{"id":29427,"date":"2024-10-27T16:18:23","date_gmt":"2024-10-27T16:18:23","guid":{"rendered":"http:\/\/atmokpo.com\/w\/?p=29427"},"modified":"2024-11-26T06:52:41","modified_gmt":"2024-11-26T06:52:41","slug":"%eb%a8%b8%ec%8b%a0%eb%9f%ac%eb%8b%9d-%eb%b0%8f-%eb%94%a5%eb%9f%ac%eb%8b%9d-%ec%95%8c%ea%b3%a0%eb%a6%ac%ec%a6%98-%ed%8a%b8%eb%a0%88%ec%9d%b4%eb%94%a9-%ed%8c%8c%ec%9d%b4%ec%8d%ac%ec%9d%84-%ec%9d%b4-3","status":"publish","type":"post","link":"https:\/\/atmokpo.com\/w\/29427\/","title":{"rendered":"\uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529, \ud30c\uc774\uc36c\uc744 \uc774\uc6a9\ud574 \ucc98\uc74c\ubd80\ud130 \uc2e0\uacbd\ub9dd\uc744 \uad6c\ucd95\ud558\ub294 \ubc29\ubc95"},"content":{"rendered":"

<\/p>\n

\ucd5c\uadfc \uba87 \ub144 \ub3d9\uc548 \uae08\uc735 \uc2dc\uc7a5\uc5d0\uc11c \uba38\uc2e0\ub7ec\ub2dd(ML)\uacfc \ub525\ub7ec\ub2dd(DL) \uae30\uc220\uc744 \uc774\uc6a9\ud55c \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc774 \uc8fc\ubaa9\ubc1b\uace0 \uc788\uc2b5\ub2c8\ub2e4. \uc774 \uae00\uc5d0\uc11c\ub294 \ud30c\uc774\uc36c\uc744 \uc774\uc6a9\ud558\uc5ec \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uae30\ubc18\uc758 \ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \ucc98\uc74c\ubd80\ud130 \ub05d\uae4c\uc9c0 \uad6c\ucd95\ud558\ub294 \ubc29\ubc95\uc744 \uc124\uba85\ud569\ub2c8\ub2e4. \uc8fc\uc81c\ub97c \uc5ec\ub7ec \uac1c\uc758 \uc139\uc158\uc73c\ub85c \ub098\ub204\uc5b4 \uc124\uba85\ud558\uba70, \uac01 \uc139\uc158\uc5d0\uc11c\ub294 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc758 \uae30\ubcf8 \uac1c\ub150, \ub370\uc774\ud130 \uc218\uc9d1, \uc804\ucc98\ub9ac, \ubaa8\ub378 \uad6c\ucd95, \ud6c8\ub828, \ud3c9\uac00, \uadf8\ub9ac\uace0 \uc2e4\uc81c \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c\uc5d0 \uc801\uc6a9\ud558\ub294 \ub2e8\uacc4\uae4c\uc9c0 \ub2e4\ub8f9\ub2c8\ub2e4.<\/p>\n

1. \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc758 \uac1c\ub150<\/h2>\n

\uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc774\ub780 \uc77c\uc815\ud55c \uaddc\uce59\uc5d0 \ub530\ub77c \uc790\ub3d9\uc73c\ub85c \uc8fc\uc2dd, \uc678\ud658, \uc120\ubb3c \ubc0f \uae30\ud0c0 \uae08\uc735 \uc790\uc0b0\uc744 \uac70\ub798\ud558\ub294 \ud504\ub85c\uc138\uc2a4\ub97c \ub9d0\ud569\ub2c8\ub2e4. \uc774\ub7f0 \uc54c\uace0\ub9ac\uc998\uc740 \uc2dc\uc7a5 \ub370\uc774\ud130\ub97c \ubd84\uc11d\ud558\uace0 \uac70\ub798 \uacb0\uc815\uc744 \ub0b4\ub9ac\ub294 \ub370 \uc788\uc5b4 \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uae30\ubc95\uc744 \ud65c\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uc774\ub97c \ud1b5\ud574 \uc778\uac04\uc758 \uc8fc\uad00\uc801 \ud310\ub2e8\uc744 \ubc30\uc81c\ud558\uace0 \ubcf4\ub2e4 \uccb4\uacc4\uc801\uc774\uace0 \uc77c\uad00\ub41c \uac70\ub798 \uc804\ub7b5\uc744 \uad6c\ud604\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n

1.1. \uba38\uc2e0\ub7ec\ub2dd\uacfc \ub525\ub7ec\ub2dd\uc758 \ucc28\uc774<\/h3>\n

\uba38\uc2e0\ub7ec\ub2dd\uc740 \ub370\uc774\ud130\ub97c \uc774\uc6a9\ud558\uc5ec \ud328\ud134\uc744 \ud559\uc2b5\ud558\uace0 \uc608\uce21\uc744 \uc218\ud589\ud558\ub294 \uc54c\uace0\ub9ac\uc998\uc744 \ub9d0\ud569\ub2c8\ub2e4. \ubc18\uba74, \ub525\ub7ec\ub2dd\uc740 \uad6c\uc870\uac00 \ubcf5\uc7a1\ud55c \uc778\uacf5\uc2e0\uacbd\ub9dd\uc744 \ud65c\uc6a9\ud558\uc5ec \ube44\uc120\ud615 \uad00\uacc4\ub97c \ubaa8\ub378\ub9c1\ud558\ub294 \uba38\uc2e0\ub7ec\ub2dd\uc758 \ud558\uc704 \ubd84\uc57c\uc785\ub2c8\ub2e4. \uc8fc\uc2dd \uc2dc\uc7a5\uacfc \uac19\uc740 \ubcf5\uc7a1\ud55c \ub370\uc774\ud130\uc5d0\uc11c\ub294 \ub525\ub7ec\ub2dd\uc774 \ub354\uc6b1 \ud6a8\uacfc\uc801\uc77c \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n

2. \ub370\uc774\ud130 \uc218\uc9d1<\/h2>\n

\ud2b8\ub808\uc774\ub529 \uc54c\uace0\ub9ac\uc998\uc744 \uad6c\ucd95\ud558\uae30 \uc704\ud574\uc11c\ub294 \uc6b0\uc120\uc801\uc73c\ub85c \uc2e0\ub8b0\ud560 \uc218 \uc788\ub294 \ub370\uc774\ud130\uac00 \ud544\uc694\ud569\ub2c8\ub2e4. \uc8fc\uc2dd \ub370\uc774\ud130\ub294 Yahoo Finance, Alpha Vantage, Quandl \ub4f1\uc758 \uc11c\ube44\uc2a4\uc5d0\uc11c \uc218\uc9d1\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \ud30c\uc774\uc36c\uc5d0\uc11c\ub294 ‘pandas_datareader’\uc640 \uac19\uc740 \ub77c\uc774\ube0c\ub7ec\ub9ac\ub97c \ud1b5\ud574 \ub370\uc774\ud130\ub97c \uc27d\uac8c \ub2e4\uc6b4\ub85c\ub4dc\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n

2.1. Yahoo Finance\uc5d0\uc11c \ub370\uc774\ud130 \uc218\uc9d1\ud558\uae30<\/h3>\n

\ub2e4\uc74c\uc740 Yahoo Finance\uc5d0\uc11c \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud558\ub294 \ucf54\ub4dc \uc608\uc81c\uc785\ub2c8\ub2e4.<\/p>\n

import pandas as pd\nfrom pandas_datareader import data as pdr\n\n# \ud2b9\uc815 \uc8fc\uc2dd\uc758 \ub370\uc774\ud130\ub97c \uc218\uc9d1\ud569\ub2c8\ub2e4. \uc608\ub97c \ub4e4\uc5b4, AAPL (Apple Inc.)\nstart = '2010-01-01'\nend = '2022-01-01'\ndf = pdr.get_data_yahoo('AAPL', start=start, end=end)\nprint(df.head())<\/code><\/pre>\n
3. \ub370\uc774\ud130 \uc804\ucc98\ub9ac<\/h2>\n
\ub370\uc774\ud130\ub97c \uc218\uc9d1\ud55c \ud6c4, \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc5d0 \uc801\ud569\ud55c \ud615\ud0dc\ub85c \uc804\ucc98\ub9ac\ud574\uc57c \ud569\ub2c8\ub2e4. \uc804\ucc98\ub9ac \uacfc\uc815\uc5d0\ub294 \uacb0\uce21\uce58 \ucc98\ub9ac, \uc815\uaddc\ud654, \ub808\uc774\ube14 \uc778\ucf54\ub529 \ub4f1\uc774 \ud3ec\ud568\ub429\ub2c8\ub2e4. \ud2b9\ud788, \uc8fc\uc2dd \ub370\uc774\ud130\uc5d0\uc11c\ub294 \ub0a0\uc9dc\ub97c \uc778\ub371\uc2a4\ub85c \uc124\uc815\ud558\uace0, \ucc28\ud2b8 \ubd84\uc11d\uc744 \uc704\ud574 \uac00\uaca9 \ub370\uc774\ud130\uc758 \ubcc0\ub3d9\uc131\uc744 \ucd94\uac00\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n
3.1. \uacb0\uce21\uce58 \ucc98\ub9ac<\/h3>\n
\uacb0\uce21\uce58\ub294 \uba38\uc2e0\ub7ec\ub2dd \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \uc800\ud558\uc2dc\ud0ac \uc218 \uc788\uc73c\ubbc0\ub85c \uc2e0\uc911\ud558\uac8c \ucc98\ub9ac\ud574\uc57c \ud569\ub2c8\ub2e4. \uac00\uc7a5 \uc77c\ubc18\uc801\uc778 \ubc29\ubc95\uc740 \uacb0\uce21\uce58\ub97c \ud3c9\uade0 \ub610\ub294 \uc911\uc704\uc218\ub85c \ub300\uccb4\ud558\ub294 \uac83\uc785\ub2c8\ub2e4.<\/p>\n
df.fillna(method='ffill', inplace=True)  # \uacb0\uce21\uce58\ub97c \uc774\uc804 \uac12\uc73c\ub85c \ub300\uccb4<\/code><\/pre>\n
3.2. \ub370\uc774\ud130 \uc815\uaddc\ud654<\/h3>\n
\ub370\uc774\ud130\uc758 \ubc94\uc704\ub97c \uc870\uc815\ud558\uc5ec \ubaa8\ub378\uc758 \uc218\ub834 \uc18d\ub3c4\ub97c \ub192\uc774\ub294 \uac83\ub3c4 \uc911\uc694\ud569\ub2c8\ub2e4. \ub370\uc774\ud130 \uc815\uaddc\ud654\ub294 \uc77c\ubc18\uc801\uc73c\ub85c Min-Max \uc2a4\ucf00\uc77c\ub9c1 \ub610\ub294 Z-\uc2a4\ucf54\uc5b4 \uc815\uaddc\ud654\ub97c \ud1b5\ud574 \uc218\ud589\ub429\ub2c8\ub2e4.<\/p>\n
from sklearn.preprocessing import MinMaxScaler\n\nscaler = MinMaxScaler()\nscaled_data = scaler.fit_transform(df[['Close']])  # \uc885\uac00 \ub370\uc774\ud130 \uc815\uaddc\ud654<\/code><\/pre>\n
4. \uc2e0\uacbd\ub9dd \ubaa8\ub378 \uad6c\ucd95<\/h2>\n
\uc774\uc81c \uc804\ucc98\ub9ac\ub41c \ub370\uc774\ud130\ub97c \ubc14\ud0d5\uc73c\ub85c \uc2e0\uacbd\ub9dd \ubaa8\ub378\uc744 \uad6c\ucd95\ud569\ub2c8\ub2e4. \uc5ec\uae30\uc11c\ub294 Keras\uc640 TensorFlow \ub77c\uc774\ube0c\ub7ec\ub9ac\ub97c \uc0ac\uc6a9\ud558\uc5ec \uac04\ub2e8\ud55c \ub2e4\uce35 \ud37c\uc149\ud2b8\ub860(MLP) \ubaa8\ub378\uc744 \uc0dd\uc131\ud574\ubcf4\uaca0\uc2b5\ub2c8\ub2e4.<\/p>\n
4.1. \ubaa8\ub378 \uc124\uacc4<\/h3>\n
\ubaa8\ub378\uc740 \uc5ec\ub7ec \uac1c\uc758 \ub808\uc774\uc5b4\ub85c \uad6c\uc131\ub418\uba70, \uac01 \ub808\uc774\uc5b4\ub294 \ud65c\uc131\ud654 \ud568\uc218\ub97c \uc0ac\uc6a9\ud558\uc5ec \ube44\uc120\ud615\uc131\uc744 \ucd94\uac00\ud569\ub2c8\ub2e4. \uc544\ub798\ub294 \uae30\ubcf8\uc801\uc778 \ubaa8\ub378\uc758 \uc124\uacc4 \uc608\uc81c\uc785\ub2c8\ub2e4.<\/p>\n
from keras.models import Sequential\nfrom keras.layers import Dense\n\nmodel = Sequential()\nmodel.add(Dense(64, activation='relu', input_dim=X_train.shape[1]))  # \uccab \ubc88\uc9f8 \uc740\ub2c9\uce35\nmodel.add(Dense(32, activation='relu'))  # \ub450 \ubc88\uc9f8 \uc740\ub2c9\uce35\nmodel.add(Dense(1, activation='linear'))  # \ucd9c\ub825\uce35 (\ud68c\uadc0 \ubb38\uc81c\uc774\ubbc0\ub85c \uc120\ud615 \ud65c\uc131\ud654 \ud568\uc218 \uc0ac\uc6a9)<\/code><\/pre>\n
4.2. \ubaa8\ub378 \ucef4\ud30c\uc77c<\/h3>\n
\ubaa8\ub378\uc744 \ucef4\ud30c\uc77c\ud560 \ub54c\ub294 \uc190\uc2e4 \ud568\uc218\uc640 \ucd5c\uc801\ud654 \uc54c\uace0\ub9ac\uc998\uc744 \uc9c0\uc815\ud574\uc57c \ud569\ub2c8\ub2e4. \ud68c\uadc0 \ubb38\uc81c\uc5d0\uc11c\ub294 \uc77c\ubc18\uc801\uc73c\ub85c \ud3c9\uade0\uc81c\uacf1\uc624\ucc28(MSE)\uac00 \uc0ac\uc6a9\ub429\ub2c8\ub2e4.<\/p>\n
model.compile(loss='mean_squared_error', optimizer='adam')<\/code><\/pre>\n
5. \ubaa8\ub378 \ud6c8\ub828<\/h2>\n
\ubaa8\ub378\uc744 \ud6c8\ub828\uc2dc\ud0a4\uae30 \uc704\ud574\uc11c\ub294 \ud6c8\ub828 \ub370\uc774\ud130\ub97c \ud559\uc2b5\ud558\uace0 \uac80\uc99d \ub370\uc774\ud130\ub97c \ud1b5\ud574 \uc131\ub2a5\uc744 \ud3c9\uac00\ud574\uc57c \ud569\ub2c8\ub2e4. \uc77c\ubc18\uc801\uc73c\ub85c \ud6c8\ub828 \ub370\uc774\ud130\uc640 \uac80\uc99d \ub370\uc774\ud130\ub97c 80:20\uc758 \ube44\uc728\ub85c \ub098\ub204\uc5b4 \uc0ac\uc6a9\ud569\ub2c8\ub2e4.<\/p>\n
history = model.fit(X_train, y_train, epochs=100, batch_size=32, validation_split=0.2)<\/code><\/pre>\n
6. \ubaa8\ub378 \ud3c9\uac00<\/h2>\n
\ud6c8\ub828\uc774 \uc644\ub8cc\ub41c \ud6c4, \ud14c\uc2a4\ud2b8 \ub370\uc774\ud130\ub97c \uc0ac\uc6a9\ud558\uc5ec \ubaa8\ub378\uc758 \uc131\ub2a5\uc744 \ud3c9\uac00\ud569\ub2c8\ub2e4. RMSE(root mean square error)\uc640 \uac19\uc740 \uc9c0\ud45c\ub97c \uc0ac\uc6a9\ud558\uc5ec \ubaa8\ub378\uc758 \uc608\uce21 \ub2a5\ub825\uc744 \ud310\ub2e8\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n
from sklearn.metrics import mean_squared_error\nimport numpy as np\n\ny_pred = model.predict(X_test)\nrmse = np.sqrt(mean_squared_error(y_test, y_pred))\nprint(f'RMSE: {rmse}')  # RMSE \ucd9c\ub825<\/code><\/pre>\n
7. \uc2e4\uc81c \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c\uc5d0 \uc801\uc6a9<\/h2>\n
\ubaa8\ub378\uc758 \uc131\ub2a5\uc774 \ucda9\ubd84\ud788 \uc88b\ub2e4\uba74, \uc774\ub97c \uc2e4\uc81c \uac70\ub798\uc5d0 \uc801\uc6a9\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4. \uadf8\ub7ec\ub098 \uc2e4\uc81c \ud2b8\ub808\uc774\ub529 \uc2dc\uc2a4\ud15c\uc744 \uad6c\ucd95\ud558\ub294 \uac83\uc740 \uc0c1\ub2f9\ud55c \ub09c\uc774\ub3c4\ub97c \ub3d9\ubc18\ud558\uba70, \ub9ac\uc2a4\ud06c \uad00\ub9ac \ubc0f \uc2ac\ub9ac\ud53c\uc9c0\uc640 \uac19\uc740 \uc5ec\ub7ec \uc694\uc18c\ub97c \uace0\ub824\ud574\uc57c \ud569\ub2c8\ub2e4.<\/p>\n
7.1. \ub9ac\uc2a4\ud06c \uad00\ub9ac<\/h3>\n
\ub9ac\uc2a4\ud06c \uad00\ub9ac\ub294 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc5d0\uc11c \uac00\uc7a5 \uc911\uc694\ud55c \uc694\uc18c \uc911 \ud558\ub098\uc785\ub2c8\ub2e4. \ud1b5\uc0c1\uc801\uc73c\ub85c \ud3ec\uc9c0\uc158 \ud06c\uae30, \uc190\uc808\ub9e4 \ubc0f \uc774\uc775 \uc2e4\ud604 \uc218\uce58\ub97c \uc124\uc815\ud558\uc5ec \ub9ac\uc2a4\ud06c\ub97c \uad00\ub9ac\ud569\ub2c8\ub2e4.<\/p>\n
7.2. \uc8fc\ubb38 \uc2e4\ud589<\/h3>\n
\uc8fc\ubb38 \uc2e4\ud589 \uacfc\uc815\uc5d0\uc11c\ub294 Broker API\ub97c \ud65c\uc6a9\ud558\uc5ec \uc2e4\uc81c \uc2dc\uc7a5\uc5d0\uc11c \uc8fc\ubb38\uc744 \ubc1c\uc1a1\ud569\ub2c8\ub2e4. ‘ccxt’\uc640 \uac19\uc740 \ub77c\uc774\ube0c\ub7ec\ub9ac\ub97c \uc0ac\uc6a9\ud558\uba74 \ub2e4\uc591\ud55c \uac70\ub798\uc18c\uc758 API\uc640 \uc27d\uac8c \uc5f0\uacb0\ud560 \uc218 \uc788\uc2b5\ub2c8\ub2e4.<\/p>\n
import ccxt\n\nexchange = ccxt.binance()  # \uac70\ub798\uc18c \uc124\uc815\norder = exchange.create_market_order('BTC\/USDT', 'buy', amount)  # \uc2dc\uc7a5\uac00 \uc8fc\ubb38 \uc2e4\ud589<\/code><\/pre>\n
\uacb0\ub860<\/h2>\n
\uc774\ubc88 \uae00\uc5d0\uc11c\ub294 \uba38\uc2e0\ub7ec\ub2dd \ubc0f \ub525\ub7ec\ub2dd \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc744 \uc704\ud574 \ud30c\uc774\uc36c\uc744 \uc0ac\uc6a9\ud558\uc5ec \uc2e0\uacbd\ub9dd \ubaa8\ub378\uc744 \ucc98\uc74c\ubd80\ud130 \ub05d\uae4c\uc9c0 \uad6c\ucd95\ud558\ub294 \ubc29\ubc95\uc744 \uc54c\uc544\ubcf4\uc558\uc2b5\ub2c8\ub2e4. \ub370\uc774\ud130 \uc218\uc9d1, \uc804\ucc98\ub9ac, \ubaa8\ub378 \uad6c\ucd95 \ubc0f \ud6c8\ub828, \ud3c9\uac00 \ubc0f \uc2e4\uc81c \ud2b8\ub808\uc774\ub529\uc5d0 \uc774\ub974\uae30\uae4c\uc9c0 \ubaa8\ub4e0 \uacfc\uc815\uc744 \uc124\uba85\ud558\uc600\uc2b5\ub2c8\ub2e4. \uc774\ub7ec\ud55c \ud504\ub85c\uc138\uc2a4\ub97c \ud1b5\ud574 \uc54c\uace0\ub9ac\uc998 \ud2b8\ub808\uc774\ub529\uc758 \uae30\ucd08\ub97c \ub2e4\uc9c0\uace0, \ub354 \ub098\uc544\uac00 \uc790\uc2e0\ub9cc\uc758 \uac70\ub798 \uc804\ub7b5\uc744 \uc9c4\ud654\uc2dc\ud0ac \uc218 \uc788\ub294 \uae30\ud68c\ub97c \uc81c\uacf5\ud558\uae38 \ubc14\ub78d\ub2c8\ub2e4.<\/p>\n
\ucc38\uace0\ubb38\ud5cc<\/h2>\n