|
--- |
|
datasets: |
|
- relbert/semeval2012_relational_similarity_v6 |
|
model-index: |
|
- name: relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1 |
|
results: |
|
- task: |
|
name: Relation Mapping |
|
type: sorting-task |
|
dataset: |
|
name: Relation Mapping |
|
args: relbert/relation_mapping |
|
type: relation-mapping |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.8430952380952381 |
|
- task: |
|
name: Analogy Questions (SAT full) |
|
type: multiple-choice-qa |
|
dataset: |
|
name: SAT full |
|
args: relbert/analogy_questions |
|
type: analogy-questions |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.3582887700534759 |
|
- task: |
|
name: Analogy Questions (SAT) |
|
type: multiple-choice-qa |
|
dataset: |
|
name: SAT |
|
args: relbert/analogy_questions |
|
type: analogy-questions |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.3649851632047478 |
|
- task: |
|
name: Analogy Questions (BATS) |
|
type: multiple-choice-qa |
|
dataset: |
|
name: BATS |
|
args: relbert/analogy_questions |
|
type: analogy-questions |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.4280155642023346 |
|
- task: |
|
name: Analogy Questions (Google) |
|
type: multiple-choice-qa |
|
dataset: |
|
name: Google |
|
args: relbert/analogy_questions |
|
type: analogy-questions |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.532 |
|
- task: |
|
name: Analogy Questions (U2) |
|
type: multiple-choice-qa |
|
dataset: |
|
name: U2 |
|
args: relbert/analogy_questions |
|
type: analogy-questions |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.3333333333333333 |
|
- task: |
|
name: Analogy Questions (U4) |
|
type: multiple-choice-qa |
|
dataset: |
|
name: U4 |
|
args: relbert/analogy_questions |
|
type: analogy-questions |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.3101851851851852 |
|
- task: |
|
name: Lexical Relation Classification (BLESS) |
|
type: classification |
|
dataset: |
|
name: BLESS |
|
args: relbert/lexical_relation_classification |
|
type: relation-classification |
|
metrics: |
|
- name: F1 |
|
type: f1 |
|
value: 0.8460147657073979 |
|
- name: F1 (macro) |
|
type: f1_macro |
|
value: 0.8315897128108677 |
|
- task: |
|
name: Lexical Relation Classification (CogALexV) |
|
type: classification |
|
dataset: |
|
name: CogALexV |
|
args: relbert/lexical_relation_classification |
|
type: relation-classification |
|
metrics: |
|
- name: F1 |
|
type: f1 |
|
value: 0.8084507042253521 |
|
- name: F1 (macro) |
|
type: f1_macro |
|
value: 0.5269777075808457 |
|
- task: |
|
name: Lexical Relation Classification (EVALution) |
|
type: classification |
|
dataset: |
|
name: BLESS |
|
args: relbert/lexical_relation_classification |
|
type: relation-classification |
|
metrics: |
|
- name: F1 |
|
type: f1 |
|
value: 0.6424702058504875 |
|
- name: F1 (macro) |
|
type: f1_macro |
|
value: 0.6178608994596904 |
|
- task: |
|
name: Lexical Relation Classification (K&H+N) |
|
type: classification |
|
dataset: |
|
name: K&H+N |
|
args: relbert/lexical_relation_classification |
|
type: relation-classification |
|
metrics: |
|
- name: F1 |
|
type: f1 |
|
value: 0.913612019197329 |
|
- name: F1 (macro) |
|
type: f1_macro |
|
value: 0.7738790468743169 |
|
- task: |
|
name: Lexical Relation Classification (ROOT09) |
|
type: classification |
|
dataset: |
|
name: ROOT09 |
|
args: relbert/lexical_relation_classification |
|
type: relation-classification |
|
metrics: |
|
- name: F1 |
|
type: f1 |
|
value: 0.8693199623942337 |
|
- name: F1 (macro) |
|
type: f1_macro |
|
value: 0.864532922094076 |
|
|
|
--- |
|
# relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1 |
|
|
|
RelBERT fine-tuned from [roberta-base](https://huggingface.co/roberta-base) on |
|
[relbert/semeval2012_relational_similarity_v6](https://huggingface.co/datasets/relbert/semeval2012_relational_similarity_v6). |
|
Fine-tuning is done via [RelBERT](https://github.com/asahi417/relbert) library (see the repository for more detail). |
|
It achieves the following results on the relation understanding tasks: |
|
- Analogy Question ([dataset](https://huggingface.co/datasets/relbert/analogy_questions), [full result](https://huggingface.co/relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1/raw/main/analogy.json)): |
|
- Accuracy on SAT (full): 0.3582887700534759 |
|
- Accuracy on SAT: 0.3649851632047478 |
|
- Accuracy on BATS: 0.4280155642023346 |
|
- Accuracy on U2: 0.3333333333333333 |
|
- Accuracy on U4: 0.3101851851851852 |
|
- Accuracy on Google: 0.532 |
|
- Lexical Relation Classification ([dataset](https://huggingface.co/datasets/relbert/lexical_relation_classification), [full result](https://huggingface.co/relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1/raw/main/classification.json)): |
|
- Micro F1 score on BLESS: 0.8460147657073979 |
|
- Micro F1 score on CogALexV: 0.8084507042253521 |
|
- Micro F1 score on EVALution: 0.6424702058504875 |
|
- Micro F1 score on K&H+N: 0.913612019197329 |
|
- Micro F1 score on ROOT09: 0.8693199623942337 |
|
- Relation Mapping ([dataset](https://huggingface.co/datasets/relbert/relation_mapping), [full result](https://huggingface.co/relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1/raw/main/relation_mapping.json)): |
|
- Accuracy on Relation Mapping: 0.8430952380952381 |
|
|
|
|
|
### Usage |
|
This model can be used through the [relbert library](https://github.com/asahi417/relbert). Install the library via pip |
|
```shell |
|
pip install relbert |
|
``` |
|
and activate model as below. |
|
```python |
|
from relbert import RelBERT |
|
model = RelBERT("relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1") |
|
vector = model.get_embedding(['Tokyo', 'Japan']) # shape of (1024, ) |
|
``` |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- model: roberta-base |
|
- max_length: 64 |
|
- mode: mask |
|
- data: relbert/semeval2012_relational_similarity_v6 |
|
- split: train |
|
- split_eval: validation |
|
- template_mode: manual |
|
- loss_function: triplet |
|
- classification_loss: False |
|
- temperature_nce_constant: 0.05 |
|
- temperature_nce_rank: {'min': 0.01, 'max': 0.05, 'type': 'linear'} |
|
- epoch: 9 |
|
- batch: 128 |
|
- lr: 5e-06 |
|
- lr_decay: False |
|
- lr_warmup: 1 |
|
- weight_decay: 0 |
|
- random_seed: 1 |
|
- exclude_relation: None |
|
- n_sample: 320 |
|
- gradient_accumulation: 8 |
|
- relation_level: None |
|
|
|
The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-triplet-1/raw/main/trainer_config.json). |
|
|
|
### Reference |
|
If you use any resource from RelBERT, please consider to cite our [paper](https://aclanthology.org/2021.eacl-demos.7/). |
|
|
|
``` |
|
|
|
@inproceedings{ushio-etal-2021-distilling-relation-embeddings, |
|
title = "{D}istilling {R}elation {E}mbeddings from {P}re-trained {L}anguage {M}odels", |
|
author = "Ushio, Asahi and |
|
Schockaert, Steven and |
|
Camacho-Collados, Jose", |
|
booktitle = "EMNLP 2021", |
|
year = "2021", |
|
address = "Online", |
|
publisher = "Association for Computational Linguistics", |
|
} |
|
|
|
``` |
|
|
