added the pretrained model

imedslab · Jun 20, 2018 · fcc99af · fcc99af
1 parent c2de9d5
commit fcc99af
Show file tree

Hide file tree

Showing 18 changed files with 18 additions and 6 deletions.
diff --git a/.gitattributes b/.gitattributes
@@ -0,0 +1 @@
+*.pth filter=lfs diff=lfs merge=lfs -text
diff --git a/README.md b/README.md
@@ -13,8 +13,8 @@ Here we present the training codes and the pretrained models from each of our ex
 
 To train the networks, we used Ubuntu 14.04, CUDA 8.0 and CuDNN v.6. Below please find the other dependencies which need to be installed:
 
-* Python 3
-* pytorch with CUDA support
+* Python 3.6
+* pytorch < 0.4.0 with CUDA support
 * PIL
 * matplotlib
 * Jupyter Notebook (to work with attention maps)
@@ -31,17 +31,16 @@ However, you can run the codes as they are, just use the parameters fixed in the
 ## Attention maps examples
 Our model learns localized radiological findings as we imposed prior anatomical knowledge to teh network architecture. Here are some examples of attention maps and predictions (Kellgren-Lawrence grade 2 ground truth):
 
-<img src="https://github.com/lext/DeepKnee/blob/master/pics/15_2_R_1_1_1_3_1_0_own.jpg" width="260"/> <img src="https://github.com/lext/DeepKnee/blob/master/pics/235_2_R_3_3_0_0_1_1_own.jpg" width="260"/>  <img src="https://github.com/lext/DeepKnee/blob/master/pics/77_2_R_2_0_0_0_0_1_own.jpg" width="260"/> 
+<img src="https://github.com/lext/DeepKnee/blob/master/pics/15_2_R_1_1_1_3_1_0_own.jpg" width="260"/> <img src="https://github.com/lext/DeepKnee/blob/master/pics/235_2_R_3_3_0_0_1_1_own.jpg" width="260"/>  <img src="https://github.com/lext/DeepKnee/blob/master/pics/77_2_R_2_0_0_0_0_1_own.jpg" width="260"/>
 
-## TODO
+## What is in here
 
 - [x] Codes for the main experiements (Supplementary information of the article)
-- [ ] Pre-trained models
+- [x] Pre-trained models
 - [ ] Datasets generation scripts
 - [ ] MOST and OAI cohorts bounding box annotations
 
 
 ## License
 
 This code is freely available only for research purpuses.
-
diff --git a/snapshots_knee_grading/.DS_Store b/snapshots_knee_grading/.DS_Store
diff --git a/snapshots_knee_grading/2017_10_10_12_30_42/args.pkl b/snapshots_knee_grading/2017_10_10_12_30_42/args.pkl
diff --git a/snapshots_knee_grading/2017_10_10_12_30_42/epoch_64.pth b/snapshots_knee_grading/2017_10_10_12_30_42/epoch_64.pth
diff --git a/snapshots_knee_grading/2017_10_10_12_30_42/logs.npy b/snapshots_knee_grading/2017_10_10_12_30_42/logs.npy
diff --git a/snapshots_knee_grading/2017_10_10_12_30_46/args.pkl b/snapshots_knee_grading/2017_10_10_12_30_46/args.pkl
diff --git a/snapshots_knee_grading/2017_10_10_12_30_46/epoch_97.pth b/snapshots_knee_grading/2017_10_10_12_30_46/epoch_97.pth
diff --git a/snapshots_knee_grading/2017_10_10_12_30_46/logs.npy b/snapshots_knee_grading/2017_10_10_12_30_46/logs.npy
diff --git a/snapshots_knee_grading/2017_10_10_12_30_49/args.pkl b/snapshots_knee_grading/2017_10_10_12_30_49/args.pkl
diff --git a/snapshots_knee_grading/2017_10_10_12_30_49/epoch_96.pth b/snapshots_knee_grading/2017_10_10_12_30_49/epoch_96.pth
diff --git a/snapshots_knee_grading/2017_10_10_12_30_49/logs.npy b/snapshots_knee_grading/2017_10_10_12_30_49/logs.npy
diff --git a/snapshots_knee_grading/mean_std.npy b/snapshots_knee_grading/mean_std.npy
diff --git a/snapshots_knee_grading_resnet/.DS_Store b/snapshots_knee_grading_resnet/.DS_Store
diff --git a/snapshots_knee_grading_resnet/2017_10_17_13_10_14/args.pkl b/snapshots_knee_grading_resnet/2017_10_17_13_10_14/args.pkl
diff --git a/snapshots_knee_grading_resnet/2017_10_17_13_10_14/epoch_12.pth b/snapshots_knee_grading_resnet/2017_10_17_13_10_14/epoch_12.pth
diff --git a/snapshots_knee_grading_resnet/2017_10_17_13_10_14/logs.npy b/snapshots_knee_grading_resnet/2017_10_17_13_10_14/logs.npy
diff --git a/snapshots_knee_grading_resnet/mean_std.npy b/snapshots_knee_grading_resnet/mean_std.npy