search

facebookresearch / diffq

DiffQ performs differentiable quantization using pseudo quantization noise. It can automatically tune the number of bits used per weight or group of weights, in order to achieve a given trade-off between model size and accuracy.
Other
234 stars 14 forks source link

require 'override' keyword #3

Open xieyi4650 opened 3 years ago

xieyi4650 commented 3 years ago

🐛 Bug Report

run cd examples/cifar/ and run flowing readme, but got the flowing error:

envs/diffq/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:389: UserWarning: In config.yaml: Invalid over riding of hydra/job_logging: Default list overrides requires 'override' keyword. See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

change examples/cifar/conf/config.yaml from   1 defaults:   2   - hydra/job_logging: colorlog   3   - hydra/hydra_logging: colorlog

into

  1 defaults:   2   - override hydra/job_logging: colorlog   3   - override hydra/hydra_logging: colorlog can fix the issue

To Reproduce

(Write your steps here:)

pip install . make examples cd examples/cifar pip install -r requirements ./train.py db.name=cifar100 model=mobilenet quant.bits=3 quant.qat=True

Expected behavior

(Write what you thought would happen.)

Actual Behavior

(Write what happened. Add screenshots, if applicable.)

Your Environment

adefossez commented 3 years ago

Sorry about that, this was due to an update in the Hydra dependency since the release. Can you try uninstalling hydra and reinstalling version < 1.1:

pip uninstall hydra_core
pip install hydra_core<1.1
xieyi4650 commented 3 years ago

@adefossez Thanks. I have fix it by adding 'override' keyword in config.yaml create issue is just a reminder

adefossez commented 3 years ago

I wouldn't advise doing so. The way Hydra parses configs has largely changed in version 1.1, and it is likely the configuration will not have the same meaning if you use hydra_core=1.1. In order to reproduce the experiments from the paper, please use a virutalenv with hydra_core<1.1.