Update README.md
Browse files
README.md
CHANGED
@@ -1,45 +1,58 @@
|
|
1 |
-
---
|
2 |
-
dataset_info:
|
3 |
-
features:
|
4 |
-
- name: tokens
|
5 |
-
sequence: string
|
6 |
-
- name: ner_tags
|
7 |
-
sequence:
|
8 |
-
class_label:
|
9 |
-
names:
|
10 |
-
'0': O
|
11 |
-
'1': B-PER
|
12 |
-
'2': I-PER
|
13 |
-
'3': B-ORG
|
14 |
-
'4': I-ORG
|
15 |
-
'5': B-LOC
|
16 |
-
'6': I-LOC
|
17 |
-
'7': B-TIME
|
18 |
-
'8': I-TIME
|
19 |
-
'9': B-TTL
|
20 |
-
'10': I-TTL
|
21 |
-
splits:
|
22 |
-
- name: train
|
23 |
-
num_bytes: 2138256
|
24 |
-
num_examples: 3465
|
25 |
-
download_size: 546138
|
26 |
-
dataset_size: 2138256
|
27 |
-
configs:
|
28 |
-
- config_name: default
|
29 |
-
data_files:
|
30 |
-
- split: train
|
31 |
-
path: data/train-*
|
32 |
-
task_categories:
|
33 |
-
- token-classification
|
34 |
-
language:
|
35 |
-
- am
|
36 |
-
size_categories:
|
37 |
-
- 1K<n<10K
|
38 |
-
---
|
39 |
|
40 |
# Amharic Named Entity Recognition Dataset
|
41 |
This dataset can be used to train models for Named Entity Recognition.
|
42 |
|
43 |
### Dataset Source
|
44 |
|
45 |
-
https://github.com/uhh-lt/ethiopicmodels/blob/master/am/data/NER/train.txt
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
dataset_info:
|
3 |
+
features:
|
4 |
+
- name: tokens
|
5 |
+
sequence: string
|
6 |
+
- name: ner_tags
|
7 |
+
sequence:
|
8 |
+
class_label:
|
9 |
+
names:
|
10 |
+
'0': O
|
11 |
+
'1': B-PER
|
12 |
+
'2': I-PER
|
13 |
+
'3': B-ORG
|
14 |
+
'4': I-ORG
|
15 |
+
'5': B-LOC
|
16 |
+
'6': I-LOC
|
17 |
+
'7': B-TIME
|
18 |
+
'8': I-TIME
|
19 |
+
'9': B-TTL
|
20 |
+
'10': I-TTL
|
21 |
+
splits:
|
22 |
+
- name: train
|
23 |
+
num_bytes: 2138256
|
24 |
+
num_examples: 3465
|
25 |
+
download_size: 546138
|
26 |
+
dataset_size: 2138256
|
27 |
+
configs:
|
28 |
+
- config_name: default
|
29 |
+
data_files:
|
30 |
+
- split: train
|
31 |
+
path: data/train-*
|
32 |
+
task_categories:
|
33 |
+
- token-classification
|
34 |
+
language:
|
35 |
+
- am
|
36 |
+
size_categories:
|
37 |
+
- 1K<n<10K
|
38 |
+
---
|
39 |
|
40 |
# Amharic Named Entity Recognition Dataset
|
41 |
This dataset can be used to train models for Named Entity Recognition.
|
42 |
|
43 |
### Dataset Source
|
44 |
|
45 |
+
https://github.com/uhh-lt/ethiopicmodels/blob/master/am/data/NER/train.txt
|
46 |
+
|
47 |
+
### Finetuned Models
|
48 |
+
|
49 |
+
The following models were finetuned using this dataset. The reported precision, recall, and f1 metrics are macro averages.
|
50 |
+
|
51 |
+
|Model|Size (# params)| Precision | Recall | F1 |
|
52 |
+
| --- | ------------- | --------- |------- | -- |
|
53 |
+
|bert-medium-amharic|40.5M|0.64|0.73|0.68|
|
54 |
+
|bert-small-amharic|27.8M|0.64|0.72|0.68|
|
55 |
+
|bert-mini-amharic|10.7M|0.60|0.67|0.64|
|
56 |
+
|bert-tiny-amharic|4.18M|0.50|0.59|0.54|
|
57 |
+
|xlm-roberta-base|279M|0.69|0.79|0.73|
|
58 |
+
|am-roberta|443M|0.67|0.72|0.69|
|