asahi417 commited on
Commit
a301578
1 Parent(s): 536b753

model update

Browse files
Files changed (1) hide show
  1. README.md +9 -9
README.md CHANGED
@@ -2,7 +2,7 @@
2
  datasets:
3
  - relbert/conceptnet_high_confidence
4
  model-index:
5
- - name: relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce
6
  results:
7
  - task:
8
  name: Relation Mapping
@@ -14,7 +14,7 @@ model-index:
14
  metrics:
15
  - name: Accuracy
16
  type: accuracy
17
- value: 0.8696428571428572
18
  - task:
19
  name: Analogy Questions (SAT full)
20
  type: multiple-choice-qa
@@ -153,27 +153,27 @@ model-index:
153
  value: 0.9012451685993444
154
 
155
  ---
156
- # relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce
157
 
158
  RelBERT fine-tuned from [roberta-large](https://huggingface.co/roberta-large) on
159
  [relbert/conceptnet_high_confidence](https://huggingface.co/datasets/relbert/conceptnet_high_confidence).
160
  Fine-tuning is done via [RelBERT](https://github.com/asahi417/relbert) library (see the repository for more detail).
161
  It achieves the following results on the relation understanding tasks:
162
- - Analogy Question ([dataset](https://huggingface.co/datasets/relbert/analogy_questions), [full result](https://huggingface.co/relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/analogy.json)):
163
  - Accuracy on SAT (full): 0.5026737967914439
164
  - Accuracy on SAT: 0.5074183976261127
165
  - Accuracy on BATS: 0.7837687604224569
166
  - Accuracy on U2: 0.4868421052631579
167
  - Accuracy on U4: 0.5717592592592593
168
  - Accuracy on Google: 0.914
169
- - Lexical Relation Classification ([dataset](https://huggingface.co/datasets/relbert/lexical_relation_classification), [full result](https://huggingface.co/relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/classification.json)):
170
  - Micro F1 score on BLESS: 0.9169805635076088
171
  - Micro F1 score on CogALexV: 0.8615023474178404
172
  - Micro F1 score on EVALution: 0.6917659804983749
173
  - Micro F1 score on K&H+N: 0.9652917854907144
174
  - Micro F1 score on ROOT09: 0.9025383892196804
175
- - Relation Mapping ([dataset](https://huggingface.co/datasets/relbert/relation_mapping), [full result](https://huggingface.co/relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/relation_mapping.json)):
176
- - Accuracy on Relation Mapping: 0.8696428571428572
177
 
178
 
179
  ### Usage
@@ -184,7 +184,7 @@ pip install relbert
184
  and activate model as below.
185
  ```python
186
  from relbert import RelBERT
187
- model = RelBERT("relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce")
188
  vector = model.get_embedding(['Tokyo', 'Japan']) # shape of (1024, )
189
  ```
190
 
@@ -211,7 +211,7 @@ The following hyperparameters were used during training:
211
  - n_sample: 640
212
  - gradient_accumulation: 8
213
 
214
- The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/relbert/relbert-roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/trainer_config.json).
215
 
216
  ### Reference
217
  If you use any resource from RelBERT, please consider to cite our [paper](https://aclanthology.org/2021.eacl-demos.7/).
 
2
  datasets:
3
  - relbert/conceptnet_high_confidence
4
  model-index:
5
+ - name: relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce
6
  results:
7
  - task:
8
  name: Relation Mapping
 
14
  metrics:
15
  - name: Accuracy
16
  type: accuracy
17
+ value: 0.844484126984127
18
  - task:
19
  name: Analogy Questions (SAT full)
20
  type: multiple-choice-qa
 
153
  value: 0.9012451685993444
154
 
155
  ---
156
+ # relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce
157
 
158
  RelBERT fine-tuned from [roberta-large](https://huggingface.co/roberta-large) on
159
  [relbert/conceptnet_high_confidence](https://huggingface.co/datasets/relbert/conceptnet_high_confidence).
160
  Fine-tuning is done via [RelBERT](https://github.com/asahi417/relbert) library (see the repository for more detail).
161
  It achieves the following results on the relation understanding tasks:
162
+ - Analogy Question ([dataset](https://huggingface.co/datasets/relbert/analogy_questions), [full result](https://huggingface.co/relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/analogy.json)):
163
  - Accuracy on SAT (full): 0.5026737967914439
164
  - Accuracy on SAT: 0.5074183976261127
165
  - Accuracy on BATS: 0.7837687604224569
166
  - Accuracy on U2: 0.4868421052631579
167
  - Accuracy on U4: 0.5717592592592593
168
  - Accuracy on Google: 0.914
169
+ - Lexical Relation Classification ([dataset](https://huggingface.co/datasets/relbert/lexical_relation_classification), [full result](https://huggingface.co/relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/classification.json)):
170
  - Micro F1 score on BLESS: 0.9169805635076088
171
  - Micro F1 score on CogALexV: 0.8615023474178404
172
  - Micro F1 score on EVALution: 0.6917659804983749
173
  - Micro F1 score on K&H+N: 0.9652917854907144
174
  - Micro F1 score on ROOT09: 0.9025383892196804
175
+ - Relation Mapping ([dataset](https://huggingface.co/datasets/relbert/relation_mapping), [full result](https://huggingface.co/relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/relation_mapping.json)):
176
+ - Accuracy on Relation Mapping: 0.844484126984127
177
 
178
 
179
  ### Usage
 
184
  and activate model as below.
185
  ```python
186
  from relbert import RelBERT
187
+ model = RelBERT("relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce")
188
  vector = model.get_embedding(['Tokyo', 'Japan']) # shape of (1024, )
189
  ```
190
 
 
211
  - n_sample: 640
212
  - gradient_accumulation: 8
213
 
214
+ The full configuration can be found at [fine-tuning parameter file](https://huggingface.co/relbert/roberta-large-conceptnet-hc-mask-prompt-b-nce/raw/main/trainer_config.json).
215
 
216
  ### Reference
217
  If you use any resource from RelBERT, please consider to cite our [paper](https://aclanthology.org/2021.eacl-demos.7/).