qurator-spk/sbb_pixelwise_segmentation
1
0
Fork 0
mirror of https://github.com/qurator-spk/sbb_pixelwise_segmentation.git synced 2025-06-07 19:05:24 +02:00
Code Issues Releases Wiki Activity

Update README.md

Browse source
This commit is contained in:
vahidrezanezhad 2021-06-23 07:55:36 -04:00 • committed by GitHub
parent 040d3cfef2
commit e69846363b
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 10 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

17
README.md
View file

@ -39,12 +39,15 @@ If you have an image label with height and width of 10, for a binary case the fi
The train and evaluation folders should contain subfolders of images and labels. The train and evaluation folders should contain subfolders of images and labels.
The output folder should be an empty folder where the output model will be written to. The output folder should be an empty folder where the output model will be written to.
### Patches ### Parameter configuration
If you want to train your model with patches, the height and width of * patches: If you want to break input images into smaller patches (input size of the model) you need to set this parameter to ``true``. In the case that the model should see the image once, like page extraction, patches should be set to ``false``.
the patches should be defined and also the number of batches (how many patches * n_batch: Number of batches at each iteration.
should be seen by the model in each iteration). * n_classes: Number of classes. In the case of binary classification this should be 2.
* n_epochs: Number of epochs.
In the case that the model should see the image once, like page extraction, * input_height: This indicates the height of model's input.
patches should be set to ``false``. * input_width: This indicates the width of model's input.
* weight_decay: Weight decay of l2 regularization of model layers.
* augmentation: If you want to apply any kind of augmentation this parameter should first set to ``true``.
* flip_aug: If ``true``, different types of filp will applied on image. Type of flips is given by "flip_index" in train.py file.