IT_World

현 작업 상황을 기록하고 싶던 중 tensorboard, plt, scalar 고민하다가 log text file로 남기기로 결정했다. from tensorflow.python.client import device_lib device_lib.list_local_devices() 를 사용하면 CPU랑 GPU 모두 출력되야 하는데 cpu만 출력되는 문제 발생 device_lib.list_local_devices () no gpu print(device_lib.list_local_devices()) none log 정보를 파일로 저장하기 위해 코드를 돌리던 중 IndexError: list index out of range 에러발생 logger.INFO:OS : Linux logger.INFO:OS Versi..

How to Visualize Filters and Feature Maps in Convolutional Neural Networks Deep learning neural networks are generally opaque, meaning that although they can make useful and skillful predictions, it is not […] machinelearningmastery.com [Deep learning]컨볼 루션 신경망에서 필터 및 기능 맵을 시각화하는 방법 -2- machinelearningmastery.com/how-to-visualize-filters-and-feature-maps-in-convolutional-neural-networks/ How to ..

machinelearningmastery.com/how-to-visualize-filters-and-feature-maps-in-convolutional-neural-networks/ How to Visualize Filters and Feature Maps in Convolutional Neural Networks Deep learning neural networks are generally opaque, meaning that although they can make useful and skillful predictions, it is not […] machinelearningmastery.com niniit.tistory.com/12 [Deep learning]컨볼 루션 신경망에서 필터 및 기능 맵..

machinelearningmastery.com/how-to-visualize-filters-and-feature-maps-in-convolutional-neural-networks/ How to Visualize Filters and Feature Maps in Convolutional Neural Networks Deep learning neural networks are generally opaque, meaning that although they can make useful and skillful predictions, it is not […] machinelearningmastery.com 블로그를 참조하여 컨볼 루션 신경망에서 특정 필터에 대한 시각화를 개발하는 방법 컨볼 루션 신경망에서 특..

지난 예측 프로그램을 가지고 종가 예측해볼 것이다. import numpy as np import pandas as pd kospi = np.load('/kospi.npy') samsung = np.load('./samsung.npy') print(kospi) print(samsung) print(kospi.shape) print(samsung.shape) def split_xy5(dataset, time_steps, y_column): x, y = list(), list() for i in range(len(dataset)): x_end_number = i + time_steps y_end_number = x_end_number + y_column if y_end_number > len(dataset)..