Please make sure that this is a build/installation issue. As per our GitHub Policy, we only address code/doc bugs, performance issues, feature requests and build/installation issues on GitHub. tag:build_template

System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Windows 10
• Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device: N/A
• TensorFlow installed from (source or binary): binary
• TensorFlow version: 2.4.0
• Python version: 3.8.8
• Installed using virtualenv? pip? conda?: pip inside a virtualenv.
• Bazel version (if compiling from source): N/A
• GCC/Compiler version (if compiling from source): N/A
• CUDA/cuDNN version: v11.0.3_451.82 / cudnn-11.0-windows-x64-v8.0.4.30.zip
• GPU model and memory: 2080 8G -Nvidia driver version: 465.89

Describe the problem

A similar issue as to #46606, #43193, and #45055. When attempting to run python -c "import tensorflow as tf;print(tf.reduce_sum(tf.random.normal([1000, 1000])))", I get the following trace;

2021-05-02 16:52:00.618495: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found 2021-05-02 16:52:00.618653: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. 2021-05-02 16:52:02.101881: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set 2021-05-02 16:52:02.102613: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library nvcuda.dll 2021-05-02 16:52:02.130031: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties: pciBusID: 0000:2d:00.0 name: NVIDIA GeForce RTX 2080 computeCapability: 7.5 coreClock: 1.845GHz coreCount: 46 deviceMemorySize: 8.00GiB deviceMemoryBandwidth: 417.23GiB/s 2021-05-02 16:52:02.130365: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudart64_110.dll'; dlerror: cudart64_110.dll not found 2021-05-02 16:52:02.130609: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cublas64_11.dll'; dlerror: cublas64_11.dll not found 2021-05-02 16:52:02.130862: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cublasLt64_11.dll'; dlerror: cublasLt64_11.dll not found 2021-05-02 16:52:02.131045: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cufft64_10.dll'; dlerror: cufft64_10.dll not found 2021-05-02 16:52:02.131225: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'curand64_10.dll'; dlerror: curand64_10.dll not found 2021-05-02 16:52:02.131484: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cusolver64_10.dll'; dlerror: cusolver64_10.dll not found 2021-05-02 16:52:02.131796: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cusparse64_11.dll'; dlerror: cusparse64_11.dll not found 2021-05-02 16:52:02.132066: W tensorflow/stream_executor/platform/default/dso_loader.cc:60] Could not load dynamic library 'cudnn64_8.dll'; dlerror: cudnn64_8.dll not found 2021-05-02 16:52:02.132188: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1757] 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... 2021-05-02 16:52:02.132804: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 2021-05-02 16:52:02.133414: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix: 2021-05-02 16:52:02.133546: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267] 2021-05-02 16:52:02.133646: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set tf.Tensor(-115.45033, shape=(), dtype=float32)

Provide the exact sequence of commands / steps that you executed before running into the problem

I continue to get the above issue despite following the steps and suggestions in each of the other similar issues.

I tried adding the environmental variables through cmd, which wasn’t mirrored in the GUI so I added them there as well. This did not fix the issue.

I re-installed the Microsoft Visual C++ Redistributable for Visual Studio 2015, 2017, and 2019. This did not fix the issue.

I enabled long paths. This did not fix the issue.

I was already on a 64bit release of Python 3.8.8, which the venv I’m using was created with.

I’ve restarted the terminal and restarted my PC several times, which did not fix the issue.

I created the C:\tools\cuda folder and extracted the contents of the cuDNN there, and ensured that it was on PATH. This confuses me since it’s not mentioned as a necessity in the installation process of cuDNN itself. Either way, this did not fix the issue.

Any other info / logs deviceQuery running. All the items in PATH. The contents of the CUDA bin folder, showing the supposedly missing files exist.

Thank you in advance for your help.