Update README.md
Browse files
README.md
CHANGED
@@ -202,9 +202,15 @@ Input normalization was performed
|
|
202 |
* VerticalFlip(p=0.5)
|
203 |
* HorizontalFlip(p=0.5)
|
204 |
* RandomRotate90(p=0.5)
|
|
|
|
|
|
|
205 |
* Seed: 2022
|
206 |
* Batch size: 10
|
|
|
|
|
207 |
* Optimizer : SGD
|
|
|
208 |
* Learning rate : 0.02
|
209 |
* Class Weights :
|
210 |
* 1: [1, building]
|
@@ -232,6 +238,12 @@ Input normalization was performed
|
|
232 |
|
233 |
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
234 |
|
|
|
|
|
|
|
|
|
|
|
|
|
235 |
{{ speeds_sizes_times | default("[More Information Needed]", true)}}
|
236 |
|
237 |
## Evaluation
|
|
|
202 |
* VerticalFlip(p=0.5)
|
203 |
* HorizontalFlip(p=0.5)
|
204 |
* RandomRotate90(p=0.5)
|
205 |
+
* Input normalization (mean=0 | std=1):
|
206 |
+
* norm_means: [105.08, 110.87, 101.82, 106.38, 53.26]
|
207 |
+
* norm_stds: [52.17, 45.38, 44, 39.69, 79.3]
|
208 |
* Seed: 2022
|
209 |
* Batch size: 10
|
210 |
+
* Number of epochs : 200
|
211 |
+
* Early stopping : patience 30 and val_loss as monitor criterium
|
212 |
* Optimizer : SGD
|
213 |
+
* Schaeduler : mode = "min", factor = 0.5, patience = 10, cooldown = 4, min_lr = 1e-7
|
214 |
* Learning rate : 0.02
|
215 |
* Class Weights :
|
216 |
* 1: [1, building]
|
|
|
238 |
|
239 |
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
240 |
|
241 |
+
The FLAIR-INC_RVBIE_resnet34_unet_15cl_norm model was trained on a HPC/AI resources provided by GENCI-IDRIS (Grant 2022-A0131013803). 16 V100 GPUs were requested ( 4 nodes, 4 GPUS per node).
|
242 |
+
|
243 |
+
FLAIR-INC_RVBIE_resnet34_unet_15cl_norm was obtained for num_epoch=76 with corresponding val_loss=0.56.
|
244 |
+
|
245 |
+
|
246 |
+
|
247 |
{{ speeds_sizes_times | default("[More Information Needed]", true)}}
|
248 |
|
249 |
## Evaluation
|