I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)
主要原因，训练使用内存或显存不足：
1、修改：Batch大小。
2、修改每个加载数据大小
3、减少训练网络复杂程度，减少训练网络涉及到得参数数量，减少训练网络的步骤等。
https://stackoverflow.com/questions/63886762/
tensorflow
-none-of-the-
mlir
-
optimization
-
pass
es-are-
enab
led
-re
gis
tere
d-1
The I at the beginning of the message means this is an informational log message.
tensorflow
前面的I表示的是信息日志消息，不是错误。
MLIR
is being used as
cuda跟
tensorflow
-gpu版本没有冲突，只要使用
tensorflow
-gpu，就会报错：None of the
MLIR
Optimization
Pass
es are
enab
led
(re
gis
tere
d 2)就是程序刚开始运行就这样结束了可能的原因：
tensorflow
版本和
tensorflow
-gpu版本两个版本有代差，而
tensorflow
默认选择版本高的版本来计算。
解决办法
：卸载
tensorflow
版本，conda list可能查看不出来，直接命令行卸载
参考链接：http://
使用
tensorflow
，点击程序运行，原来的提示和现在的稍有不同。
None of the
MLIR
Optimization
Pass
es are
enab
led
(re
gis
tere
d 2)
原来是re
gis
tere
d 1
layout fai
led
: Invalid argument: Size of values 0 does not match size of permutation 4 @ fanin shape inRefiner/tf.where/SelectV2_1-2-Transpo
2023-07-13 14:44:21.978453: 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: AVX AVX2
To
enab
le them in other operations, rebuild
TensorFlow
with the appropriate compiler flags.
2023-07-13 14:44:22.835121: I
tensorflow
/core/common_runtime/gpu/gpu_device.
cc
:1510] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 3475 MB memory: -> device: 0, name: NVIDIA GeForce RTX 3060 Laptop GPU, pci bus id: 0000:01:00.0, compute ca
pa
bility: 8.6
2023-07-13 14:44:22.954245: I
tensorflow
/compiler/
mlir
/
mlir
_graph_
optimization
_
pass
.
cc
:
185
] None of the
MLIR
Optimization
Pass
es are
enab
led
(re
gis
tere
d 2)
pyswarms.single.global_best: 0%| |0/502023-06-12 09:15:36.800789: 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: AVX AVX2 To
enab
le them in other operations, rebuild
TensorFlow
with the appropriate compiler flags. 2023-06-12 09:15:37.280722: I
tensorflow
/core/common_runtime/gpu/gpu_device.
cc
:1510] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 9426 MB memory: -> device: 0, name: NVIDIA GeForce RTX 3080 Ti, pci bus id: 0000:01:00.0, compute ca
pa
bility: 8.6 2023-06-12 09:15:37.633600: I
tensorflow
/compiler/
mlir
/
mlir
_graph_
optimization
_
pass
.
cc
:
185
] None of the
MLIR
Optimization
Pass
es are
enab
led
(re
gis
tere
d 2) 2023-06-12 09:15:38.355549: F .\
tensorflow
/core/util/gpu_launch_config.h:129] Check fai
led
: work_element_count > 0 (0 vs. 0)
2022-01-02 18:41:16.826148: I
tensorflow
/compiler/
mlir
/
mlir
_graph_
optimization
_
pass
.
cc
:
185
] None of the
MLIR
Optimization
Pass
es are
enab
led
(re
gis
tere
d 2)
Traceback (most recent call last):
File "E:/Code/PyCharm/
TensorFlow
学习/Keras/自定义评估指标.py", line 59,.
Mac M1 关于“conda安装的
tensorflow
报错”解决方法问题描述导致问题的原因问题解决
最近发现之前在Mac m1安装的
tensorflow
无法使用了，这里的无法使用不是指无法import之类的错误，而是使用相关模块时代码报错，报错如下（小伙伴们可以参考）：
2021-09-03 17:05:21.327506: I
tensorflow
/compiler/
mlir
/
mlir
_graph_
optimization
_
pass
.
cc
:116] None of the
MLIR
optim
LaserNet README批注LaserNet in
Tensorflow
2.0环境配置Instructions:What works:What's still not done:Example Results
LaserNet in
Tensorflow
2.0
In this project I’ve done my best to implement the work from Uber ATG on LaserNet. LaserNet is a 3D object detector for
INFO:
tensorflow
:Done running local_init_op.
I0518 10:08:16.107131 139741801011008 session_manager.py:534] Done running local_init_op.
INFO:
tensorflow
:Calling checkpoint listeners before saving checkpoint 298000...
optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)解决办法
pytorch RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available
pygame 飞机大战 精灵AttributeError: ‘HeroPlane‘ object has no attribute ‘image‘解决方法
2401_84088696:
清华大学环境镜像使用方法
洒去犹能化碧涛:
optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)解决办法
吾生也有呀:
esp32 CMT130-V1.0 PS 240*240屏幕显示动态图片的实验
CSDN-Ada助手:
optimization_pass.cc:185] None of the MLIR Optimization Passes are enabled (registered 2)解决办法
callmeleige:
ERROR: Could not find a version that satisfies the requirement notebook (from versions: none) 解决办法
jupyter安装完成后运行代码没有反应
