Datasets:
Tasks:
Token Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
named-entity-recognition
Languages:
English
Size:
1K - 10K
License:
fixed readme
Browse files
README.md
CHANGED
@@ -3,15 +3,9 @@ annotations_creators:
|
|
3 |
- expert-generated
|
4 |
language_creators:
|
5 |
- found
|
6 |
-
<<<<<<< HEAD
|
7 |
-
languages:
|
8 |
-
- en
|
9 |
-
licenses:
|
10 |
-
=======
|
11 |
language:
|
12 |
- en
|
13 |
license:
|
14 |
-
>>>>>>> 297571f844c69c59b0a7d6325ad12c86b64aa523
|
15 |
- cc-by-4.0
|
16 |
multilinguality:
|
17 |
- monolingual
|
@@ -25,10 +19,7 @@ task_ids:
|
|
25 |
- named-entity-recognition
|
26 |
---
|
27 |
# Dataset for the first <a href="https://ui.adsabs.harvard.edu/WIESP/" style="color:blue">Workshop on Information Extraction from Scientific Publications (WIESP/2022)</a>.
|
28 |
-
<<<<<<< HEAD
|
29 |
-
=======
|
30 |
**(NOTE: loading from the Huggingface Dataset Hub directly does not work. You need to clone the repository locally.)**
|
31 |
-
>>>>>>> 297571f844c69c59b0a7d6325ad12c86b64aa523
|
32 |
|
33 |
## Dataset Description
|
34 |
Datasets with text fragments from astrophysics papers, provided by the [NASA Astrophysical Data System](https://ui.adsabs.harvard.edu/) with manually tagged astronomical facilities and other entities of interest (e.g., celestial objects).
|
@@ -58,11 +49,7 @@ with open("./WIESP2022-NER-DEV.jsonl", 'r') as f:
|
|
58 |
from datasets import Dataset
|
59 |
wiesp_dev_from_json = Dataset.from_json(path_or_paths="./WIESP2022-NER-DEV.jsonl")
|
60 |
```
|
61 |
-
<<<<<<< HEAD
|
62 |
-
(NOTE: currently loading from the Huggingface Dataset Hub directly does not work. You need to clone the repository locally)
|
63 |
-
=======
|
64 |
(NOTE: loading from the Huggingface Dataset Hub directly does not work. You need to clone the repository locally.)
|
65 |
-
>>>>>>> 297571f844c69c59b0a7d6325ad12c86b64aa523
|
66 |
|
67 |
How to compute your scores on the training data:
|
68 |
1. format your predictions as a list of dictionaries, each with the same `"unique_id"` and `"tokens"` keys from the dataset, as well as the list of predicted NER tags under the `"pred_ner_tags"` key (see `WIESP2022-NER-DEV-sample-predictions.jsonl` for an example).
|
@@ -82,10 +69,7 @@ To get scores on the validation data, zip your predictions file (a single `.json
|
|
82 |
βββ WIESP2022-NER-DEV-sample-predictions.jsonl : an example file with properly formatted predictions on the development data.
|
83 |
βββ WIESP2022-NER-VALIDATION-NO-LABELS.jsonl : 1366 samples for validation without the NER labels. Used for the WIESP2022 workshop.
|
84 |
βββ README.MD: this file.
|
85 |
-
<<<<<<< HEAD
|
86 |
-
=======
|
87 |
βββ tag_definitions.txt: short descriptions and examples of the tags used in the task.
|
88 |
-
>>>>>>> 297571f844c69c59b0a7d6325ad12c86b64aa523
|
89 |
βββ scoring-scripts/ : scripts used to evaluate submissions.
|
90 |
βββ compute_MCC.py : computes the Matthews correlation coefficient between two datasets.
|
91 |
βββ compute_seqeval.py : computes the seqeval scores (precision, recall, f1, overall and for each class) between two datasets.
|
|
|
3 |
- expert-generated
|
4 |
language_creators:
|
5 |
- found
|
|
|
|
|
|
|
|
|
|
|
6 |
language:
|
7 |
- en
|
8 |
license:
|
|
|
9 |
- cc-by-4.0
|
10 |
multilinguality:
|
11 |
- monolingual
|
|
|
19 |
- named-entity-recognition
|
20 |
---
|
21 |
# Dataset for the first <a href="https://ui.adsabs.harvard.edu/WIESP/" style="color:blue">Workshop on Information Extraction from Scientific Publications (WIESP/2022)</a>.
|
|
|
|
|
22 |
**(NOTE: loading from the Huggingface Dataset Hub directly does not work. You need to clone the repository locally.)**
|
|
|
23 |
|
24 |
## Dataset Description
|
25 |
Datasets with text fragments from astrophysics papers, provided by the [NASA Astrophysical Data System](https://ui.adsabs.harvard.edu/) with manually tagged astronomical facilities and other entities of interest (e.g., celestial objects).
|
|
|
49 |
from datasets import Dataset
|
50 |
wiesp_dev_from_json = Dataset.from_json(path_or_paths="./WIESP2022-NER-DEV.jsonl")
|
51 |
```
|
|
|
|
|
|
|
52 |
(NOTE: loading from the Huggingface Dataset Hub directly does not work. You need to clone the repository locally.)
|
|
|
53 |
|
54 |
How to compute your scores on the training data:
|
55 |
1. format your predictions as a list of dictionaries, each with the same `"unique_id"` and `"tokens"` keys from the dataset, as well as the list of predicted NER tags under the `"pred_ner_tags"` key (see `WIESP2022-NER-DEV-sample-predictions.jsonl` for an example).
|
|
|
69 |
βββ WIESP2022-NER-DEV-sample-predictions.jsonl : an example file with properly formatted predictions on the development data.
|
70 |
βββ WIESP2022-NER-VALIDATION-NO-LABELS.jsonl : 1366 samples for validation without the NER labels. Used for the WIESP2022 workshop.
|
71 |
βββ README.MD: this file.
|
|
|
|
|
72 |
βββ tag_definitions.txt: short descriptions and examples of the tags used in the task.
|
|
|
73 |
βββ scoring-scripts/ : scripts used to evaluate submissions.
|
74 |
βββ compute_MCC.py : computes the Matthews correlation coefficient between two datasets.
|
75 |
βββ compute_seqeval.py : computes the seqeval scores (precision, recall, f1, overall and for each class) between two datasets.
|