zhang19991111/scibert-spanmarker-STEM-NER

SpanMarker with allenai/scibert_scivocab_uncased on my-data

This is a SpanMarker model that can be used for Named Entity Recognition. This SpanMarker model uses allenai/scibert_scivocab_uncased as the underlying encoder.

Model Details

Model Description

• Model Type: SpanMarker
• Encoder: allenai/scibert_scivocab_uncased
• Maximum Sequence Length: 256 tokens
• Maximum Entity Length: 8 words
• Language: en
• License: cc-by-sa-4.0

Model Sources

• Repository: SpanMarker on GitHub
• Thesis: SpanMarker For Named Entity Recognition

Model Labels

Label	Examples
Data	"an overall mitochondrial", "defect", "Depth time - series"
Material	"cross - shore measurement locations", "the subject 's fibroblasts", "COXI , COXII and COXIII subunits"
Method	"EFSA", "an approximation", "in vitro"
Process	"translation", "intake", "a significant reduction of synthesis"

Evaluation

Metrics

Label	Precision	Recall	F1
all	0.6981	0.6732	0.6854
Data	0.6269	0.6402	0.6335
Material	0.8085	0.7562	0.7815
Method	0.4211	0.4	0.4103
Process	0.6891	0.6488	0.6683

Uses

Direct Use for Inference

from span_marker import SpanMarkerModel

# Download from the 🤗 Hub
model = SpanMarkerModel.from_pretrained("span-marker-allenai/scibert_scivocab_uncased-me")
# Run inference
entities = model.predict("In situ Peak Force Tapping AFM was employed for determining morphology and nano - mechanical properties of the surface layer .")

Downstream Use

You can finetune this model on your own dataset.

Click to expand

from span_marker import SpanMarkerModel, Trainer

# Download from the 🤗 Hub
model = SpanMarkerModel.from_pretrained("span-marker-allenai/scibert_scivocab_uncased-me")

# 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-allenai/scibert_scivocab_uncased-me-finetuned")

Training Details

Training Set Metrics

Training set	Min	Median	Max
Sentence length	3	25.6049	106
Entities per sentence	0	5.2439	22

Training Hyperparameters

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• 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: 10

Training Results

Epoch	Step	Validation Loss	Validation Precision	Validation Recall	Validation F1	Validation Accuracy
2.0134	300	0.0476	0.7297	0.5821	0.6476	0.7880
4.0268	600	0.0532	0.7537	0.6775	0.7136	0.8281
6.0403	900	0.0655	0.7162	0.7080	0.7121	0.8357
8.0537	1200	0.0761	0.7143	0.7061	0.7102	0.8251

Framework Versions

• Python: 3.10.12
• SpanMarker: 1.5.0
• Transformers: 4.36.2
• PyTorch: 2.0.1+cu118
• Datasets: 2.16.1
• Tokenizers: 0.15.0

Citation

BibTeX

@software{Aarsen_SpanMarker,
    author = {Aarsen, Tom},
    license = {Apache-2.0},
    title = {{SpanMarker for Named Entity Recognition}},
    url = {https://github.com/tomaarsen/SpanMarkerNER}
}