問題1：Could not load dynamic library cudart64_101.dll
解決方法：https://www.dll-files.com/download/1d7955354884a9058e89bb8ea34415c9/cudart64_101.dll.html?c=VElyVUtzT2Q0R2xhc2JNK1ZUa0Jvdz09
問題2：tensorflow-gpu版本使用時報錯，各種dll文件找不到
2020-06-15 20:30:54.310396: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cublas64_10.dll’; dlerror: cublas64_10.dll not found
2020-06-15 20:30:54.311457: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cufft64_10.dll’; dlerror: cufft64_10.dll not found
2020-06-15 20:30:54.312777: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘curand64_10.dll’; dlerror: curand64_10.dll not found
2020-06-15 20:30:54.314450: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cusolver64_10.dll’; dlerror: cusolver64_10.dll not found
2020-06-15 20:30:54.315749: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cusparse64_10.dll’; dlerror: cusparse64_10.dll not found
2020-06-15 20:30:54.316863: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library ‘cudnn64_7.dll’; dlerror: cudnn64_7.dll not found
2020-06-15 20:30:54.317064: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1592] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices…
解決方法：https://www.pianshen.com/article/37081444749/
問題3：RuntimeError: The Session graph is empty. Add operations to the graph before calling run().
解決方法：tf.compat.v1.disable_eager_execution()
無法執行sess.run()的原因是tensorflow版本不同導致的，tensorflow版本2.0無法兼容版本1.0.
import tensorflow as tf print(tf.__version__) tf.compat.v1.disable_eager_execution() # 保證sess.run()能夠正常運行 hello = tf.constant('hello tensorflow!') sess = tf.compat.v1.Session() # 版本2.0的函數 print(sess.run(hello))
問題4：AttributeError: module 'tensorflow' has no attribute 'mul'
解決方法：tf.mul已經在新版本中被移除，請使用 tf.multiply 代替
問題5：AttributeError: module 'tensorflow' has no attribute 'placeholder'
解決方法：
將
import tensorflow as tf
用以下代碼代替
import numpy as np #import tensorflow as tf import tensorflow.compat.v1 as tf tf.disable_v2_behavior()
問題5：Pycharm提示 Unresolved reference 的解決辦法
解決方法：
有時候a.py和b.py在一個目錄里面，但是在a.py種寫import b有時會提示Unresolved reference，Pycharm常見。
解決辦法是setting -> Project -> Project structure -> Source，點擊要加入的文件夾，
** 注意： 添加成功之后該文件見的圖標會編程藍色（加入之前是淡藍色中間一個圓圈）