Update README.md
Browse filesupdated description.
README.md
CHANGED
@@ -25,7 +25,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
25 |
# entity-recognition-general-sota-v1-finetuned-ner
|
26 |
|
27 |
This model is a fine-tuned version of [numind/entity-recognition-general-sota-v1](https://huggingface.co/numind/entity-recognition-general-sota-v1) on Babelscape/MultiNerd dataset.
|
28 |
-
The dataset if filtered on english
|
|
|
29 |
Train data - 131280 items
|
30 |
Eval data - 16410 items
|
31 |
It achieves the following results on the evaluation set:
|
@@ -37,7 +38,41 @@ It achieves the following results on the evaluation set:
|
|
37 |
|
38 |
## Model description
|
39 |
|
40 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
41 |
|
42 |
## Intended uses & limitations
|
43 |
|
|
|
25 |
# entity-recognition-general-sota-v1-finetuned-ner
|
26 |
|
27 |
This model is a fine-tuned version of [numind/entity-recognition-general-sota-v1](https://huggingface.co/numind/entity-recognition-general-sota-v1) on Babelscape/MultiNerd dataset.
|
28 |
+
The dataset if filtered on english language and sampled first 1M on train and 100k on validation.
|
29 |
+
further filtered with data containing atleast one tag from labels2ids mentioned below.
|
30 |
Train data - 131280 items
|
31 |
Eval data - 16410 items
|
32 |
It achieves the following results on the evaluation set:
|
|
|
38 |
|
39 |
## Model description
|
40 |
|
41 |
+
Trained on all tags from the MultiNERD dataset.
|
42 |
+
|
43 |
+
labels2ids = {
|
44 |
+
"O": 0,
|
45 |
+
"B-PER": 1,
|
46 |
+
"I-PER": 2,
|
47 |
+
"B-ORG": 3,
|
48 |
+
"I-ORG": 4,
|
49 |
+
"B-LOC": 5,
|
50 |
+
"I-LOC": 6,
|
51 |
+
"B-ANIM": 7,
|
52 |
+
"I-ANIM": 8,
|
53 |
+
"B-BIO": 9,
|
54 |
+
"I-BIO": 10,
|
55 |
+
"B-CEL": 11,
|
56 |
+
"I-CEL": 12,
|
57 |
+
"B-DIS": 13,
|
58 |
+
"I-DIS": 14,
|
59 |
+
"B-EVE": 15,
|
60 |
+
"I-EVE": 16,
|
61 |
+
"B-FOOD": 17,
|
62 |
+
"I-FOOD": 18,
|
63 |
+
"B-INST": 19,
|
64 |
+
"I-INST": 20,
|
65 |
+
"B-MEDIA": 21,
|
66 |
+
"I-MEDIA": 22,
|
67 |
+
"B-MYTH": 23,
|
68 |
+
"I-MYTH": 24,
|
69 |
+
"B-PLANT": 25,
|
70 |
+
"I-PLANT": 26,
|
71 |
+
"B-TIME": 27,
|
72 |
+
"I-TIME": 28,
|
73 |
+
"B-VEHI": 29,
|
74 |
+
"I-VEHI": 30,
|
75 |
+
}
|
76 |
|
77 |
## Intended uses & limitations
|
78 |
|