Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add comments #34

Open
wants to merge 23 commits into
base: master
Choose a base branch
from
Open

Add comments #34

wants to merge 23 commits into from

Conversation

nekhtiari
Copy link

Add some extra commnets.

Rodrigo Almeida and others added 23 commits July 9, 2020 11:04
Add test infrastructure
* ignore pycharm idea file

* Add img ids for S2L1C and S2L2A.
* Intial refactoring of training script.

* Rename create_random file.

* Create a data_utils for data handeling.

* Refactor create_patches to utlize data_utils.

* Use lowercase naming

* Refactor s2_tiles_superres to utlize data_utils class

* Fix some of the linting errors

* Add unittests for data_utils.

* Add two more unittests

* Update the naming of files.

* Use typing more defined parameters of the method
* Freeze keras version to be compatible with tf1

* minor refactoring

* Revert naming of saved patches.

* Uncomment the commented line

* Add tests for inference

* Black update and fix some liniting issues.
* Stub out patches tests

* Fix shape of dataset

* Explicitly pass patch size and border

* Update defaults to working params

* Add aditional asserts

* Add multiple tests

* Skip long running tests

* Add blockutils to requirements

* Skip too long tests

* Add fstring linter

* Fix unused vars

* Disable unary operator

* Add description

* Use get_patches function

* Add description

* Use get_patches

* Import patches from context

* Add get_random_patches methods separate from save method

* Fix del

* Add helper methods

* Apply patch utils

* Remove hardcoded patch sizes

* Add typing

* Remove unused variable

* Add typing
* Fix import and data_path

* Fix path to xml file

* Add data_name to class

* Handle unever array case in downsizing

* Add data_name instance var

* Format

* Format

* Add generate patches script

* Add logger

* Add sys

* Add md file for patch gen

* Add test imgs ids for L1C

* Add test images L2A

* Add test commands

* Remove xtra -

* Remove duplicated test_data

* remove /

* Fix linter issues

* Fix duplicated code

* Disable tuple unbalance error

* Disable type for var input_shape
* Remove package pinning

* Add missing reqs

* Use input path to fetch data

* Add CLI to create val train

* Add glob

* Add os lib

* Use args.path

* Return args

* Format

* Change import order

* Set channels first in layer

* Use distributed strategy for multiple gpus

* Add scope to inference

* Format

* Add skip inference if no GPU available
* Add copied over AESR model

* Simplify model creation, add tests

* Add AESR model option

* format

* Fix linter

* Add run_60 to val_patch
* Add new AESR model example (20 - 10m)

* load AESR model from file

* rename model file

* Remove deep argument

* Add another placeholder  for 60m
* Add quality metrics.

* Rename the filename

* Add evaluation script

* use load_model mechanism

* Use image-similarity-measures package.

* Remove redundant var

* Remove pdb
* Make sure patch size is the same

* Add all instructions for patch generation

* Add adpated rmse

* Add comparison with bicubic

* fix bug

* add bic flag

* Add eval code

* Format

* Add scale

* Add all metrics

* Fix linter

* Add one pixel to size of d20 due to change in patch generation

* format

* Fix out array size

* Fix linter
* Add srcnn, rednet and resnet to model zoo

* Fix linter issues

* Add choices in command line

* Add imports

* Replace PreLu with ReLu in ResNetSR
* Add uiq adapted metric

* Parametrize metrics comming from image-similarity-metrics
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant