monai
medical
katielink commited on
Commit
4528beb
·
1 Parent(s): e430b14

add RAM warning

Browse files
Files changed (3) hide show
  1. README.md +5 -0
  2. configs/metadata.json +2 -1
  3. docs/README.md +5 -0
README.md CHANGED
@@ -32,6 +32,11 @@ The training was performed with the following:
32
  - Optimizer: Novograd
33
  - Learning Rate: 0.002
34
  - Loss: DiceCELoss
 
 
 
 
 
35
 
36
  ### Input
37
  One channel
 
32
  - Optimizer: Novograd
33
  - Learning Rate: 0.002
34
  - Loss: DiceCELoss
35
+ - Dataset Manager: CacheDataset
36
+
37
+ ### Memory Consumption Warning
38
+
39
+ If you face memory issues with CacheDataset, you can either switch to a regular Dataset class or lower the caching rate `cache_rate` in the configurations within range $(0, 1)$ to minimize the System RAM requirements.
40
 
41
  ### Input
42
  One channel
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": "0.5.0",
4
  "changelog": {
 
5
  "0.5.0": "update the README file with the ONNX-TensorRT conversion",
6
  "0.4.9": "update TensorRT descriptions",
7
  "0.4.8": "update deterministic training results",
 
1
  {
2
  "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20220324.json",
3
+ "version": "0.5.1",
4
  "changelog": {
5
+ "0.5.1": "add RAM warning",
6
  "0.5.0": "update the README file with the ONNX-TensorRT conversion",
7
  "0.4.9": "update TensorRT descriptions",
8
  "0.4.8": "update deterministic training results",
docs/README.md CHANGED
@@ -25,6 +25,11 @@ The training was performed with the following:
25
  - Optimizer: Novograd
26
  - Learning Rate: 0.002
27
  - Loss: DiceCELoss
 
 
 
 
 
28
 
29
  ### Input
30
  One channel
 
25
  - Optimizer: Novograd
26
  - Learning Rate: 0.002
27
  - Loss: DiceCELoss
28
+ - Dataset Manager: CacheDataset
29
+
30
+ ### Memory Consumption Warning
31
+
32
+ If you face memory issues with CacheDataset, you can either switch to a regular Dataset class or lower the caching rate `cache_rate` in the configurations within range $(0, 1)$ to minimize the System RAM requirements.
33
 
34
  ### Input
35
  One channel