CHANfiG aims to make your configuration easier.
There are tons of configurable parameters in training a Machine Learning model.
To configure all these parameters, researchers usually need to write gigantic config files, sometimes even thousands of lines.
Most of the configs are just replicates of the default arguments of certain functions, resulting in many unnecessary declarations.
It is also very hard to alter the configurations.
One needs to navigate and open the right configuration file, make changes, save and exit.
These had wasted an uncountable1 amount of precious time
and are no doubt a crime.
argparse could relieve the burdens to some extent.
However, it takes a lot of work to make it compatible with existing config files, and its lack of nesting limits its potential.
CHANfiG would like to make a change.
You just type the alternations in the command line, and leave everything else to CHANfiG.
CHANfiG is highly inspired by YACS.
Different from the paradigm of YACS(
your code + a YACS config for experiment E (+ external dependencies + hardware + other nuisance terms ...) = reproducible experiment E),
The paradigm of CHANfiG is:
your code + command line arguments (+ optional CHANfiG config + external dependencies + hardware + other nuisance terms ...) = reproducible experiment E (+ optional CHANfiG config for experiment E)
A Config is basically a nested dict structure.
However, the default Python dict is hard to manipulate.
The only way to access a dict member is through
dict['name'], which is obviously extremely complex.
Even worse, if the dict is nested like a config, member access could be something like
Enough is enough, it is time to make a change.
We need attribute-style access, and we need it now.
dict.parent.children.name is all you need.
Although there have been some other works that achieve a similar functionality of attribute-style access to dict members.
Their Config object either uses a separate dict to store information from attribute-style access (EasyDict), which may lead to inconsistency between attribute-style access and dict-style access;
or reuse the existing
__dict__ and redirect dict-style access (ml_collections), which may result in confliction between attributes and members of Config.
To overcome the aforementioned limitations, we inherit the Python built-in
dict to create
We also introduce
Variable to allow sharing a value across multiple places, and
ConfigParser to parse command line arguments.
FlatDict improves the default
dict in 3 aspects.
FlatDict incorporates a
merge method that allows you to merge a
Iterable, or a path to the
Different from built-in
merge assign values instead of replace, which makes it works better with
dict in Python is ordered since Python 3.7, but there isn’t a built-in method to help you sort a
sort to help you manage your dict.
FlatDict comes with
intersect, which makes it very easy to compare a
FlatDict with other
Iterable, or a path.
to method similar to PyTorch Tensors.
You can simply convert all member values of
FlatDict to a certain type or pass to a device in the same way.
FlatDict also integrates
xla) methods for easier access.
yamls methods to dump
FlatDict to a file or string.
It also provides
from_yamls methods to build a
FlatDict from a string or file.
FlatDict also includes
load methods which determines the type by its extension and dump/load
FlatDict to/from a file.
To facility the needs of default values, we incorporate
DefaultDict which accepts
default_factory and works just like a
Since most Configs are in a nested structure, we further propose a
all_items methods to allow iterating over the whole nested structure at once.
NestedDict also comes with
apply_ methods, which made it easier to manipulate the nested structure.
Config extends the functionality by supporting
defrost, and by adding a built-in
ConfigParser to pare command line arguments.
Config also has
default_factory=Config() by default for convenience.
Registry extends the
NestedDict and supports
build to help you register constructors and build objects from a
Have one value for multiple names at multiple places? We got you covered.
Just wrap the value with
Variable, and one alteration will be reflected everywhere.
Variable also supports
required to ensure the correctness of the value.
To make it even easier,
Variable also supports
help to provide a description when using
ArgumentParser and provides
parse_config to parse command line arguments.
CHANfiG has great backward compatibility with previous configs.
No matter if your old config is json or yaml, you could directly read from them.
And if you are using yacs, just replace
Config and enjoy all the additional benefits that CHANfiG provides.
Moreover, if you find a name in the config is too long for command-line, you could simply call
self.add_argument with proper
dest to use a shorter name in command-line, as you do with
All you need to do is just run a line:
You could also load a default configure file and make changes based on it:
Note, you must specify
config.parse(default_config='config') to correctly load the default config.
If you have made it dump current configurations, this should be in the written file:
Define the default arguments in function, put alterations in CLI, and leave the rest to CHANfiG.
Install the most recent stable version on pypi:
Install the latest version from source:
It works the way it should have worked.
CHANfiG is multi-licensed under the following licenses:
- The Unlicense
- GNU Affero General Public License v3.0 or later
- GNU General Public License v2.0 or later
- BSD 4-Clause “Original” or “Old” License
- MIT License
- Apache License 2.0
You can choose any (one or more) of these licenses if you use this work.
SPDX-License-Identifier: Unlicense OR AGPL-3.0-or-later OR GPL-2.0-or-later OR BSD-4-Clause OR MIT OR Apache-2.0
fun fact: time is always uncountable. ↩