model update

README.md

@@ -0,0 +1,235 @@
+---
+datasets:
+- relbert/semeval2012_relational_similarity_v6
+model-index:
+- name: relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-loob-2
+  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: None
+  - 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: 0.4197860962566845
+  - 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: 0.41839762611275966
+  - 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: 0.594774874930517
+  - 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: 0.774
+  - 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: 0.40789473684210525
+  - 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: 0.41898148148148145
+  - 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: 0.9005574807895134
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.8957958532235768
+  - 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: 0.8077464788732395
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.5900936140399187
+  - 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: 0.6359696641386782
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.6206497970461441
+  - 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: 0.9577797871600473
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.8748819835358477
+  - 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: 0.8633657160764651
+    - name: F1 (macro)
+      type: f1_macro
+      value: 0.8605769477843292
+
+---
+# relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-loob-2
+
+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-loob-2/raw/main/analogy.json)):
+    - Accuracy on SAT (full): 0.4197860962566845 
+    - Accuracy on SAT: 0.41839762611275966
+    - Accuracy on BATS: 0.594774874930517
+    - Accuracy on U2: 0.40789473684210525
+    - Accuracy on U4: 0.41898148148148145
+    - Accuracy on Google: 0.774
+- 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-loob-2/raw/main/classification.json)):
+    - Micro F1 score on BLESS: 0.9005574807895134
+    - Micro F1 score on CogALexV: 0.8077464788732395
+    - Micro F1 score on EVALution: 0.6359696641386782
+    - Micro F1 score on K&H+N: 0.9577797871600473
+    - Micro F1 score on ROOT09: 0.8633657160764651
+- Relation Mapping ([dataset](https://huggingface.co/datasets/relbert/relation_mapping), [full result](https://huggingface.co/relbert/relbert-roberta-base-semeval2012-v6-mask-prompt-a-loob-2/raw/main/relation_mapping.json)):
+    - Accuracy on Relation Mapping: None 
+
+
+### 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-loob-2")
+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: info_loob
+ - 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: 2
+ - 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-loob-2/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",
+}
+
+```

analogy.json

@@ -0,0 +1 @@
+{"distance_function": "cosine_similarity", "sat/test": 0.41839762611275966, "sat/valid": 0.43243243243243246, "u2/test": 0.40789473684210525, "u2/valid": 0.375, "u4/test": 0.41898148148148145, "u4/valid": 0.5, "google/test": 0.774, "google/valid": 0.8, "bats/test": 0.594774874930517, "bats/valid": 0.6180904522613065, "sat_full": 0.4197860962566845}

classification.json

@@ -0,0 +1 @@
+{"lexical_relation_classification/BLESS": {"classifier_config": {"activation": "relu", "alpha": 0.0001, "batch_size": "auto", "beta_1": 0.9, "beta_2": 0.999, "early_stopping": false, "epsilon": 1e-08, "hidden_layer_sizes": [100], "learning_rate": "constant", "learning_rate_init": 0.001, "max_fun": 15000, "max_iter": 200, "momentum": 0.9, "n_iter_no_change": 10, "nesterovs_momentum": true, "power_t": 0.5, "random_state": 0, "shuffle": true, "solver": "adam", "tol": 0.0001, "validation_fraction": 0.1, "verbose": false, "warm_start": false}, "test/accuracy": 0.9005574807895134, "test/f1_macro": 0.8957958532235768, "test/f1_micro": 0.9005574807895134, "test/p_macro": 0.8945268113825259, "test/p_micro": 0.9005574807895134, "test/r_macro": 0.8971956945882472, "test/r_micro": 0.9005574807895134}, "lexical_relation_classification/CogALexV": {"classifier_config": {"activation": "relu", "alpha": 0.0001, "batch_size": "auto", "beta_1": 0.9, "beta_2": 0.999, "early_stopping": false, "epsilon": 1e-08, "hidden_layer_sizes": [100], "learning_rate": "constant", "learning_rate_init": 0.001, "max_fun": 15000, "max_iter": 200, "momentum": 0.9, "n_iter_no_change": 10, "nesterovs_momentum": true, "power_t": 0.5, "random_state": 0, "shuffle": true, "solver": "adam", "tol": 0.0001, "validation_fraction": 0.1, "verbose": false, "warm_start": false}, "test/accuracy": 0.8077464788732395, "test/f1_macro": 0.5900936140399187, "test/f1_micro": 0.8077464788732395, "test/p_macro": 0.6441044111094533, "test/p_micro": 0.8077464788732395, "test/r_macro": 0.5540714028639299, "test/r_micro": 0.8077464788732395}, "lexical_relation_classification/EVALution": {"classifier_config": {"activation": "relu", "alpha": 0.0001, "batch_size": "auto", "beta_1": 0.9, "beta_2": 0.999, "early_stopping": false, "epsilon": 1e-08, "hidden_layer_sizes": [100], "learning_rate": "constant", "learning_rate_init": 0.001, "max_fun": 15000, "max_iter": 200, "momentum": 0.9, "n_iter_no_change": 10, "nesterovs_momentum": true, "power_t": 0.5, "random_state": 0, "shuffle": true, "solver": "adam", "tol": 0.0001, "validation_fraction": 0.1, "verbose": false, "warm_start": false}, "test/accuracy": 0.6359696641386782, "test/f1_macro": 0.6206497970461441, "test/f1_micro": 0.6359696641386782, "test/p_macro": 0.6322579427118538, "test/p_micro": 0.6359696641386782, "test/r_macro": 0.6165577882904756, "test/r_micro": 0.6359696641386782}, "lexical_relation_classification/K&H+N": {"classifier_config": {"activation": "relu", "alpha": 0.0001, "batch_size": "auto", "beta_1": 0.9, "beta_2": 0.999, "early_stopping": false, "epsilon": 1e-08, "hidden_layer_sizes": [100], "learning_rate": "constant", "learning_rate_init": 0.001, "max_fun": 15000, "max_iter": 200, "momentum": 0.9, "n_iter_no_change": 10, "nesterovs_momentum": true, "power_t": 0.5, "random_state": 0, "shuffle": true, "solver": "adam", "tol": 0.0001, "validation_fraction": 0.1, "verbose": false, "warm_start": false}, "test/accuracy": 0.9577797871600473, "test/f1_macro": 0.8748819835358477, "test/f1_micro": 0.9577797871600473, "test/p_macro": 0.9088625506469264, "test/p_micro": 0.9577797871600473, "test/r_macro": 0.8492239514605641, "test/r_micro": 0.9577797871600473}, "lexical_relation_classification/ROOT09": {"classifier_config": {"activation": "relu", "alpha": 0.0001, "batch_size": "auto", "beta_1": 0.9, "beta_2": 0.999, "early_stopping": false, "epsilon": 1e-08, "hidden_layer_sizes": [100], "learning_rate": "constant", "learning_rate_init": 0.001, "max_fun": 15000, "max_iter": 200, "momentum": 0.9, "n_iter_no_change": 10, "nesterovs_momentum": true, "power_t": 0.5, "random_state": 0, "shuffle": true, "solver": "adam", "tol": 0.0001, "validation_fraction": 0.1, "verbose": false, "warm_start": false}, "test/accuracy": 0.8633657160764651, "test/f1_macro": 0.8605769477843292, "test/f1_micro": 0.8633657160764651, "test/p_macro": 0.8563746318155546, "test/p_micro": 0.8633657160764651, "test/r_macro": 0.8654137967096123, "test/r_micro": 0.8633657160764651}}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "relbert_output/models/semeval2012-v6.a.info_loob.mask.roberta-base.0.000005.8.0.05.320.2/best_model",
+  "_name_or_path": "roberta-base",
   "architectures": [
     "RobertaModel"
   ],

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1aa706a4cb3bf5d1e6ecb76ea2547e4e68d05a993ea43c42f6a8fec3077af852
-size 498649201
+oid sha256:e2c8d414fa97f364b2c97ed04f9ab1f810c125c0570fdd076e4b740ff5d2d284
+size 498652017

tokenizer_config.json

@@ -6,7 +6,7 @@
   "errors": "replace",
   "mask_token": "<mask>",
   "model_max_length": 512,
-  "name_or_path": "relbert_output/models/semeval2012-v6.a.info_loob.mask.roberta-base.0.000005.8.0.05.320.2/best_model",
+  "name_or_path": "roberta-base",
   "pad_token": "<pad>",
   "sep_token": "</s>",
   "special_tokens_map_file": null,

trainer_config.json

@@ -0,0 +1 @@
+{"model": "roberta-base", "max_length": 64, "mode": "mask", "data": "relbert/semeval2012_relational_similarity_v6", "split": "train", "split_eval": "validation", "template_mode": "manual", "template": "Today, I finally discovered the relation between <subj> and <obj> : <subj> is the <mask> of <obj>", "loss_function": "info_loob", "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": 2, "exclude_relation": null, "n_sample": 320, "gradient_accumulation": 8, "relation_level": null}

validation_loss.json

@@ -0,0 +1 @@
+{"split": "validation", "loss": -18.684419652409265, "data": "relbert/semeval2012_relational_similarity_v6", "exclude_relation": null, "relation_level": null}