|
--- |
|
tags: |
|
- spacy |
|
- token-classification |
|
language: |
|
- en |
|
widget: |
|
- text: "Moses received the Torah from Mount Sinai (Avot Chapter 1 Mishnah 1)" |
|
- text: "Yerushalmi Peah 23d" |
|
- text: "Rashi argues with Tosafot about this. See Rashi Bava Kamma 40a s.v. Amar." |
|
- text: "Bet Shammai says he is obligated to pay while Bet Hillel says he is exempt." |
|
model-index: |
|
- name: en_torah_ner |
|
results: |
|
- task: |
|
name: NER |
|
type: token-classification |
|
metrics: |
|
- name: NER Precision |
|
type: precision |
|
value: 0.8413793103 |
|
- name: NER Recall |
|
type: recall |
|
value: 0.87517934 |
|
- name: NER F Score |
|
type: f_score |
|
value: 0.8579465541 |
|
--- |
|
|
|
See below for technical details about the model. |
|
|
|
# Description |
|
|
|
This model is a named entity recognition model that was trained to run on text that discusses Torah topics (e.g. dvar torahs, Torah blogs, translations of classic Torah texts etc.). |
|
|
|
It detects the following types of entities: |
|
|
|
| Label | Description |
|
|---|---| |
|
| Person | Name of a person | |
|
| Group | Name of a group of people. E.g. nations (Egypt), schools (Bet Hillel, Tosafot) | |
|
| Citation | Citations to Torah texts. See notes below. | |
|
|
|
## Notes on normalization |
|
|
|
All text the model was trained on was initially put through the following normalizer: [link](https://github.com/Sefaria/Machine-Learning/blob/main/util/helper.py#L43). |
|
Results will be signicantly worse if this normalizer is not used. |
|
|
|
## Notes on citation matches |
|
|
|
- Final parentheses is not included in the match. E.g. if the citation is `Genesis (1:1)` then the final parentheses will not be included. We found that the model would get confused if the final parentheses was part of the entity. It is fairly simple to add it back in via a deterministic check. |
|
- Only the first word of a dibur hamatchil is included in the match. E.g. `Tosafot s.v. Amar Rabbi Akiva` only until the word `Amar` will be tagged. We found the model had trouble determining the end of the dibur hamatchil. |
|
- See Ref part model for a model that can break down citations into chunks so it is simpler to parse them. |
|
|
|
## Using with Sefaria-Project |
|
|
|
The [Sefaria-Project](https://github.com/Sefaria/Sefaria-Project) repo can use this model to return objects linked to objects in the Sefaria database. Non-citation entities are linked to `Topic` objects and citation entities are linked to `Ref` objects. |
|
|
|
Note, this model is designed to be used in conjunction with the corresponding [subref model](https://huggingface.co/Sefaria/en_subref_ner). That model takes citations as input and tags the parts of the citation. The below instructions explain how to integrate both of these models into Sefaria-Project. |
|
|
|
### Configuring Sefaria-Project to use this model |
|
|
|
The assumption is that Sefaria-Project is set up on your environment following the instructions in our [README](https://github.com/Sefaria/Sefaria-Project/blob/master/README.mkd). |
|
|
|
Download this repo and the [subref repo](https://huggingface.co/Sefaria/en_subref_ner). |
|
|
|
In `local_settings.py`, modify the following lines: |
|
|
|
```python |
|
ENABLE_LINKER = True |
|
|
|
RAW_REF_MODEL_BY_LANG_FILEPATH = { |
|
"en": "/path/to/en-ref-ner model" |
|
} |
|
|
|
RAW_REF_PART_MODEL_BY_LANG_FILEPATH = { |
|
"en": "/path/to/en-subref-ner model", |
|
} |
|
``` |
|
|
|
Make sure spaCy is installed. |
|
|
|
```bash |
|
pip install spacy==3.4.1 |
|
``` |
|
|
|
### Running the model with Sefaria-Project |
|
|
|
The following code shows an example of instantiating the `Linker` object which uses the ML models and running the `Linker` with input. |
|
|
|
```python |
|
import django |
|
django.setup() |
|
from sefaria.model.text import library |
|
|
|
text = "Moses received the Torah from Har Sinai (Avot Chapter 1 Mishnah 1)" |
|
linker = library.get_linker("en") |
|
doc = linker.link(text) |
|
|
|
print("Named entities") |
|
for resolved_named_entity in doc.resolved_named_entities: |
|
print("---") |
|
print("Text:", resolved_named_entity.raw_entity.text) |
|
print("Topic Slug:", resolved_named_entity.topic.slug) |
|
|
|
print("Citations") |
|
for resolved_ref in doc.resolved_refs: |
|
print("---") |
|
print("Text:", resolved_ref.raw_entity.text) |
|
print("Ref:", resolved_ref.ref.normal()) |
|
``` |
|
|
|
# Technical Details |
|
|
|
| Feature | Description | |
|
| --- | --- | |
|
| **Name** | `en_torah_ner` | |
|
| **Version** | `1.0.0` | |
|
| **spaCy** | `>=3.4.1,<3.5.0` | |
|
| **Default Pipeline** | `tok2vec`, `ner` | |
|
| **Components** | `tok2vec`, `ner` | |
|
| **Vectors** | 218765 keys, 218765 unique vectors (50 dimensions) | |
|
| **Sources** | n/a | |
|
| **License** | GPLv3.0 | |
|
| **Author** | Sefaria | |
|
|
|
### Label Scheme |
|
|
|
<details> |
|
|
|
<summary>View label scheme (3 labels for 1 components)</summary> |
|
|
|
| Component | Labels | |
|
| --- | --- | |
|
| **`ner`** | `Citation`, `Group`, `Person` | |
|
|
|
</details> |
|
|
|
### Accuracy |
|
|
|
| Type | Score | |
|
| --- | --- | |
|
| `ENTS_F` | 85.79 | |
|
| `ENTS_P` | 84.14 | |
|
| `ENTS_R` | 87.52 | |
|
| `TOK2VEC_LOSS` | 136797.07 | |
|
| `NER_LOSS` | 95967.72 | |
