Commit 07a55d13 authored by Cresson Remi's avatar Cresson Remi
Browse files

DOC: note with default placeholder with maggiori model

Showing with 2 additions and 0 deletions
+2 -0
......@@ -224,6 +224,8 @@ otbcli_TensorflowModelTrain \
-validation.source2.il $patches_labels_valid -validation.source2.name "estimated" \
```
Note that the `userplaceholders` parameter contains the *is_training* placeholder, fed with value *true* because the default value for this placeholder is *false*, and it is used in the batch normalization layers (take a look in the `create_savedmodel_maggiori17_fullyconv.py` code).
### Inference
This model can be used in fully convolutional mode only.
......
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment