Zhao Changmin
Zhao Changmin
Hey,I'm glad to inform you that your suggests really work!!! I write a python script to generate trainable dataset, sooner I will commit it for further usage.
Yes，I noticed that the model you trained is quite different between the official one，your trainning prototxts would help a lot，would you mind packing them and sending it to my email（[email protected]，thanks...
yes.you can see in model_train.py
remove the parallels config in params.py and related code in main.py
Copy that! It would be a good reference.
> hi, @dding3 can help to merge this PR?, find that conda user can not use by the original writing. hi, i think we have already met your need in...
By running mypy, you will see ``` (py37) arda@intern03:~/Project/BigDL/python/orca$ mypy --config-file test/mypy.ini src/bigdl/orca/data/utils.py src/bigdl/orca/data/ray_xshards.py Success: no issues found in 2 source files ``` if type checking passed.
``` def convert_imagenet_to_tf_records( raw_data_dir: str, output_dir: str) -> Tuple[List[str], List[str]]: ``` should return None https://github.com/intel-analytics/BigDL/pull/5299/files#diff-f7838e0bb04f8222e3c77ca2e67f75e96d88a986cf86dddbe41ebd22b2a713faR58
Add type hint to orca.data.transformer as well done
orca.data.tf.tf2_data done