Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -6,13 +6,13 @@ description: Finetuned RoBECZech model for detection of disease spans. This mode
|
|
| 6 |
license: mit
|
| 7 |
language: cs
|
| 8 |
tags:
|
| 9 |
-
-
|
|
|
|
| 10 |
- biomedical
|
|
|
|
| 11 |
- lexical semantic
|
| 12 |
-
- bionlp
|
| 13 |
- clinical ner
|
| 14 |
-
-
|
| 15 |
-
- biology
|
| 16 |
base_model: ufal/robeczech-base
|
| 17 |
pipeline_tag: token-classification
|
| 18 |
---
|
|
@@ -54,16 +54,11 @@ named_ents = le_pipe(SOME_TEXT, stride=256)
|
|
| 54 |
```
|
| 55 |
|
| 56 |
|
| 57 |
-
# Data description
|
| 58 |
|
| 59 |
-
50/50 Train/validation split on CardioCCC, a manually labeled cardiology corpus
|
| 60 |
|
| 61 |
-
#
|
| 62 |
|
| 63 |
-
|
| 64 |
-
* 2025-05-26_0002: basemodel=ufal/robeczech-base, learning rate; 1e-5, epochs; 20, decay: 1e-3, With IOB tagging
|
| 65 |
-
* 2025-05-26_0003: basemodel=ufal/robeczech-base, learning rate; 5e-5, epochs; 20, decay: 1e-3, With IOB tagging
|
| 66 |
-
* 2025-05-26_0004: basemodel=ufal/robeczech-base, learning rate; 5e-5, epochs; 20, decay: 1e-3, no IOB tagging
|
| 67 |
|
| 68 |
|
| 69 |
# Acknowledgement
|
|
@@ -72,6 +67,8 @@ This is part of the [DT4H project](https://www.datatools4heart.eu/).
|
|
| 72 |
|
| 73 |
# Doi and reference
|
| 74 |
|
|
|
|
|
|
|
| 75 |
For more details about training/eval and other scripts, see CardioNER [github repo](https://github.com/DataTools4Heart/CardioNER).
|
| 76 |
and for more information on the background, see Datatools4Heart [Huggingface](https://huggingface.co/DT4H)/[Website](https://www.datatools4heart.eu/)
|
| 77 |
|
|
|
|
| 6 |
license: mit
|
| 7 |
language: cs
|
| 8 |
tags:
|
| 9 |
+
- science
|
| 10 |
+
- biology
|
| 11 |
- biomedical
|
| 12 |
+
- span classification
|
| 13 |
- lexical semantic
|
|
|
|
| 14 |
- clinical ner
|
| 15 |
+
- bionlp
|
|
|
|
| 16 |
base_model: ufal/robeczech-base
|
| 17 |
pipeline_tag: token-classification
|
| 18 |
---
|
|
|
|
| 54 |
```
|
| 55 |
|
| 56 |
|
|
|
|
| 57 |
|
|
|
|
| 58 |
|
| 59 |
+
# Data description
|
| 60 |
|
| 61 |
+
50/50 Train/validation split on CardioCCC, a manually labeled cardiology corpus
|
|
|
|
|
|
|
|
|
|
| 62 |
|
| 63 |
|
| 64 |
# Acknowledgement
|
|
|
|
| 67 |
|
| 68 |
# Doi and reference
|
| 69 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
For more details about training/eval and other scripts, see CardioNER [github repo](https://github.com/DataTools4Heart/CardioNER).
|
| 73 |
and for more information on the background, see Datatools4Heart [Huggingface](https://huggingface.co/DT4H)/[Website](https://www.datatools4heart.eu/)
|
| 74 |
|