AI入魔之路
Rex~
Make it work -> Make it right -> Make it fast
展开
-
Only one model should be specified in tuneGrid with no resampling
起因R语言使用train函数训练模型,由于不想用交叉验证,所以可以用在trainControl(method = “none”),但是会报错 Only one model should be specified in tuneGrid with no resampling原始代码> train(bbbDescr, logBBB, method = "earth", trControl = trainControl(method = "none"))运行报错 Onl原创 2021-09-10 22:39:44 · 652 阅读 · 3 评论 -
ModuleNotFoundError: No module named ‘keras.applications.resnet50‘
报错ModuleNotFoundError: No module named ‘keras.applications.resnet50’ImportError: cannot import name ‘Sequence’ from ‘keras.utils’(/usr/local/lib/python3.7/dist-packages/keras/utils/init.pyImportError: cannot import name ‘to_categorical’ from ‘kera原创 2021-09-09 12:08:50 · 10959 阅读 · 0 评论 -
不降级解决ModuleNotFoundError: No module named ‘tensorflow.contrib‘
问题在TensorFlow2.x版本已经不能使用contrib包import tensorflow.contrib.slim as slim使用该包时报错如下 ModuleNotFoundError: No module named ‘tensorflow.contrib’解决参考链接用如下语句下载镜像包pip install --upgrade tf_slim将import tensorflow.contrib.slim as slim改为import tf_slim as sl原创 2021-01-19 09:14:26 · 43832 阅读 · 197 评论