Update README.md
Browse files
README.md
CHANGED
|
@@ -23,6 +23,7 @@ etc
|
|
| 23 |
|
| 24 |
## Architecture
|
| 25 |
This model uses an (as far as I am aware) novel architecture which stacks the lstms in parralel, then in sequence like in transformer blocks.
|
|
|
|
| 26 |
|
| 27 |
## Inference code
|
| 28 |
You need to place this in a python file in the same folder as the .pt file and then run it. change `device = torch.device("cuda:0")` to `device = torch.device("cpu")` for CPU inference
|
|
|
|
| 23 |
|
| 24 |
## Architecture
|
| 25 |
This model uses an (as far as I am aware) novel architecture which stacks the lstms in parralel, then in sequence like in transformer blocks.
|
| 26 |
+
It has 4,987,763 parameters.
|
| 27 |
|
| 28 |
## Inference code
|
| 29 |
You need to place this in a python file in the same folder as the .pt file and then run it. change `device = torch.device("cuda:0")` to `device = torch.device("cpu")` for CPU inference
|