个人中心
文章发布
退出
作者热门文章
- html - 出于某种原因，IE8 对我的 Sass 文件中继承的 html5 CSS 不友好？
- JMeter 在响应断言中使用 span 标签的问题
- html - 在 :hover and :active? 上具有不同效果的 CSS 动画
- html - 相对于居中的 html 内容固定的 CSS 重复背景？
28
4
我正在使用 ubuntu 20.04 并安装了 anaconda。根据this instruction ，我通过 conda create -n tf tensorflow-gpu
在安装必要的软件包如 scikit-learn 和 tqdm 之后，我尝试运行以下代码
## Global variables
BATCH_SIZE = 2
EPOCHS = 50
CATGORICAL = 10
PATIENCE = 10
INPUT_SHAPE = (32, 32, 3)
TARGET_SIZE = (INPUT_SHAPE[0], INPUT_SHAPE[1])
TRAIN_SIZE = 0.8
VAL_SPLIT = 0.2
STUDENT_CNN_LR = 1e-5
LR_FACTOR = 0.4
LR_PATIENCE = 3
MODEL_NAME = 'tftest.hdf5'
### Setup ###
import numpy as np
import tensorflow as tf
from glob import *
import numpy as np # linear algebra
import os
import tensorflow as tf
from sklearn.model_selection import train_test_split, StratifiedKFold, learning_curve
from sklearn.preprocessing import MinMaxScaler, StandardScaler, scale
from sklearn.metrics import roc_auc_score, confusion_matrix, accuracy_score, classification_report
from tensorflow import keras
from tensorflow.keras.applications.resnet50 import ResNet50
from tensorflow.keras.layers import Activation, Dropout, Flatten, Dense, GlobalMaxPooling2D, BatchNormalization, Input, Conv2D, MaxPool2D, GlobalAveragePooling2D
from tensorflow.keras.callbacks import ModelCheckpoint
from tensorflow.keras import metrics, layers, Sequential
from tensorflow.keras.optimizers import Adam
from tensorflow.keras.models import Model
from tensorflow.keras.losses import binary_crossentropy, categorical_crossentropy, SparseCategoricalCrossentropy
from tensorflow.keras.callbacks import ModelCheckpoint, LearningRateScheduler, EarlyStopping, ReduceLROnPlateau
from tqdm import *
from scipy.stats import norm, rankdata
### Load the dataset ###
(X_train, y_train), (X_test, y_test) = tf.keras.datasets.mnist.load_data()
# expand new axis, channel axis
X_train = np.expand_dims(X_train, axis=-1)
X_test = np.expand_dims(X_test, axis=-1)
# [optional]: we may need 3 channel (instead of 1)
X_train = np.repeat(X_train, 3, axis=-1)
X_test = np.repeat(X_test, 3, axis=-1)
# resize the input shape , i.e. old shape: 28, new shape: 32 (cifar is 32 already)
X_train = tf.image.resize(X_train, TARGET_SIZE) # if we want to resize
X_test = tf.image.resize(X_test, TARGET_SIZE) # if we want to resize
# one hot
y_train = tf.keras.utils.to_categorical(y_train, num_classes=CATGORICAL)
y_test = tf.keras.utils.to_categorical(y_test, num_classes=CATGORICAL)
data_augmentation_student = tf.keras.Sequential([
tf.keras.layers.experimental.preprocessing.Rescaling(scale=1/255),
layers.experimental.preprocessing.Normalization()
])
# images in cifar is small, use LeNet
student_CNN = keras.Sequential(
[
data_augmentation_student,
Conv2D(filters=6, kernel_size=(5,5), padding='valid', input_shape=INPUT_SHAPE, activation='tanh'),
MaxPool2D(pool_size=(2,2)),
Conv2D(filters=16, kernel_size=(5,5), padding='valid', activation='tanh'),
MaxPool2D(pool_size=(2,2)),
Flatten(),
Dense(120, activation='tanh'),
Dense(84, activation='tanh'),
Dense(CATGORICAL, activation='softmax')
],
name="student_CNN",
)
student_CNN.compile(optimizer=Adam(learning_rate=STUDENT_CNN_LR), loss='categorical_crossentropy', metrics=['accuracy'])
checkpoint = ModelCheckpoint(filepath=MODEL_NAME, monitor='val_accuracy', verbose=1,
save_best_only=True, mode='auto', save_weights_only = True)
reduceLROnPlat = ReduceLROnPlateau(monitor='val_accuracy', factor=LR_FACTOR, patience=LR_PATIENCE,
verbose=1, mode='auto', min_delta=0.0001)
early = EarlyStopping(monitor='val_accuracy', mode="auto", patience=PATIENCE)
callbacks_list = [checkpoint, reduceLROnPlat, early]
history = student_CNN.fit(X_train, y_train,
batch_size=BATCH_SIZE,
epochs=EPOCHS,
verbose=1,
callbacks=callbacks_list,
validation_split=VAL_SPLIT)
# 計算正確率
CNN_epoch = len(list(range(len(history.history['accuracy']))))
CNN_train_acc = history.history['accuracy'][-1]
CNN_val_acc = history.history['val_accuracy'][-1]
y_test_int = [np.where(r==1)[0][0] for r in y_test] # y_test is one hot, convert to integer labels
student_CNN.load_weights(MODEL_NAME)
student_prediction = student_CNN.predict(X_test, batch_size=BATCH_SIZE, verbose=1)
y_prediction_S = np.argmax(student_prediction, axis=1)
CNN_test_acc = accuracy_score(y_prediction_S, y_test_int)
我使用命令 python tftest.py 通过 GPU 运行代码。
021-09-30 17:44:53.913935: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1
2021-09-30 17:44:55.106074: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
2021-09-30 17:44:55.106578: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcuda.so.1
2021-09-30 17:44:55.134062: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:44:55.134520: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3060 computeCapability: 8.6
coreClock: 1.852GHz coreCount: 28 deviceMemorySize: 11.77GiB deviceMemoryBandwidth: 335.32GiB/s
2021-09-30 17:44:55.134534: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1
2021-09-30 17:44:55.135370: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10
2021-09-30 17:44:55.135393: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.10
2021-09-30 17:44:55.136296: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2021-09-30 17:44:55.136437: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2021-09-30 17:44:55.137322: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
2021-09-30 17:44:55.137817: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.10
2021-09-30 17:44:55.139781: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7
2021-09-30 17:44:55.139853: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:44:55.140351: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:44:55.140850: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
2021-09-30 17:44:55.141045: 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: SSE4.1 SSE4.2 AVX AVX2 AVX512F FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
2021-09-30 17:44:55.141798: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:44:55.142243: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1720] Found device 0 with properties:
pciBusID: 0000:01:00.0 name: NVIDIA GeForce RTX 3060 computeCapability: 8.6
coreClock: 1.852GHz coreCount: 28 deviceMemorySize: 11.77GiB deviceMemoryBandwidth: 335.32GiB/s
2021-09-30 17:44:55.142255: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1
2021-09-30 17:44:55.142264: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10
2021-09-30 17:44:55.142270: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublasLt.so.10
2021-09-30 17:44:55.142275: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcufft.so.10
2021-09-30 17:44:55.142280: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcurand.so.10
2021-09-30 17:44:55.142287: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusolver.so.10
2021-09-30 17:44:55.142293: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcusparse.so.10
2021-09-30 17:44:55.142299: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7
2021-09-30 17:44:55.142325: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:44:55.142767: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:44:55.143181: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1862] Adding visible gpu devices: 0
2021-09-30 17:44:55.143199: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudart.so.10.1
但是代码在这里卡了4分钟，然后就报错了
2021-09-30 17:48:33.025621: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1261] Device interconnect StreamExecutor with strength 1 edge matrix:
2021-09-30 17:48:33.025643: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1267] 0
2021-09-30 17:48:33.025647: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1280] 0: N
2021-09-30 17:48:33.025797: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:48:33.026250: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:48:33.026668: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:941] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-09-30 17:48:33.027075: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1406] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10758 MB memory) -> physical GPU (device: 0, name: NVIDIA GeForce RTX 3060, pci bus id: 0000:01:00.0, compute capability: 8.6)
2021-09-30 17:48:33.027256: I tensorflow/compiler/jit/xla_gpu_device.cc:99] Not creating XLA devices, tf_xla_enable_xla_devices not set
2021-09-30 17:48:33.038331: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 737280000 exceeds 10% of free system memory.
2021-09-30 17:48:33.669046: W tensorflow/core/framework/cpu_allocator_impl.cc:80] Allocation of 589824000 exceeds 10% of free system memory.
2021-09-30 17:48:33.838103: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
2021-09-30 17:48:33.856360: I tensorflow/core/platform/profile_utils/cpu_utils.cc:112] CPU Frequency: 2592000000 Hz
Epoch 1/50
2021-09-30 17:48:34.107817: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcublas.so.10
2021-09-30 17:49:42.529567: I tensorflow/stream_executor/platform/default/dso_loader.cc:49] Successfully opened dynamic library libcudnn.so.7
2021-09-30 18:00:47.141606: W tensorflow/stream_executor/gpu/asm_compiler.cc:63] Running ptxas --version returned 256
2021-09-30 18:00:47.200580: W tensorflow/stream_executor/gpu/redzone_allocator.cc:314] Internal: ptxas exited with non-zero error code 256, output:
Relying on driver to perform ptx compilation.
Modify $PATH to customize ptxas location.
This message will be only logged once.
2021-09-30 18:00:47.881462: W tensorflow/core/framework/op_kernel.cc:1763] OP_REQUIRES failed at cwise_op_gpu_base.cc:89 : Internal: Failed to load in-memory CUBIN: CUDA_ERROR_NO_BINARY_FOR_GPU: no kernel image is available for execution on the device
Traceback (most recent call last):
File "/home/benny/Documents/DL2021/tensorflow_test/tftest.py", line 91, in <module>
history = student_CNN.fit(X_train, y_train,
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/keras/engine/training.py", line 1100, in fit
tmp_logs = self.train_function(iterator)
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py", line 828, in __call__
result = self._call(*args, **kwds)
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/eager/def_function.py", line 888, in _call
return self._stateless_fn(*args, **kwds)
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/eager/function.py", line 2942, in __call__
return graph_function._call_flat(
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/eager/function.py", line 1918, in _call_flat
return self._build_call_outputs(self._inference_function.call(
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/eager/function.py", line 555, in call
outputs = execute.execute(
File "/home/benny/anaconda3/envs/tf/lib/python3.9/site-packages/tensorflow/python/eager/execute.py", line 59, in quick_execute
tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
tensorflow.python.framework.errors_impl.InternalError: Failed to load in-memory CUBIN: CUDA_ERROR_NO_BINARY_FOR_GPU: no kernel image is available for execution on the device
[[node student_CNN/conv2d/Tanh (defined at /Documents/DL2021/tensorflow_test/tftest.py:91) ]] [Op:__inference_train_function_953]
Function call stack:
train_function
我的 GPU 是 Geforce RTX3060，CUDA 版本是 114.4 我试过重建环境。这是我的一些包裹 list 。谢谢!
# packages in environment at /home/benny/anaconda3/envs/tf:
#
# Name Version Build Channel
_libgcc_mutex 0.1 main
_openmp_mutex 4.5 1_gnu
_tflow_select 2.1.0 gpu
absl-py 0.13.0 py39h06a4308_0
aiohttp 3.7.4.post0 py39h7f8727e_2
argon2-cffi 20.1.0 py39h27cfd23_1
astor 0.8.1 py39h06a4308_0
astunparse 1.6.3 py_0
async-timeout 3.0.1 py39h06a4308_0
async_generator 1.10 pyhd3eb1b0_0
attrs 21.2.0 pyhd3eb1b0_0
backcall 0.2.0 pyhd3eb1b0_0
blas 1.0 mkl
bleach 4.0.0 pyhd3eb1b0_0
blinker 1.4 py39h06a4308_0
brotlipy 0.7.0 py39h27cfd23_1003
c-ares 1.17.1 h27cfd23_0
ca-certificates 2021.7.5 h06a4308_1
cachetools 4.2.2 pyhd3eb1b0_0
certifi 2021.5.30 py39h06a4308_0
cffi 1.14.6 py39h400218f_0
chardet 4.0.0 py39h06a4308_1003
charset-normalizer 2.0.4 pyhd3eb1b0_0
click 8.0.1 pyhd3eb1b0_0
coverage 5.5 py39h27cfd23_2
cryptography 3.4.7 py39hd23ed53_0
cudatoolkit 10.1.243 h6bb024c_0
cudnn 7.6.5 cuda10.1_0
cupti 10.1.168 0
cython 0.29.24 py39h295c915_0
dataclasses 0.8 pyh6d0b6a4_7
dbus 1.13.18 hb2f20db_0
debugpy 1.4.1 py39h295c915_0
decorator 5.0.9 pyhd3eb1b0_0
defusedxml 0.7.1 pyhd3eb1b0_0
entrypoints 0.3 py39h06a4308_0
expat 2.4.1 h2531618_2
fontconfig 2.13.1 h6c09931_0
freetype 2.10.4 h5ab3b9f_0
gast 0.4.0 pyhd3eb1b0_0
glib 2.69.1 h5202010_0
google-auth 1.33.0 pyhd3eb1b0_0
google-auth-oauthlib 0.4.4 pyhd3eb1b0_0
google-pasta 0.2.0 pyhd3eb1b0_0
grpcio 1.36.1 py39h2157cd5_1
gst-plugins-base 1.14.0 h8213a91_2
gstreamer 1.14.0 h28cd5cc_2
h5py 2.10.0 py39hec9cf62_0
hdf5 1.10.6 hb1b8bf9_0
icu 58.2 he6710b0_3
idna 3.2 pyhd3eb1b0_0
importlib-metadata 4.8.1 py39h06a4308_0
importlib_metadata 4.8.1 hd3eb1b0_0
intel-openmp 2021.3.0 h06a4308_3350
ipykernel 6.2.0 py39h06a4308_1
ipython 7.27.0 py39hb070fc8_0
ipython_genutils 0.2.0 pyhd3eb1b0_1
ipywidgets 7.6.4 pyhd3eb1b0_0
jedi 0.18.0 py39h06a4308_1
jinja2 3.0.1 pyhd3eb1b0_0
joblib 1.0.1 pyhd8ed1ab_0 conda-forge
jpeg 9d h7f8727e_0
jsonschema 3.2.0 pyhd3eb1b0_2
jupyter 1.0.0 py39h06a4308_7
jupyter_client 7.0.1 pyhd3eb1b0_0
jupyter_console 6.4.0 pyhd3eb1b0_0
jupyter_core 4.7.1 py39h06a4308_0
jupyterlab_pygments 0.1.2 py_0
jupyterlab_widgets 1.0.0 pyhd3eb1b0_1
keras-preprocessing 1.1.2 pyhd3eb1b0_0
ld_impl_linux-64 2.35.1 h7274673_9
libblas 3.9.0 11_linux64_mkl conda-forge
libcblas 3.9.0 11_linux64_mkl conda-forge
libffi 3.3 he6710b0_2
libgcc-ng 9.3.0 h5101ec6_17
libgfortran-ng 7.5.0 ha8ba4b0_17
libgfortran4 7.5.0 ha8ba4b0_17
libgomp 9.3.0 h5101ec6_17
libpng 1.6.37 hbc83047_0
libprotobuf 3.17.2 h4ff587b_1
libsodium 1.0.18 h7b6447c_0
libstdcxx-ng 9.3.0 hd4cf53a_17
libuuid 1.0.3 h1bed415_2
libxcb 1.14 h7b6447c_0
libxml2 2.9.12 h03d6c58_0
markdown 3.3.4 py39h06a4308_0
markupsafe 2.0.1 py39h27cfd23_0
matplotlib-inline 0.1.2 pyhd3eb1b0_2
mistune 0.8.4 py39h27cfd23_1000
mkl 2021.3.0 h06a4308_520
mkl-service 2.4.0 py39h7f8727e_0
mkl_fft 1.3.0 py39h42c9631_2
mkl_random 1.2.2 py39h51133e4_0
multidict 5.1.0 py39h27cfd23_2
nbclient 0.5.3 pyhd3eb1b0_0
nbconvert 6.1.0 py39h06a4308_0
nbformat 5.1.3 pyhd3eb1b0_0
ncurses 6.2 he6710b0_1
nest-asyncio 1.5.1 pyhd3eb1b0_0
notebook 6.4.3 py39h06a4308_0
numpy 1.20.3 py39hf144106_0
numpy-base 1.20.3 py39h74d4b33_0
oauthlib 3.1.1 pyhd3eb1b0_0
openssl 1.1.1l h7f8727e_0
opt_einsum 3.3.0 pyhd3eb1b0_1
packaging 21.0 pyhd3eb1b0_0
pandocfilters 1.4.3 py39h06a4308_1
parso 0.8.2 pyhd3eb1b0_0
pcre 8.45 h295c915_0
pexpect 4.8.0 pyhd3eb1b0_3
pickleshare 0.7.5 pyhd3eb1b0_1003
pip 21.2.4 py37h06a4308_0
prometheus_client 0.11.0 pyhd3eb1b0_0
prompt-toolkit 3.0.17 pyhca03da5_0
prompt_toolkit 3.0.17 hd3eb1b0_0
protobuf 3.17.2 py39h295c915_0
ptyprocess 0.7.0 pyhd3eb1b0_2
pyasn1 0.4.8 pyhd3eb1b0_0
pyasn1-modules 0.2.8 py_0
pycparser 2.20 py_2
pygments 2.10.0 pyhd3eb1b0_0
pyjwt 2.1.0 py39h06a4308_0
pyopenssl 20.0.1 pyhd3eb1b0_1
pyparsing 2.4.7 pyhd3eb1b0_0
pyqt 5.9.2 py39h2531618_6
pyrsistent 0.18.0 py39h7f8727e_0
pysocks 1.7.1 py39h06a4308_0
python 3.9.7 h12debd9_1
python-dateutil 2.8.2 pyhd3eb1b0_0
python-flatbuffers 1.12 pyhd3eb1b0_0
python_abi 3.9 2_cp39 conda-forge
pyzmq 22.2.1 py39h295c915_1
qt 5.9.7 h5867ecd_1
qtconsole 5.1.1 pyhd3eb1b0_0
qtpy 1.10.0 pyhd3eb1b0_0
readline 8.1 h27cfd23_0
requests 2.26.0 pyhd3eb1b0_0
requests-oauthlib 1.3.0 py_0
rsa 4.7.2 pyhd3eb1b0_1
scikit-learn 0.24.2 py39h4dfa638_0 conda-forge
scipy 1.7.1 py39h292c36d_2
send2trash 1.8.0 pyhd3eb1b0_1
setuptools 58.0.4 py39h06a4308_0
sip 4.19.13 py39h2531618_0
six 1.16.0 pyhd3eb1b0_0
sqlite 3.36.0 hc218d9a_0
tensorboard 2.4.0 pyhc547734_0
tensorboard-plugin-wit 1.6.0 py_0
tensorflow 2.4.1 gpu_py39h8236f22_0
tensorflow-base 2.4.1 gpu_py39h29c2da4_0
tensorflow-estimator 2.6.0 pyh7b7c402_0
tensorflow-gpu 2.4.1 h30adc30_0
termcolor 1.1.0 py39h06a4308_1
terminado 0.9.4 py39h06a4308_0
testpath 0.5.0 pyhd3eb1b0_0
threadpoolctl 2.2.0 pyh8a188c0_0 conda-forge
tk 8.6.11 h1ccaba5_0
tornado 6.1 py39h27cfd23_0
tqdm 4.62.2 pyhd3eb1b0_1
traitlets 5.1.0 pyhd3eb1b0_0
typing-extensions 3.10.0.2 hd3eb1b0_0
typing_extensions 3.10.0.2 pyh06a4308_0
tzdata 2021a h5d7bf9c_0
urllib3 1.26.6 pyhd3eb1b0_1
wcwidth 0.2.5 pyhd3eb1b0_0
webencodings 0.5.1 py39h06a4308_1
werkzeug 2.0.1 pyhd3eb1b0_0
wheel 0.37.0 pyhd3eb1b0_1
widgetsnbextension 3.5.1 py39h06a4308_0
wrapt 1.12.1 py39he8ac12f_1
xz 5.2.5 h7b6447c_0
yarl 1.6.3 py39h27cfd23_0
zeromq 4.3.4 h2531618_0
zipp 3.5.0 pyhd3eb1b0_0
zlib 1.2.11 h7b6447c_3
最佳答案
这可能是您使用的软件包版本有问题。 Tensorflow official site表示 tensorflow 2.4 最多只兼容 python 3.8。
根据您的环境版本，您似乎正在使用 tensorflow 2.4.1 以及 python 3.9。
尝试将你的 python 版本降级到 3.8
关于python - Tensorflow 不能用 cudnn 的问题，我们在Stack Overflow上找到一个类似的问题： https://stackoverflow.com/questions/69391384/
28
4
0
在 Ubuntu 16.04 上按照标准命令安装 Lua 和其他依赖项: conda install lua=5.2 lua-science -c alexbw 我在行中遇到错误: require '
我正在尝试在 python 3 中创建机器学习。但后来我试图编译我的代码，我在 Cuda 10.0/cuDNN 7.5.0 中遇到了这个错误，有人可以帮我解决这个问题吗？ RTX 2080 我在:喀拉
我在安装了 CUDA 7.5 并正常工作的 Ubuntu 系统上使用 Python 和 IDE Pycharm。 我刚刚将 CUDNN 文件与我的常规 CUDA 安装合并。 现在，当我从 Tensor
我正在尝试配置 theano 以在我的 Windows 机器上使用 gpu。我已经将 .theanorc 设置为使用 device= gpu 但是当我运行一些应该使用 gpu 的代码时，我收到以下错误
当我为Windows编译caffe(64位，发行版，2013年，nvidia 750，opencv 3.1，cuDNN版本5.1)时，出现以下错误 "Error 13 error C1083: Can
我正在尝试理解和调试我的代码。我尝试使用在 GPU 上的 tf2.0/tf.keras 下开发的 CNN 模型进行预测，但得到了那些错误消息。 有人可以帮我修吗？ 这是我的环境配置 enviromen
我现在在 C++ 中使用 Cuda 有一段时间了，我想试试 cuDNN。我想直接使用 C++，但我大多只能找到基于不同平台(如 Caffè 或 TensorFlow)的示例和教程。这是否意味着我不能在
我不确定这是否是这个问题的正确堆栈交换，但这里是。 我已经安装了最新的 CUDA 驱动程序和 Tensorflow 1.14，但是当我尝试训练卷积层时，Tensorflow 说它找不到实现，因为它无法
我需要找到有关提供给 cudnnConvolutionForward、cudnnConvolutionBackwardData、cudnnConvolutionBackwardFilter 函数系列的
我在尝试运行前馈 torch.nn.Conv2d 时收到此消息，得到以下堆栈跟踪: ----------------------------------------------------------
我正在尝试加载 NSynth 权重，我正在使用 tf 版本 1.7.0 from magenta.models.nsynth import utils from magenta.models.nsyn
我搜索了很多地方，但我得到的只是如何安装它，而不是如何验证它是否已安装。我可以验证我的 NVIDIA 驱动程序是否已安装，并且 CUDA 是否已安装，但我不知道如何验证 CuDNN 是否已安装。非常感
库德恩:https://developer.nvidia.com/cudnn 我登录并完成 NVIDIA 希望您完成的所有任务；然而，当需要下载文件时，我似乎不知道如何通过 wget 和命令行来完成它
我编写了一个简单的应用程序来测试 cudnn rnn api 并检查我的理解是否正确； 代码是这样的， int layernum = 1; int batchnum = 32; int hiddenS
CuDNN 安装程序似乎在查找错误版本的 CUDA。我究竟做错了什么？完整的故事: Ubuntu 16.04 安装了两个版本的 CUDA，9.0 和 9.1。/usr/lib/cuda 链接到 9.1
我有以下基于 Theano example 的代码: from theano import function, config, shared, sandbox import theano.tensor
我在Windows机器(Win10 Pro 64位，i7-7700，8GB内存，GTX-1060-6GB)中使用cupy和Spyder3.3.6和Python 3.7.5。 cupy、chainer、
我正在运行example Keras 的 kaggle_otto_nn.py，后端为 theano。 在下面的打印输出中，第 5 行，有这样的内容: CNMeM is enabled with ini
cuDNN 安装手册说 ALL PLATFORMS Extract the cuDNN archive to a directory of your choice, referred to below
我正在使用 ubuntu 20.04 并安装了 anaconda。根据this instruction ，我通过 conda create -n tf tensorflow-gpu 创建一个环境 在安
我是一名优秀的程序员,十分优秀！