clarified the size of smaps refiner

README.md

@@ -7,9 +7,9 @@ datasets:
 - fastMRI
 ---
 
-This model was used to achieve the 2nd highest submission in terms of PSNR on the fastMRI dataset (see https://fastmri.org/leaderboards/).
+This model was used to achieve the 3rd highest submission in terms of PSNR on the fastMRI dataset (see https://fastmri.org/leaderboards/).
 It is a base model for acceleration factor 4.
-The model uses 25 iterations and a medium MWCNN.
+The model uses 25 iterations and a medium MWCNN, and a big sensitivity maps refiner.
 
 ## Model description
 For more details, see https://arxiv.org/abs/2010.07290.
@@ -82,7 +82,7 @@ The training procedure is described in https://arxiv.org/abs/2010.07290.
 This section is WIP.
 
 ## Evaluation results
-On the fastMRI validation dataset
+On the fastMRI validation dataset, the same model with a smaller sensitivity maps refiner gives the following results for 30 validation volumes per contrast:
 
 | Contrast | T1     | T2     | FLAIR  | T1-POST |
 |----------|--------|--------|--------|---------|