update TRT memory requirement in readme
Browse files- README.md +1 -1
- configs/metadata.json +2 -1
- docs/README.md +1 -1
README.md
CHANGED
|
@@ -40,7 +40,7 @@ Alternatively, we provide [resampled nifti images](https://drive.google.com/driv
|
|
| 40 |
## Training configuration
|
| 41 |
The training was performed with the following:
|
| 42 |
|
| 43 |
-
- GPU: at least 16GB GPU memory
|
| 44 |
- Actual Model Input: 192 x 192 x 80
|
| 45 |
- AMP: True
|
| 46 |
- Optimizer: Adam
|
|
|
|
| 40 |
## Training configuration
|
| 41 |
The training was performed with the following:
|
| 42 |
|
| 43 |
+
- GPU: at least 16GB GPU memory, requires 32G when exporting TRT model
|
| 44 |
- Actual Model Input: 192 x 192 x 80
|
| 45 |
- AMP: True
|
| 46 |
- Optimizer: Adam
|
configs/metadata.json
CHANGED
|
@@ -1,7 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
|
| 3 |
-
"version": "0.5.
|
| 4 |
"changelog": {
|
|
|
|
| 5 |
"0.5.7": "add dataset dir example",
|
| 6 |
"0.5.6": "add the ONNX-TensorRT way of model conversion",
|
| 7 |
"0.5.5": "update retrained validation results and training curve",
|
|
|
|
| 1 |
{
|
| 2 |
"schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
|
| 3 |
+
"version": "0.5.8",
|
| 4 |
"changelog": {
|
| 5 |
+
"0.5.8": "update TRT memory requirement in readme",
|
| 6 |
"0.5.7": "add dataset dir example",
|
| 7 |
"0.5.6": "add the ONNX-TensorRT way of model conversion",
|
| 8 |
"0.5.5": "update retrained validation results and training curve",
|
docs/README.md
CHANGED
|
@@ -33,7 +33,7 @@ Alternatively, we provide [resampled nifti images](https://drive.google.com/driv
|
|
| 33 |
## Training configuration
|
| 34 |
The training was performed with the following:
|
| 35 |
|
| 36 |
-
- GPU: at least 16GB GPU memory
|
| 37 |
- Actual Model Input: 192 x 192 x 80
|
| 38 |
- AMP: True
|
| 39 |
- Optimizer: Adam
|
|
|
|
| 33 |
## Training configuration
|
| 34 |
The training was performed with the following:
|
| 35 |
|
| 36 |
+
- GPU: at least 16GB GPU memory, requires 32G when exporting TRT model
|
| 37 |
- Actual Model Input: 192 x 192 x 80
|
| 38 |
- AMP: True
|
| 39 |
- Optimizer: Adam
|