Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- spacy
|
| 4 |
+
- token-classification
|
| 5 |
+
- text-classification
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
model-index:
|
| 9 |
+
- name: en_healthsea
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: NER
|
| 13 |
+
type: token-classification
|
| 14 |
+
metrics:
|
| 15 |
+
- name: NER Precision
|
| 16 |
+
type: precision
|
| 17 |
+
value: 80.77
|
| 18 |
+
- name: NER Recall
|
| 19 |
+
type: recall
|
| 20 |
+
value: 79.92
|
| 21 |
+
- name: NER F Score
|
| 22 |
+
type: f_score
|
| 23 |
+
value: 80.34
|
| 24 |
+
---
|
| 25 |
+
|
| 26 |
+
# Welcome to Healthsea ✨
|
| 27 |
+
Create better access to health with machine learning and natural language processing. This is the trained healthsea pipeline for analyzing user reviews to supplements by extracting their effects on health. This pipeline features a trained NER model and a custom Text Classification model with Clause Segmentation and Blinding capabilities.
|
| 28 |
+
|
| 29 |
+
> In our [blog post](explosion.ai) you can read more about the architecture of healthsea and you can also visit the [healthsea repository](https://github.com/thomashacker/healthsea) for all the training workflows, custom components and training data.
|
| 30 |
+
|
| 31 |
+
| Feature | Description |
|
| 32 |
+
| --- | --- |
|
| 33 |
+
| **Name** | `en_healthsea` |
|
| 34 |
+
| **Version** | `0.0.0` |
|
| 35 |
+
| **spaCy** | `>=3.2.0,<3.3.0` |
|
| 36 |
+
| **Default Pipeline** | `sentencizer`, `tok2vec`, `ner`, `benepar`, `segmentation`, `clausecat`, `aggregation` |
|
| 37 |
+
| **Components** | `sentencizer`, `tok2vec`, `ner`, `benepar`, `segmentation`, `clausecat`, `aggregation` |
|
| 38 |
+
| **Vectors** | 684830 keys, 684830 unique vectors (300 dimensions) |
|
| 39 |
+
| **Sources** | n/a |
|
| 40 |
+
| **License** | MIT |
|
| 41 |
+
| **Author** | [Explosion](explosion.ai) |
|
| 42 |
+
|
| 43 |
+
### Label Scheme
|
| 44 |
+
|
| 45 |
+
<details>
|
| 46 |
+
|
| 47 |
+
<summary>View label scheme (6 labels for 2 components)</summary>
|
| 48 |
+
|
| 49 |
+
| Component | Labels |
|
| 50 |
+
| --- | --- |
|
| 51 |
+
| **`ner`** | `BENEFIT`, `CONDITION` |
|
| 52 |
+
| **`clausecat`** | `POSITIVE`, `NEUTRAL`, `NEGATIVE`, `ANAMNESIS` |
|
| 53 |
+
|
| 54 |
+
</details>
|
| 55 |
+
|
| 56 |
+
### Accuracy
|
| 57 |
+
|
| 58 |
+
| Type | Score |
|
| 59 |
+
| --- | --- |
|
| 60 |
+
| `ENTS_F` | 80.34 |
|
| 61 |
+
| `ENTS_P` | 80.77 |
|
| 62 |
+
| `ENTS_R` | 79.92 |
|
| 63 |
+
| `CATS_SCORE` | 74.87 |
|
| 64 |
+
| `CATS_MICRO_P` | 82.17 |
|
| 65 |
+
| `CATS_MICRO_R` | 80.85 |
|
| 66 |
+
| `CATS_MICRO_F` | 81.51 |
|
| 67 |
+
| `CATS_MACRO_P` | 78.01 |
|
| 68 |
+
| `CATS_MACRO_R` | 72.41 |
|
| 69 |
+
| `CATS_MACRO_F` | 74.87 |
|
| 70 |
+
| `CATS_MACRO_AUC` | 92.76 |
|
| 71 |
+
| `CATS_LOSS` | 297.22 |
|