Jamal Mohd-Yusof

Default_utils.py specifies ‘save_path’ as a variable for this purpose. I believe we decided against ‘save’ as a keyword because its intent is ambiguous (is it a boolean ‘save = True’,...

We also have output_dir which is used in some benchmarks. I believe the intent of this was to be used with the experiment directories (EXP00/RUN00 etc) to standardize paths when...

All the imports should be tensorflow.keras in the candle library and in the code. Have you looked at the p2b1 imports?

See, for example, p3b1. The line fpath = fetch_data(gParameters) is basically what you want to run separately from the 'run' command. We can modify fetch_file to allow a different base...

So probably a command like this should work for both tickets: python benchmark --dl_only --basedir='/scratch/candle/'

Whatever we choose for keywords we can make part of the standard parser, so just decide on ones that don't conflict with other standard (keras/neon/etc) keywords. --data_dir is fine (we...

How strict is the 'don't import Keras' restriction? We need to be able to read the default_model file to get data locations, as well as import the command line parser,...

Just curious what this means. We still haven't done the final Benchmark and candle_lib releases (I don't have permissions) but want to make sure things work properly

Is there anything in the performance callback that is Intel-specific? I may lift it into candle_lib unless there are objections (and after testing)

OK. This branch is still using /common instead of candle_lib so I will not try to merge it. We should try to fix the offload syntax so that xpu, gpu...