Update documentation for Hunflair2 release #3410
Conversation
…ow warnings if hunflair v1 models are loaded
flair/models/multitask_model.py
Outdated
| @@ -260,6 +260,14 @@ def _fetch_model(model_name) -> str: | |||
|
|
|||
| cache_dir = Path("models") | |||
| if model_name in model_map: | |||
| if model_name.startswith("hunflair") or model_name == "bioner": | |||
There was a problem hiding this comment.
"hunflair2" also starts with "hunflair", so I think this warning would always be printed.
There was a problem hiding this comment.
True. However, HunFlair2 will not be loaded as MultitaskModel. I fix it anyway.
| @@ -781,6 +781,14 @@ def _fetch_model(model_name) -> str: | |||
| elif model_name in hu_model_map: | |||
| model_path = cached_path(hu_model_map[model_name], cache_dir=cache_dir) | |||
|
|
|||
| if model_name.startswith("hunflair"): | |||
There was a problem hiding this comment.
True. However, HunFlair2 will not be loaded as SequenceTaggerModel. I fix it anyway.
Added full description
Fixed some syntax errors
Fixed descriptions
| @@ -648,6 +649,8 @@ def p(text: str) -> str: | |||
| emb = emb / torch.norm(emb) | |||
| dense_embeddings.append(emb.cpu().numpy()) | |||
| sent.clear_embeddings() | |||
|
|
|||
| # empty cuda cache if device is a cuda device | |||
| if flair.device.type == "cuda": | |||
There was a problem hiding this comment.
@sg-wbi Is this really required?
| # Sanity conversion: if flair.device was set as a string, convert to torch.device | ||
| if isinstance(flair.device, str): | ||
| flair.device = torch.device(flair.device) | ||
|
|
||
| if flair.device.type == "cuda": |
There was a problem hiding this comment.
@sg-wbi Is this really required?
|
Great thanks! I am not sure either. @helpmefindaname / @mariosaenger can you remember why this was added? |
|
Hi @alanakbik @sg-wbi |
|
Alright, thanks! |
This PR updates the documentation for HunFlair lifting it to new release. More specifically the PR contains: