\nSA\/DL Easy\u3092\u4f7f\u3046\u3068\u3001\u74b0\u5883\u69cb\u7bc9\u306e\u5fc5\u8981\u3082\u306a\u304f\u3001\u4ee5\u4e0b\u306e\u3088\u3046\u306aScript\u3082\u66f8\u304b\u305a\u306b\u3001Deep Learning\u306e\u4e16\u754c\u304c\u305f\u3084\u3059\u304f\u4f7f\u3048\u308b\u3088\u3046\u306b\u306a\u308a\u307e\u3059\u3002<\/p>\n
\u4e0b\u8a18\u3092\u81ea\u8eab\u306e\u30d1\u30bd\u30b3\u30f3\u3067\u5b9f\u884c\u3055\u308c\u308b\u5834\u5408\u306b\u306f\u3001Python\u306eScript\u3092\u4fdd\u5b58\u3057\u305f\u30d5\u30a9\u30eb\u30c0\u30fc\u5185\u306bpath\u304c\u901a\u3063\u3066\u3044\u308b\u3053\u3068\u3001\u4fdd\u5b58\u3057\u305f\u5165\u529b\u30c7\u30fc\u30bf\u306epath\u3001\u5b66\u7fd2\u7d50\u679c\u3068\u30c6\u30b9\u30c8\u7d50\u679c\u304c\u4fdd\u5b58\u3055\u308c\u3066\u3044\u308b\u30d5\u30a9\u30eb\u30c0\u30fc\u306epath\u306a\u3069\u3092\u9593\u9055\u3048\u306a\u3044\u3088\u3046\u306b\u6307\u5b9a\u3057\u3066\u304f\u3060\u3055\u3044\u3002<\/p>\n
———————————————————————————————
\n#\u3000LecChip\u306e\u30c7\u30fc\u30bf\u3092\u4f7f\u3063\u3066\u3001\u4f8b\u3048\u3070\u3001\u7cd6\u9396\u306e\u69cb\u9020\u3084\u7d30\u80de\u7a2e\u306e\u5224\u65ad\u3092\u5b66\u7fd2\u3055\u305b\u308b\u305f\u3081\u306eDeep Learning Python Script\u4f8b<\/p>\n
from __future__ import print_function
\nimport numpy as np
\nimport csv
\nimport pandas
\nfrom keras.datasets import mnist
\nfrom keras.models import Sequential
\nfrom keras.layers.core import Dense, Dropout, Activation
\nfrom keras.optimizers import RMSprop
\nfrom keras.utils import np_utils
\nfrom make_tensorboard import make_tensorboard<\/p>\n
np.random.seed(1671) # \u518d\u73fe\u6027\u3092\u826f\u304f\u3059\u308b\u305f\u3081\u306b<\/p>\n
# \u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u69cb\u6210\u3068\u5b66\u7fd2\u306e\u3055\u305b\u65b9
\nNB_EPOCH = 100 # \u4f55\u56de\u5b66\u7fd2\u3055\u305b\u308b\u304b\u3001\u9069\u5f53\u306b\u6c7a\u3081\u3066\u304f\u3060\u3055\u3044
\nBATCH_SIZE = 2\u3000\u3000# \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3092\u5e7e\u3064\u304b\u306e\u30b5\u30d6\u30bb\u30c3\u30c8\u306b\u5206\u3051\u308b
\nVERBOSE = 1
\nNB_CLASSES = 2 # \u6700\u7d42\u7684\u306a\u51fa\u529b\u6570
\nOPTIMIZER = RMSprop() # \u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc
\nN_HIDDEN = 45\u3000\u3000# \u96a0\u308c\u5c64\u306e\u30ce\u30fc\u30c9\u6570\u3001\u3053\u3053\u3067\u306fLecChip\u306e\u30ec\u30af\u30c1\u30f3\u6570\u306b\u5408\u308f\u305b\u306645\u3068\u3057\u3066\u3044\u308b
\nVALIDATION_SPLIT = 0.2 # \u5b66\u7fd2\u30c7\u30fc\u30bf\u306e\u5185\u3001\u4f55\u5272\u3092\u30c6\u30b9\u30c8\u30c7\u30fc\u30bf\u3068\u3057\u3066\u4f7f\u3046\u304b
\nDROPOUT = 0.3
\nLECTINS = 45<\/p>\n
def drop(df):
\nreturn df[pandas.to_numeric(df.iloc[:, 2], errors=’coerce’).notnull()]<\/p>\n
#\u3000\u30c7\u30fc\u30bf\u306f\u6700\u5927\u5024\u3092\uff11\u3068\u3059\u308b\u3088\u3046\u306b\u898f\u683c\u5316\u3059\u308b
\ndef normalize_column(d):
\ndmax = np.max(d)
\ndmin = np.min(d)
\nreturn (np.log10(d + 1.0) – np.log10(dmin + 1.0)) \/ \\
\n(np.log10(dmax + 1.0) – np.log10(dmin + 1.0))<\/p>\n
def normalize(data):
\nreturn np.apply_along_axis(normalize_column, 0, data)<\/p>\n
# \u5165\u529b\u3059\u308b\u30c7\u30fc\u30bf\u306fCSV\u30d5\u30a1\u30a4\u30eb\u306e\u5f62\u5f0f\u3068\u3059\u308b
\ndf1 = drop(pandas.read_csv(r’c:\\Users\\Masao\\Anaconda3\\DL_scripts\\cell.csv’)).reset_index(drop=True)
\nX_train = normalize(df1.iloc[:, 2:].astype(np.float64))
\nfamily_column = df1.iloc[:, 1]
\nfamily_list = sorted(list(set(family_column)))
\nY_train = np.array([family_list.index(f) for f in family_column])<\/p>\n
df2 = drop(pandas.read_csv(r’c:\\Users\\Masao\\Anaconda3\\DL_scripts\\cell_test.csv’)).reset_index(drop=True)
\nX_test = normalize(df2.iloc[:, 2:].astype(np.float64))
\nfamilyt_column = df2.iloc[:, 1]
\nfamilyt_list = sorted(list(set(familyt_column)))
\nY_test = np.array([familyt_list.index(f) for f in familyt_column])<\/p>\n
print(X_train.shape[0], ‘train samples’)
\nprint(X_test.shape[0], ‘test samples’)<\/p>\n
# convert class vectors to binary class matrices
\nY_train = np_utils.to_categorical(Y_train, NB_CLASSES)
\nY_test = np_utils.to_categorical(Y_test, NB_CLASSES)<\/p>\n
print(X_train)
\nprint(Y_train)
\nprint(X_test)
\nprint(Y_test)<\/p>\n
# \u30cb\u30e5\u30fc\u30e9\u30eb\u30cd\u30c3\u30c8\u30ef\u30fc\u30af\u306e\u5177\u4f53\u7684\u306a\u69cb\u6210\u4f8b
\n# \u96a0\u308c\u5c64\u306f\uff12\u5c64
\n# \u5165\u529b\u306fLecChip\u306e\u30c7\u30fc\u30bf\uff0845\u30ec\u30af\u30c1\u30f3\u3092\u4f7f\u7528\uff09
\n# \u6700\u7d42\u5c64\u306fsoftmax\u3067\u6d3b\u6027\u5316<\/p>\n
model = Sequential()
\nmodel.add(Dense(N_HIDDEN, input_shape=(LECTINS,)))
\nmodel.add(Activation(‘relu’))
\nmodel.add(Dropout(DROPOUT))
\nmodel.add(Dense(N_HIDDEN))
\nmodel.add(Activation(‘relu’))
\nmodel.add(Dropout(DROPOUT))
\nmodel.add(Dense(NB_CLASSES))
\nmodel.add(Activation(‘softmax’))
\nmodel.summary()<\/p>\n
#\u3000\u5b66\u7fd2\u3068\u30c6\u30b9\u30c8\u7d50\u679c\u72b6\u6cc1\u3092Tensorboard\u306b\u51fa\u529b\u3057\u3066\u53ef\u8996\u5316\u3055\u305b\u308b
\ncallbacks = [make_tensorboard(set_dir_name=’Glycan_Profile’)]<\/p>\n
model.compile(loss=’categorical_crossentropy’,
\noptimizer=OPTIMIZER,
\nmetrics=[‘accuracy’])<\/p>\n
model.fit(X_train, Y_train,
\nbatch_size=BATCH_SIZE, epochs=NB_EPOCH,
\ncallbacks=callbacks,
\nverbose=VERBOSE, validation_split=VALIDATION_SPLIT)<\/p>\n
score = model.evaluate(X_test, Y_test, verbose=VERBOSE)
\nprint(“\\nTest score:”, score[0])
\nprint(‘Test accuracy:’, score[1])<\/p>\n
————————————————————————————
\n# Tensorboard\u3092\u4f7f\u3046\u305f\u3081\u306ePython Script<\/p>\n
# -*- coding: utf-8 -*-
\nfrom __future__ import absolute_import
\nfrom __future__ import unicode_literals
\nfrom time import gmtime, strftime
\nfrom keras.callbacks import TensorBoard
\nimport os<\/p>\n
def make_tensorboard(set_dir_name=”):
\nymdt = strftime(“%a_%d_%b_%Y_%H_%M_%S”, gmtime())
\ndirectory_name = ymdt
\nlog_dir = set_dir_name + ‘_’ + directory_name
\nos.mkdir(log_dir)
\ntensorboard = TensorBoard(log_dir=log_dir, write_graph=True, )
\nreturn tensorboard<\/p>\n
————————————————————————————
\nTensorboard\u3092\u7528\u3044\u3066\u53ef\u8996\u5316\u3055\u305b\u308b\u306b\u306f\u3001make_tensorboard.py\u3092\u8d70\u3089\u305b\u305f\u3046\u3048\u3067\u3001
\n$ tensorboard –logdir=.\/Glycan_Profile_Mon_10_Feb_2025_23_06_26\uff08.\/\u30c7\u30fc\u30bf\u304c\u8a18\u9332\u3055\u308c\u3066\u3044\u308b\u30d5\u30a9\u30eb\u30c0\u30fc\uff09\u3092\u8d70\u3089\u305b
\nhttp:\/\/localhost:6006\/\u306b\u30a2\u30af\u30bb\u30b9\u3057\u307e\u3059\u3002<\/p>\n
(base) PS C:\\Users\\masao\\Anaconda3\\DL_Scripts> python make_tensorboard.py
\nUsing TensorFlow backend.
\n(base) PS C:\\Users\\masao\\Anaconda3\\DL_Scripts> tensorboard –logdir=.\/Glycan_Profile_Mon_10_Feb_2025_23_06_26
\nServing TensorBoard on localhost; to expose to the network, use a proxy or pass –bind_all
\nTensorBoard 2.0.2 at http:\/\/localhost:6006\/ (Press CTRL+C to quit)<\/p>\n
![\"\"](\"https:\/\/www.emukk.com\/WP\/wp-content\/uploads\/2025\/02\/tensorboard.jpg\")
<\/p>\n
—————————————————————————————————-
\nLecChip\u306e\u30c7\u30fc\u30bf\u306f\u3001\u4e0b\u8a18\u306e\u3088\u3046\u306aCSV\u5f62\u5f0f\u3068\u3057\u307e\u3059\u3002
\n\u5de6\u7aef\u304b\u3089\u3001\u30b5\u30f3\u30d7\u30eb\u540d\u3001\u30d5\u30a1\u30df\u30ea\u30fc\u540d\uff08\u6559\u5e2b\u30c7\u30fc\u30bf\u3068\u306a\u308a\u307e\u3059\uff09\u3001\u5404\u7a2e\u30ec\u30af\u30c1\u30f3\u306e\u6570\u5024\u304c\u4e26\u3093\u3067\u3044\u307e\u3059\u3002<\/p>\n
![\"\"](\"https:\/\/www.emukk.com\/WP\/wp-content\/uploads\/2025\/02\/LecChip-csv.jpg\")
<\/p>\n
—————————————————————————————————-
\n\u5b66\u7fd2\u304c\u7d42\u308f\u3063\u305f\u3089\u3001\u30e2\u30c7\u30eb\u306e\u4fdd\u5b58\u3092\u3057\u3066\u304a\u304f\u3079\u304d\u3067\u3057\u3087\u3046\u3002
\n\u30e2\u30c7\u30eb\u304c\u4fdd\u5b58\u3057\u3066\u3042\u308c\u3070\u5fa9\u5143\u3082\u3067\u304d\u307e\u3059\u3057\u3001\u672a\u77e5\u30c7\u30fc\u30bf\u3092\u4e0e\u3048\u3066\u4e88\u6e2c\u3055\u305b\u308b\u3053\u3068\u304c\u51fa\u6765\u307e\u3059\u3002
\n\u3053\u306e\u8fba\u308a\u306eScript\u306f\u5225\u9014\u66f8\u304f\u3053\u3068\u306b\u3057\u307e\u3059\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"
LecChip\uff08\u30ec\u30af\u30c1\u30f3\u30de\u30a4\u30af\u30ed\u30a2\u30ec\u30a4\uff09\u306e\u30c7\u30fc\u30bf\u3092\u4f7f\u7528\u3057\u3066\u3001\u7cd6\u9396\u306e\u69cb\u9020\u3001\u7d30\u80de\u7a2e\u306a\u3069\u3092\u5224\u5225\u3055\u305b\u308b\u305f\u3081\u306b\u306f\u3001Deep Learning\u3092\u4f7f\u7528\u3057\u3066\u6ca2\u5c71\u306e\u30c7\u30fc\u30bf\u3092\u6a5f\u68b0\u5b66\u7fd2\u3055\u305b\u308b\u65b9\u6cd5\u304c\u6709\u52b9\u3067\u3059\u3002 \u3053\u308c\u3092\u884c\u3046\u70ba\u306b\u5fc5\u8981\u306a\u4e8b\u524d\u6e96\u5099\u306f\u3001\u4ee5<\/p><\/div>\n