During fairseq training:
reason:
sacrableu wrong version.
terms of settlement:
pip install sacrebleu==1.5.1
Read More:
- Please note that the module “sacrableu” has no “compute blue” attribute
- [error] attributeerror: module ‘SIP’ has no attribute ‘setapi‘
- Attributeerror: module “Seaborn” has no attribute “lineplot”
- AttributeError：module“seaborn” has no attribute “lineplot”
- AttributeError:module ‘seaborn‘ has no attribute ‘tsplot‘
- Solving attributeerror: module ‘Matplotlib’ has no attribute ‘artist’
- Error reporting: attributeerror: module ‘******* has no attribute’ ******* and other problems are solved
- Solving attributeerror: module ‘Matplotlib’ has no attribute ‘plot‘
- AttributeError: module ‘keras.backend‘ has no attribute ‘eager‘
- AttributeError: module ‘onnxruntime‘ has no attribute ‘InferenceSession‘
- AttributeError: module ‘pkg_resources‘ has no attribute ‘declare_namespace‘
- AttributeError: module ‘os‘ has no attribute ‘fork‘
- AttributeError: module ‘scipy.misc’ has no attribute ‘imread’
- Attributeerror: ‘module’ object has no attribute ‘handlers’ — Python sub module import problem
- AttributeError: module ‘os’ has no attribute ‘mknod’
- Solving attributeerror: module ‘urllib’ has no attribute ‘request’
- AttributeError: module…ops‘ has no attribute ‘_TensorLike‘， ValueError: `updates` argument..eager
- Attributeerror: the solution of module ‘urllib’ has no attribute ‘quote’
- AttributeError: module ‘numpy‘ has no attribute ‘getbuffer‘
- AttributeError: module ‘torch’ has no attribute’_six’ [The problem is solved after restart]