我正在为 MNIST 数据集做一个简单的 CNN 算法，但遇到了这样的问题。我知道在 tensorflow 中可能有数百万个原因导致这种情况发生，但如果有人知道该怎么做，我真的很高兴。
所以这是输出：
2021-06-06 00:48:30.854140: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudart64_110.dll
2021-06-06 00:48:34.334009: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2021-06-06 00:48:34.335490: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library nvcuda.dll
2021-06-06 00:48:35.401858: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce GTX 1650 computeCapability: 7.5
coreClock: 1.56GHz coreCount: 16 deviceMemorySize: 4.00GiB deviceMemoryBandwidth: 119.24GiB/s
2021-06-06 00:48:35.402113: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudart64_110.dll
2021-06-06 00:48:35.409854: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublas64_11.dll
2021-06-06 00:48:35.410015: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublasLt64_11.dll
2021-06-06 00:48:35.414546: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cufft64_10.dll
2021-06-06 00:48:35.416442: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library curand64_10.dll
2021-06-06 00:48:35.423427: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusolver64_10.dll
2021-06-06 00:48:35.427473: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusparse64_11.dll
2021-06-06 00:48:35.428431: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudnn64_8.dll
2021-06-06 00:48:35.428643: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
2021-06-06 00:48:35.429282: 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.
2021-06-06 00:48:35.430229: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: GeForce GTX 1650 computeCapability: 7.5
coreClock: 1.56GHz coreCount: 16 deviceMemorySize: 4.00GiB deviceMemoryBandwidth: 119.24GiB/s
2021-06-06 00:48:35.430583: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudart64_110.dll
2021-06-06 00:48:35.430727: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublas64_11.dll
2021-06-06 00:48:35.430854: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublasLt64_11.dll
2021-06-06 00:48:35.430981: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cufft64_10.dll
2021-06-06 00:48:35.431101: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library curand64_10.dll
2021-06-06 00:48:35.431228: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusolver64_10.dll
2021-06-06 00:48:35.431346: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cusparse64_11.dll
2021-06-06 00:48:35.431522: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudnn64_8.dll
2021-06-06 00:48:35.432695: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
2021-06-06 00:48:36.075767: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
2021-06-06 00:48:36.075939: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267] 0
2021-06-06 00:48:36.076035: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0: N
2021-06-06 00:48:36.076312: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 2903 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1650, pci bus id: 0000:01:00.0, compute capability: 7.5)
2021-06-06 00:48:36.077384: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
conv_1 (Conv2D) (None, 28, 28, 32) 832
_________________________________________________________________
pool_1 (MaxPooling2D) (None, 14, 14, 32) 0
_________________________________________________________________
conv_2 (Conv2D) (None, 14, 14, 64) 51264
_________________________________________________________________
flatten (Flatten) (None, 12544) 0
_________________________________________________________________
fc_1 (Dense) (None, 1024) 12846080
_________________________________________________________________
dropout (Dropout) (None, 1024) 0
_________________________________________________________________
fc_2 (Dense) (None, 10) 10250
=================================================================
Total params: 12,908,426
Trainable params: 12,908,426
Non-trainable params: 0
_________________________________________________________________
Epoch 1/80
2021-06-06 00:48:36.946333: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
2021-06-06 00:48:37.020602: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublas64_11.dll
2021-06-06 00:48:37.638591: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cublasLt64_11.dll
2021-06-06 00:48:38.228603: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library cudnn64_8.dll
Process finished with exit code -1073740791 (0xC0000409)
正如你所看到的，我的算法的第一个 epoch 开始了，然后程序突然中断，给出了这个输出。 tf 版本：2.4.1
答案 0 :(得分：0)
将 tf 更新到 2.5.0 是解决方案...