Dariusz Kajtoch
Dariusz Kajtoch
Thanks for response @riccardomurri! I am using google cloud heavily and I constantly experience overall resource restrictions (i can have many machines but total summarized resource cost cannot exceed fixed...
Thanks @ndmckinley! This is definitely an option. Anyway, do you consider multi-regional cluster in future releases? As far as the google cloud is concerned, I saw that you are using...
Do you include full model with both input and output layers (center and context word embeddings)?
thanks it worked!
The same question! What is the status of that feature? In the paper https://arxiv.org/pdf/1811.02084.pdf you mention "**Implementation of SPMD programming on CPU/GPU clusters**" (Future Work). Is the project dead? @adarob...
/assign @ark-kun
@googlebot I signed it!
@Ark-kun I changed return type to JsonObject and updated ml engine train test I build docker image locally and run some poc pipeline with single task to test if labels...
@Ark-kun I need to check it once again since I am passing `Dict` in tests and in pipeline so maybe there is no conversion. BTW in ml engine train yaml...
@Ark-kun sorry for such a delay. I did some tests with labels and it seems to work both with Dict and Json string. I added logger to `_train.py` in order...