64, activation='relu'),\n layers.Dense(num_classes, activation='softmax')\n])\n\nmodel.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n<\/code><\/pre>\n<\/div>\n\n8. \ubaa8\ub378 \ud559\uc2b5 \ubc0f \ud3c9\uac00<\/h2>\n
\n \ubaa8\ub378 \ud559\uc2b5 \uc2dc \ud559\uc2b5 \ub370\uc774\ud130\uc640 \uac80\uc99d \ub370\uc774\ud130\ub97c \ub098\ub204\uc5b4 \uc0ac\uc6a9\ud569\ub2c8\ub2e4.
\n \uc77c\ubc18\uc801\uc73c\ub85c 70%\ub294 \ud559\uc2b5\uc5d0, 30%\ub294 \uac80\uc99d\uc5d0 \uc0ac\uc6a9\ud569\ub2c8\ub2e4.
\n \ubaa8\ub378\uc758 \uc131\ub2a5\uc740 \uc815\ud655\ub3c4(accuracy), \uc815\ubc00\ub3c4(precision),
\n \uc7ac\ud604\uc728(recall) \ub4f1 \ub2e4\uc591\ud55c \uc9c0\ud45c\ub97c \ud1b5\ud574 \ud3c9\uac00\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.\n <\/p>\n<\/div>\n