Skip to content

Commit

Permalink
Merge pull request #221 from BerkeleyLab/doc-README-update-training_c…
Browse files Browse the repository at this point in the history
…onfiguration

doc(README): "tensor names" in JSON configuration
  • Loading branch information
rouson authored Oct 29, 2024
2 parents 18b6e6e + 28280e6 commit 9a277f2
Showing 1 changed file with 19 additions and 14 deletions.
33 changes: 19 additions & 14 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -142,22 +142,27 @@ To see the expected file format, run the [print-training-configuration] example
which should produce output like the following:
```
Project is up to date
{
"hyperparameters": {
"mini-batches" : 10,
"learning rate" : 1.50000000,
"optimizer" : "adam"
}
,
"network configuration": {
"skip connections" : false,
"nodes per layer" : [2,72,2],
"activation function" : "sigmoid"
}
}
{
"hyperparameters": {
"mini-batches" : 10,
"learning rate" : 1.5,
"optimizer" : "adam"
}
,
"network configuration": {
"skip connections" : false,
"nodes per layer" : [2,72,2],
"activation function" : "sigmoid"
}
,
"tensor names": {
"inputs" : ["pressure","temperature"],
"outputs" : ["saturated mixing ratio"]
}
}
```
The Fiats JSON file format is fragile: splitting or combining lines breaks the file reader.
Files with added or removed white space or reordered whole objects ("hyperparameters" or "network configuration") should work.
Files with added or removed white space or reordered whole objects ("hyperparameters", "network configuration", or "tensor names") should work.
A future release will leverage the [rojff] JSON interface to allow for more flexible file formatting.

### Training a neural network
Expand Down

0 comments on commit 9a277f2

Please sign in to comment.