AGarioud commited on
Commit
781941e
1 Parent(s): 1e4b930

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +20 -19
README.md CHANGED
@@ -108,7 +108,7 @@ pipeline_tag: image-segmentation
108
  - **Compute infrastructure:**
109
  - software: python, pytorch-lightning
110
  - hardware: HPC/AI resources provided by GENCI-IDRIS
111
- - **License:** : Apache 2.0
112
 
113
  ---
114
 
@@ -192,24 +192,25 @@ Statistics of the TRAIN+VALIDATION set :
192
 
193
  #### Training Hyperparameters
194
 
195
- * Model architecture: Unet (implementation from the [Segmentation Models Pytorch library](https://segmentation-modelspytorch.readthedocs.io/en/latest/docs/api.html#unet))
196
- * Encoder : mit-b5 pre-trained with ImageNet
197
- * Augmentation :
198
- * VerticalFlip(p=0.5)
199
- * HorizontalFlip(p=0.5)
200
- * RandomRotate90(p=0.5)
201
- * Input normalization (mean=0 | std=1):
202
- * norm_means: [105.08, 110.87, 101.82]
203
- * norm_stds: [52.17, 45.38, 44]
204
- * Seed: 2022
205
- * Batch size: 10
206
- * Number of epochs : 200
207
- * Early stopping : patience 30 and val_loss as monitor criterium
208
- * Optimizer : SGD
209
- * Schaeduler : mode = "min", factor = 0.5, patience = 10, cooldown = 4, min_lr = 1e-7
210
- * Learning rate : 0.02
211
- * Class Weights : [1-building: 1.0 , 2-pervious surface: 1.0 , 3-impervious surface: 1.0 , 4-bare soil: 1.0 , 5-water: 1.0 , 6-coniferous: 1.0 , 7-deciduous: 1.0 , 8-brushwood: 1.0 , 9-vineyard: 1.0 , 10-herbaceous vegetation: 1.0 , 11-agricultural land: 1.0 , 12-plowed land: 1.0 , 13-swimming_pool: 1.0 , 14-snow: 1.0 , 15-clear cut: 0.0 , 16-mixed: 0.0 , 17-ligneous: 0.0 , 18-greenhouse: 1.0 , 19-other: 0.0]
212
-
 
213
 
214
  #### Speeds, Sizes, Times
215
 
 
108
  - **Compute infrastructure:**
109
  - software: python, pytorch-lightning
110
  - hardware: HPC/AI resources provided by GENCI-IDRIS
111
+ - **License:** : Etalab 2.0
112
 
113
  ---
114
 
 
192
 
193
  #### Training Hyperparameters
194
 
195
+ ```yaml
196
+ - Model architecture: Unet (implementation from the [Segmentation Models Pytorch library](https://segmentation-modelspytorch.readthedocs.io/en/latest/docs/api.html#unet))
197
+ - Encoder : mit-b5 pre-trained with ImageNet
198
+ - Augmentation :
199
+ - VerticalFlip(p=0.5)
200
+ - HorizontalFlip(p=0.5)
201
+ - RandomRotate90(p=0.5)
202
+ - Input normalization (mean=0 | std=1):
203
+ - norm_means: [105.08, 110.87, 101.82]
204
+ - norm_stds: [52.17, 45.38, 44]
205
+ - Seed: 2022
206
+ - Batch size: 10
207
+ - Number of epochs : 200
208
+ - Early stopping : patience 30 and val_loss as monitor criterium
209
+ - Optimizer : SGD
210
+ - Schaeduler : mode = "min", factor = 0.5, patience = 10, cooldown = 4, min_lr = 1e-7
211
+ - Learning rate : 0.02
212
+ - Class Weights : [1-building: 1.0 , 2-pervious surface: 1.0 , 3-impervious surface: 1.0 , 4-bare soil: 1.0 , 5-water: 1.0 , 6-coniferous: 1.0 , 7-deciduous: 1.0 , 8-brushwood: 1.0 , 9-vineyard: 1.0 , 10-herbaceous vegetation: 1.0 , 11-agricultural land: 1.0 , 12-plowed land: 1.0 , 13-swimming_pool: 1.0 , 14-snow: 1.0 , 15-clear cut: 0.0 , 16-mixed: 0.0 , 17-ligneous: 0.0 , 18-greenhouse: 1.0 , 19-other: 0.0]
213
+ ```
214
 
215
  #### Speeds, Sizes, Times
216