File size: 8,687 Bytes
c298231 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 | ---
library_name: span-marker
tags:
- span-marker
- token-classification
- ner
- named-entity-recognition
- generated_from_span_marker_trainer
metrics:
- precision
- recall
- f1
widget:
- text: The Bengal tiger is the most common subspecies of tiger, constituting approximately
80% of the entire tiger population, and is found in Bangladesh, Bhutan, Myanmar,
Nepal, and India.
- text: In other countries, it is a non-commissioned rank (e.g. Spain, Italy, France,
the Netherlands and the Indonesian Police ranks).
- text: The filling consists of fish, pork and bacon, and is seasoned with salt (unless
the pork is already salted).
- text: This stood until August 20, 1993 when it was beaten by one 1 / 100th of a
second by Colin Jackson of Great Britain in Stuttgart, Germany, a subsequent record
that stood for 13 years.
- text: Ann Patchett ’s novel " Bel Canto ", was another creative influence that helped
her manage a plentiful cast of characters.
pipeline_tag: token-classification
model-index:
- name: SpanMarker
results:
- task:
type: token-classification
name: Named Entity Recognition
dataset:
name: Unknown
type: unknown
split: eval
metrics:
- type: f1
value: 0.9130661114003124
name: F1
- type: precision
value: 0.9148758606300855
name: Precision
- type: recall
value: 0.9112635078969243
name: Recall
---
# SpanMarker
This is a [SpanMarker](https://github.com/tomaarsen/SpanMarkerNER) model that can be used for Named Entity Recognition.
## Model Details
### Model Description
- **Model Type:** SpanMarker
<!-- - **Encoder:** [Unknown](https://huggingface.co/unknown) -->
- **Maximum Sequence Length:** 256 tokens
- **Maximum Entity Length:** 6 words
<!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
<!-- - **Language:** Unknown -->
<!-- - **License:** Unknown -->
### Model Sources
- **Repository:** [SpanMarker on GitHub](https://github.com/tomaarsen/SpanMarkerNER)
- **Thesis:** [SpanMarker For Named Entity Recognition](https://raw.githubusercontent.com/tomaarsen/SpanMarkerNER/main/thesis.pdf)
### Model Labels
| Label | Examples |
|:------|:-------------------------------------------------------------------------|
| ANIM | "vertebrate", "moth", "G. firmus" |
| BIO | "Aspergillus", "Cladophora", "Zythiostroma" |
| CEL | "pulsar", "celestial bodies", "neutron star" |
| DIS | "social anxiety disorder", "insulin resistance", "Asperger syndrome" |
| EVE | "Spanish Civil War", "National Junior Angus Show", "French Revolution" |
| FOOD | "Neera", "Bellini ( cocktail )", "soju" |
| INST | "Apple II", "Encyclopaedia of Chess Openings", "Android" |
| LOC | "Kīlauea", "Hungary", "Vienna" |
| MEDIA | "CSI : Crime Scene Investigation", "Big Comic Spirits", "American Idol" |
| MYTH | "Priam", "Oźwiena", "Odysseus" |
| ORG | "San Francisco Giants", "Arm Holdings", "RTÉ One" |
| PER | "Amelia Bence", "Tito Lusiardo", "James Cameron" |
| PLANT | "vernal squill", "Sarracenia purpurea", "Drosera rotundifolia" |
| TIME | "prehistory", "Age of Enlightenment", "annual paid holiday" |
| VEHI | "Short 360", "Ferrari 355 Challenge", "Solution F / Chretien Helicopter" |
## Uses
### Direct Use for Inference
```python
from span_marker import SpanMarkerModel
# Download from the 🤗 Hub
model = SpanMarkerModel.from_pretrained("span_marker_model_id")
# Run inference
entities = model.predict("Ann Patchett ’s novel \" Bel Canto \", was another creative influence that helped her manage a plentiful cast of characters.")
```
### Downstream Use
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
```python
from span_marker import SpanMarkerModel, Trainer
# Download from the 🤗 Hub
model = SpanMarkerModel.from_pretrained("span_marker_model_id")
# Specify a Dataset with "tokens" and "ner_tag" columns
dataset = load_dataset("conll2003") # For example CoNLL2003
# Initialize a Trainer using the pretrained model & dataset
trainer = Trainer(
model=model,
train_dataset=dataset["train"],
eval_dataset=dataset["validation"],
)
trainer.train()
trainer.save_model("span_marker_model_id-finetuned")
```
</details>
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
<!--
## Bias, Risks and Limitations
*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
-->
<!--
### Recommendations
*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
-->
## Training Details
### Training Set Metrics
| Training set | Min | Median | Max |
|:----------------------|:----|:--------|:----|
| Sentence length | 2 | 21.6493 | 237 |
| Entities per sentence | 0 | 1.5369 | 36 |
### Training Hyperparameters
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 1
- mixed_precision_training: Native AMP
### Training Results
| Epoch | Step | Validation Loss | Validation Precision | Validation Recall | Validation F1 | Validation Accuracy |
|:------:|:-----:|:---------------:|:--------------------:|:-----------------:|:-------------:|:-------------------:|
| 0.0576 | 1000 | 0.0142 | 0.8714 | 0.7729 | 0.8192 | 0.9698 |
| 0.1153 | 2000 | 0.0107 | 0.8316 | 0.8815 | 0.8558 | 0.9744 |
| 0.1729 | 3000 | 0.0092 | 0.8717 | 0.8797 | 0.8757 | 0.9780 |
| 0.2306 | 4000 | 0.0082 | 0.8811 | 0.8886 | 0.8848 | 0.9798 |
| 0.2882 | 5000 | 0.0084 | 0.8523 | 0.9163 | 0.8831 | 0.9790 |
| 0.3459 | 6000 | 0.0079 | 0.8700 | 0.9113 | 0.8902 | 0.9802 |
| 0.4035 | 7000 | 0.0070 | 0.9107 | 0.8859 | 0.8981 | 0.9822 |
| 0.4611 | 8000 | 0.0069 | 0.9259 | 0.8797 | 0.9022 | 0.9827 |
| 0.5188 | 9000 | 0.0067 | 0.9061 | 0.8965 | 0.9013 | 0.9829 |
| 0.5764 | 10000 | 0.0066 | 0.9034 | 0.8996 | 0.9015 | 0.9829 |
| 0.6341 | 11000 | 0.0064 | 0.9160 | 0.8996 | 0.9077 | 0.9839 |
| 0.6917 | 12000 | 0.0066 | 0.8952 | 0.9121 | 0.9036 | 0.9832 |
| 0.7494 | 13000 | 0.0062 | 0.9165 | 0.9009 | 0.9086 | 0.9841 |
| 0.8070 | 14000 | 0.0062 | 0.9010 | 0.9121 | 0.9065 | 0.9835 |
| 0.8647 | 15000 | 0.0062 | 0.9084 | 0.9127 | 0.9105 | 0.9842 |
| 0.9223 | 16000 | 0.0060 | 0.9151 | 0.9098 | 0.9125 | 0.9846 |
| 0.9799 | 17000 | 0.0060 | 0.9149 | 0.9113 | 0.9131 | 0.9848 |
### Framework Versions
- Python: 3.8.16
- SpanMarker: 1.5.0
- Transformers: 4.29.0.dev0
- PyTorch: 1.10.1
- Datasets: 2.15.0
- Tokenizers: 0.13.2
## Citation
### BibTeX
```
@software{Aarsen_SpanMarker,
author = {Aarsen, Tom},
license = {Apache-2.0},
title = {{SpanMarker for Named Entity Recognition}},
url = {https://github.com/tomaarsen/SpanMarkerNER}
}
```
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
--> |