Issue rykov8#2: Changed architecture a little to have dense layer aft…

…er pool6; Issue rykov8#6: If no box has iou with prior > 0.5 the best prior is assigned; Added some augmentation to generator in SSD_training.ipynb
pzhokhov · Nov 28, 2016 · 2363880 · 2363880
1 parent 6782258
commit 2363880
Show file tree

Hide file tree

Showing 5 changed files with 287 additions and 178 deletions.
diff --git a/README.md b/README.md
@@ -2,4 +2,4 @@
 For more details, please refer to [arXiv paper](http://arxiv.org/abs/1512.02325).
 For forward pass for 300x300 model, please, follow `SSD.ipynb` for examples. For training procedure for 300x300 model, please, follow `SSD_training.ipynb` for examples.
 
-Weights are ported from the original models and are available [here](https://mega.nz/#F!7RowVLCL!q3cEVRK9jyOSB9el3SssIA).
+Weights are ported from the original models and are available [here](https://mega.nz/#F!7RowVLCL!q3cEVRK9jyOSB9el3SssIA). You need `weights_SSD300.hdf5`, `weights_300x300_old.hdf5` is for the old version of architecture with 3x3 convolution for `pool6`.
diff --git a/SSD.ipynb b/SSD.ipynb