Improve structure and readability for FCNHead

[edwardyehuang]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

The first conv in the previous implementation is unnecessary since the input channels can be determined when passing parameters.

Modification

The first conv is combined into the loops. Thus, the readability and structure are improved..

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
3. If the modification has potential influence on downstream projects, this PR should be tested with downstream projects, like MMDet or MMDet3D.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[codecov]

Codecov Report

Base: 89.14% // Head: 89.08% // Decreases project coverage by -0.05% ⚠️

Coverage data is based on head (f4ea3ff) compared to base (6c746fa).
Patch coverage: 100.00% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2142      +/-   ##
==========================================
- Coverage   89.14%   89.08%   -0.06%     
==========================================
  Files         145      145              
  Lines        8716     8670      -46     
  Branches     1471     1383      -88     
==========================================
- Hits         7770     7724      -46     
  Misses        704      704              
  Partials      242      242              

Flag	Coverage Δ
unittests	89.08% <100.00%> (-0.06%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
mmseg/models/decode_heads/fcn_head.py	100.00% <100.00%> (ø)
mmseg/models/segmentors/base.py	56.03% <0.00%> (-3.17%)	⬇️
mmseg/core/hook/wandblogger_hook.py	15.71% <0.00%> (-1.77%)	⬇️
mmseg/models/decode_heads/cascade_decode_head.py	92.30% <0.00%> (-0.55%)	⬇️
mmseg/models/necks/fpn.py	57.14% <0.00%> (-0.55%)	⬇️
mmseg/ops/encoding.py	91.66% <0.00%> (-0.44%)	⬇️
mmseg/models/backbones/resnest.py	83.72% <0.00%> (-0.25%)	⬇️
mmseg/models/losses/focal_loss.py	75.24% <0.00%> (-0.25%)	⬇️
mmseg/models/backbones/vit.py	90.64% <0.00%> (-0.22%)	⬇️
mmseg/datasets/cityscapes.py	85.00% <0.00%> (-0.19%)	⬇️
... and 13 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.