Les corpus CREMMA désignent un ensemble de corpus de vérité de terrain produit dans le cadre du projet CREMMA (Consortium pour la Reconnaissance des Écritures Manuscrites des Matériaux Anciens).
| """ | |
| Here is a dice loss for keras which is smoothed to approximate a linear (L1) loss. | |
| It ranges from 1 to 0 (no error), and returns results similar to binary crossentropy | |
| """ | |
| # define custom loss and metric functions | |
| from keras import backend as K | |
| def dice_coef(y_true, y_pred, smooth=1): |
| # This script was created to convert a directory full | |
| # of markdown files into rst equivalents. It uses | |
| # pandoc to do the conversion. | |
| # | |
| # 1. Install pandoc from http://johnmacfarlane.net/pandoc/ | |
| # 2. Copy this script into the directory containing the .md files | |
| # 3. Ensure that the script has execute permissions | |
| # 4. Run the script | |
| # | |
| # By default this will keep the original .md file |