huajsj commented on a change in pull request #9108:
URL: https://github.com/apache/tvm/pull/9108#discussion_r722917455
##########
File path: python/tvm/contrib/pipeline_executor.py
##########
@@ -501,17 +538,18 @@ class PipelineExecutorFactoryModule(object):
"""
- def __init__(self, pipeline_mods, mods_config):
- mods, config = self.graph_executor_create(pipeline_mods, mods_config)
- assert (
- pipeline_executor_enabled()
- ), "Pipeline executor is not enabled. Please \
- re-build TVM with USE_PIPELINE_EXECUTOR=ON"
- pipeline_create = tvm._ffi.get_global_func(
+ def __init__(self, pipeline_libs=None, mods_config=None):
+ self.pipeline_libs = pipeline_libs
+ self.mods_config = mods_config
+ self.pipeline_create = tvm._ffi.get_global_func(
Review comment:
change allow_missing into False, then would throw a error when get
function failed.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
