Edoardo-BS
commited on
Commit
•
16384fe
1
Parent(s):
d33b02b
Update README.md
Browse filescorrected typo, clarified input format
README.md
CHANGED
@@ -24,7 +24,7 @@ This dataset, counting 2.4 million ECGs from millions of patients in 4 countries
|
|
24 |
|
25 |
## Usage
|
26 |
|
27 |
-
**Input signals must be 5-second 12-lead ECGs sampled at 100 HZ**
|
28 |
|
29 |
```python
|
30 |
import torch
|
@@ -46,7 +46,7 @@ path = "path/to/your/finetuned-hubert-ecg-model.pt"
|
|
46 |
checkpoint = torch.load(path, map_location='cpu')
|
47 |
config = checkpoint['model_config']
|
48 |
hubert_ecg = HuBERTECG(config)
|
49 |
-
hubert_ecg = HuBERTForECGClassification(
|
50 |
hubert_ecg.load_state_dict(checkpoint['model_state_dict']) # fine-tuned model ready to be used or further fine-tuned
|
51 |
```
|
52 |
|
|
|
24 |
|
25 |
## Usage
|
26 |
|
27 |
+
**Input signals must be 5-second 12-lead ECGs sampled at 100 HZ. The leads are concatenated to each other**
|
28 |
|
29 |
```python
|
30 |
import torch
|
|
|
46 |
checkpoint = torch.load(path, map_location='cpu')
|
47 |
config = checkpoint['model_config']
|
48 |
hubert_ecg = HuBERTECG(config)
|
49 |
+
hubert_ecg = HuBERTForECGClassification(hubert_ecg)
|
50 |
hubert_ecg.load_state_dict(checkpoint['model_state_dict']) # fine-tuned model ready to be used or further fine-tuned
|
51 |
```
|
52 |
|