\u4e0a\u8a18Tips \u304c\u898b\u3064\u304b\u3063\u305f\u306e\u3067\u8a66\u3057\u305f\u3053\u3068\u3092\u30e1\u30e2\u3057\u307e\u3059\u3002<\/p>\n\n\n\n
\u30a8\u30c7\u30a3\u30b7\u30e7\u30f3 Windows10 Pro
\u30d0\u30fc\u30b8\u30e7\u30f3\u30001903
\u30de\u30b6\u30fc\u30dc\u30fc\u30c9 Asrock Z370 Pro4
CPU\u3000\u3000\u3000\u3000Intel Core i7 8700K
\u30e1\u30e2\u30ea\u3000\u3000\u300016.0GB
GPU\u3000\u3000\u3000\u3000Nvidia Quadro P2000
<\/p>\n\n\n\n
\u307e\u305a\u306f\u4e0b\u8a18\u624b\u9806\u306b\u3066Anaconda Prompt\u3092\u7ba1\u7406\u8005\u3068\u3057\u3066\u5b9f\u884c\u3057\u3066Anaconda\u3092\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3002\uff08\u53c2\u8003\uff1aWindows\u74b0\u5883\u3067\u306eAnaconda\u306e\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8<\/a>\uff09<\/p>\n\n\n <\/p>\n\n\n\n Microsoft Visual Studio 2019 Community \u7121\u6599\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9 \u304b\u3089C++ \u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb<\/p>\n\n\n\n <\/p>\n\n\n\n \u7d9a\u3044\u3066Nvidia CUDA toolkit10.1 \u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3001\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb \u30d0\u30fc\u30b8\u30e7\u30f3\u306f\u6700\u65b0\u306e10.2 \u3060\u3068dll\u30a8\u30e9\u30fc\u304c\u3067\u305f\u306e\u3067\u3001\u3072\u3068\u3064\u4e0b\u306e\u30b0\u30ec\u30fc\u30c9\u3067\u3088\u304b\u3063\u305f\u307f\u305f\u3044\u3067\u3059\u3002<\/p>\n\n\n\n <\/p>\n\n\n\n https:\/\/developer.nvidia.com\/rdp\/cudnn-download<\/a> \u304b\u3089 cuDNN v7.6.5 [November 5th, 2019] for CUDA 10.1\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3001\u89e3\u51cd\u3057\u3066 <\/p>\n\n\n\n <\/p>\n\n\n\n bin,include, lib \u5404\u30d5\u30a9\u30eb\u30c0\u3092 C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v10.1 \u306b\u4e0a\u66f8\u304d\u30b3\u30d4\u30fc\u3002<\/p>\n\n\n\n <\/p>\n\n\n\n Anaconda Navigator \u306e[Environments]\u304b\u3089\u4eee\u60f3\u74b0\u5883\u3092[Create]\u3001tensorflow \u3068\u5165\u529b\u3057\u3066 [Create]<\/p>\n\n\n\n [Not Installed] \u304b\u3089anaconda \u95a2\u9023\u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u691c\u7d22\u3057\u3066\u3001[Apply] \u3059\u308b\u3068\u4eee\u60f3\u74b0\u5883\u306banaconda \u306e\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u304c\u5b8c\u4e86\u3057\u307e\u3059\u3002<\/p>\n\n\n\n Anaconda Prompt (tensorflow) \u3092\u30b9\u30bf\u30fc\u30c8\u30e1\u30cb\u30e5\u30fc\u304b\u3089\u7ba1\u7406\u8005\u3068\u3057\u3066\u8d77\u52d5\u3057\u3066tensorflow-gpu, keras \u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3002<\/p>\n\n\n\n \u8ffd\u8a18\uff09Anaconda Navigator \u306eEnvironments \u304b\u3089GUI\u3067tensorflow\u95a2\u9023\u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u3082\u3067\u304d\u307e\u3059\u3051\u3069\u30d0\u30fc\u30b8\u30e7\u30f3\u9055\u3044\u3067\u30a8\u30e9\u30fc\u304c\u3067\u308b\u306e\u3067\u3059\u3079\u3066\u4fe1\u983c\u306f\u3067\u304d\u307e\u305b\u3093\u3002<\/p>\n\n\n \u6700\u521d\u3001pip install keras \u3057\u305f\u306e\u3067\u3059\u304c\u3001AtrributeError \u304c\u51fa\u305f\u306e\u3067\u4e00\u65e6\u3001pip uninstall keras \u3057\u3066\u518d\u5ea6\u3001pip install -U \u30aa\u30d7\u30b7\u30e7\u30f3\u3059\u308b\u3068\u4e0a\u624b\u304f\u884c\u304d\u307e\u3057\u305f\u3002\uff08\u307e\u3050\u308c\u5f53\u305f\u308a\u3067\u3057\u3087\u3046\u304b\uff1f\uff09\u30b9\u30bf\u30fc\u30c8\u30e1\u30cb\u30e5\u30fc\u304b\u3089Jupyter Notebook (tensorflow)\u3092\u8d77\u52d5\u3057\u3066\u30d0\u30fc\u30b8\u30e7\u30f3\u78ba\u8a8d\u3002<\/p>\n \u3081\u3067\u305f\u304fUsing TensorFlow backend. \u304c\u8868\u793a\u3055\u308c\u307e\u3057\u305f\u3002<\/p>\n\n\n\n TensorFlow 2.x \u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u304b\u3089TensorFlow 1.x \u306e\u30b3\u30fc\u30c9\u3092\u5b9f\u884c\u3057\u305f\u3044\u3068\u304d\u306f<\/p>\n\n\n\n \u3068\u3059\u308b\u3068\u30cf\u30c3\u30d4\u30fc\u306b\u306a\u308c\u308b\u3088\u3046\u3067\u3057\u305f\u3002\uff08\u4e0b\u8a18\u30b5\u30f3\u30d7\u30eb\u30b3\u30fc\u30c9\u8ee2\u8a18\uff09<\/p>\n\n\n \u7a81\u7136\u3067\u3059\u304cAnaconda \u306b\u30c7\u30a3\u30fc\u30d7\u30e9\u30fc\u30cb\u30f3\u30b0\u306e\u30e9\u30a4\u30d6\u30e9\u30ea\u30fcTensorFlow \u74b0\u5883\u3092\u69cb\u7bc9\u3057\u305f\u3044\u3068\u601d\u3044\u3001Google\u5148\u751f\u306b\u8a0a\u3044\u305f\u3068\u3053\u308d \u521d\u5fc3\u8005\u304cGPU\u642d\u8f09Windows10\u306bPython + Anaconda + T … \u7d9a\u304d\u3092\u8aad\u3080 TensorFlow2.1.0 \u74b0\u5883\u69cb\u7bc9(Windows10)<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[98,104,44],"tags":[],"class_list":["post-6778","post","type-post","status-publish","format-standard","hentry","category-python","category-tensorflow","category-windows10"],"_links":{"self":[{"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6778","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=6778"}],"version-history":[{"count":11,"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6778\/revisions"}],"predecessor-version":[{"id":6806,"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6778\/revisions\/6806"}],"wp:attachment":[{"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6778"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6778"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.stuffy.site\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6778"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Anaconda\u672c\u4f53\u3092\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3059\u308b\n> conda update -n base conda\n\nAnaconda\u3059\u3079\u3066\u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3059\u308b\n> conda update --all\n\nAnaconda\u306ePython\u3092\u30a2\u30c3\u30d7\u30c7\u30fc\u30c8\u3059\u308b\n> conda update python
Python\u306e\u30d0\u30fc\u30b8\u30e7\u30f3\u78ba\u8a8d
> python --version
Python 3.7.6\n\n<\/pre>\n\n\n
https:\/\/visualstudio.microsoft.com\/ja\/downloads\/<\/a><\/p>\n\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-251-1024x572.png\")
<\/a><\/figure>\n\n\n\n
https:\/\/developer.nvidia.com\/cuda-toolkit-archive<\/a><\/p>\n\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-252.png\")
<\/a><\/figure>\n\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-254-1-1024x675.png\")
<\/a><\/figure>\n\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-253-1-1024x614.png\")
<\/a><\/figure>\n\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-256.png\")
<\/figcaption><\/figure>\n\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-255-1024x555.png\")
<\/a><\/figure>\n\n\n\n> pip install tensorflow-gpu\n> pip install -U tensorflow keras<\/pre>\n
pip list | find \"tensor\"
\ntensorboard 2.1.0 \ntensorflow 2.1.0 \ntensorflow-estimator 2.1.0 \ntensorflow-gpu 2.1.0 \ntensorflow-gpu-estimator 2.1.0\n
import keras\nprint(keras.__version__)
\nUsing TensorFlow backend.\n2.3.1<\/pre>\n\n\n![\"\"](\"https:\/\/www.stuffy.site\/wordpress\/wp-content\/uploads\/2020\/01\/2020-01-258-2.png\")
<\/a><\/figure>\n\n\n\nimport tensorflow.compat.v1 as tf\ntf.disable_v2_behavior()<\/pre>\n\n\n\n
import tensorflow.compat.v1 as tf\nimport numpy as np\n\ntf.disable_v2_behavior()\n# Create 100 phony x, y data points in NumPy, y = x * 0.1 + 0.3\nx_data = np.random.rand(100).astype(np.float32)\ny_data = x_data * 0.1 + 0.3\n\n# Try to find values for W and b that compute y_data = W * x_data + b\n# (We know that W should be 0.1 and b 0.3, but TensorFlow will\n# figure that out for us.)\nW = tf.Variable(tf.random.uniform([1], -1.0, 1.0))\nb = tf.Variable(tf.zeros([1]))\ny = W * x_data + b\n\n# Minimize the mean squared errors.\nloss = tf.reduce_mean(tf.square(y - y_data))\noptimizer = tf.train.GradientDescentOptimizer(0.5)\ntrain = optimizer.minimize(loss)\n\n# Before starting, initialize the variables. We will 'run' this first.\ninit = tf.global_variables_initializer()\n\n# Launch the graph.\nsess = tf.Session()\nsess.run(init)\n\n# Fit the line.\nfor step in range(201):\n sess.run(train)\n if step % 20 == 0:\n print(step, sess.run(W), sess.run(b))\n\n# Learns best fit is W: [0.1], b: [0.3]<\/pre>","protected":false},"excerpt":{"rendered":"