mmcv/docs/utils.md

Utils

Config

Config class is used for manipulating config and config files. It supports loading configs from multiple file formats including python, json and yaml. It provides dict-like apis to get and set values.

Here is an example of the config file test.py.

a = 1
b = dict(b1=[0, 1, 2], b2=None)
c = (1, 2)
d = 'string'

To load and use configs

>>> cfg = Config.fromfile('test.py')
>>> print(cfg)
>>> dict(a=1,
...      b=dict(b1=[0, 1, 2], b2=None),
...      c=(1, 2),
...      d='string')

For all format configs, inheritance is supported. To reuse fields in other config files, specify _base_='./config_a.py' or a list of configs _base_=['./config_a.py', './config_b.py']. Here are 4 examples of config inheritance.

config_a.py

a = 1
b = dict(b1=[0, 1, 2], b2=None)

Inherit from base config without overlaped keys.

config_b.py

_base_ = './config_a.py'
c = (1, 2)
d = 'string'

>>> cfg = Config.fromfile('./config_b.py')
>>> print(cfg)
>>> dict(a=1,
...      b=dict(b1=[0, 1, 2], b2=None),
...      c=(1, 2),
...      d='string')

New fields in config_b.py are combined with old fields in config_a.py

Inherit from base config with overlaped keys.

config_c.py

_base_ = './config_a.py'
b = dict(b2=1)
c = (1, 2)

>>> cfg = Config.fromfile('./config_c.py')
>>> print(cfg)
>>> dict(a=1,
...      b=dict(b1=[0, 1, 2], b2=1),
...      c=(1, 2))

b.b2=None in config_a is replaced with b.b2=1 in config_c.py.

Inherit from base config with ignored fields.

config_d.py

_base_ = './config_a.py'
b = dict(_delete_=True, b2=None, b3=0.1)
c = (1, 2)

>>> cfg = Config.fromfile('./config_d.py')
>>> print(cfg)
>>> dict(a=1,
...      b=dict(b2=None, b3=0.1),
...      c=(1, 2))

You may also set _delete_=True to ignore some fields in base configs. All old keys b1, b2, b3 in b are replaced with new keys b2, b3.

Inherit from multiple base configs (the base configs should not contain the same keys).

config_e.py

c = (1, 2)
d = 'string'

config_f.py

_base_ = ['./config_a.py', './config_e.py']

>>> cfg = Config.fromfile('./config_f.py')
>>> print(cfg)
>>> dict(a=1,
...      b=dict(b1=[0, 1, 2], b2=None),
...      c=(1, 2),
...      d='string')

ProgressBar

If you want to apply a method to a list of items and track the progress, track_progress is a good choice. It will display a progress bar to tell the progress and ETA.

import mmcv

def func(item):
    # do something
    pass

tasks = [item_1, item_2, ..., item_n]

mmcv.track_progress(func, tasks)

The output is like the following.

There is another method track_parallel_progress, which wraps multiprocessing and progress visualization.

mmcv.track_parallel_progress(func, tasks, 8)  # 8 workers

If you want to iterate or enumerate a list of items and track the progress, track_iter_progress is a good choice. It will display a progress bar to tell the progress and ETA.

import mmcv

tasks = [item_1, item_2, ..., item_n]

for task in mmcv.track_iter_progress(tasks):
    # do something like print
    print(task)

for i, task in enumerate(mmcv.track_iter_progress(tasks)):
    # do something like print
    print(i)
    print(task)

Timer

It is convinient to compute the runtime of a code block with Timer.

import time

with mmcv.Timer():
    # simulate some code block
    time.sleep(1)

or try with since_start() and since_last_check(). This former can return the runtime since the timer starts and the latter will return the time since the last time checked.

timer = mmcv.Timer()
# code block 1 here
print(timer.since_start())
# code block 2 here
print(timer.since_last_check())
print(timer.since_start())