update model card README.md
Browse files
README.md
CHANGED
|
@@ -14,43 +14,43 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
-
- Loss: 0.
|
| 18 |
-
- Benefactive Precision:
|
| 19 |
-
- Benefactive Recall: 0.
|
| 20 |
-
- Benefactive F1: 0.
|
| 21 |
- Benefactive Number: 2
|
| 22 |
-
- Causator Precision:
|
| 23 |
- Causator Recall: 1.0
|
| 24 |
-
- Causator F1:
|
| 25 |
- Causator Number: 12
|
| 26 |
-
- Cause Precision: 0
|
| 27 |
-
- Cause Recall: 0.
|
| 28 |
-
- Cause F1: 0.
|
| 29 |
- Cause Number: 5
|
| 30 |
-
- Contrsubject Precision: 0.
|
| 31 |
-
- Contrsubject Recall: 0.
|
| 32 |
-
- Contrsubject F1: 0.
|
| 33 |
- Contrsubject Number: 9
|
| 34 |
-
- Deliberative Precision:
|
| 35 |
- Deliberative Recall: 1.0
|
| 36 |
-
- Deliberative F1:
|
| 37 |
- Deliberative Number: 4
|
| 38 |
-
- Experiencer Precision: 0.
|
| 39 |
-
- Experiencer Recall: 0.
|
| 40 |
-
- Experiencer F1: 0.
|
| 41 |
- Experiencer Number: 79
|
| 42 |
-
- Object Precision: 0.
|
| 43 |
-
- Object Recall: 0.
|
| 44 |
-
- Object F1: 0.
|
| 45 |
- Object Number: 149
|
| 46 |
-
- Predicate Precision: 0.
|
| 47 |
-
- Predicate Recall: 0.
|
| 48 |
-
- Predicate F1: 0.
|
| 49 |
- Predicate Number: 260
|
| 50 |
-
- Overall Precision: 0.
|
| 51 |
-
- Overall Recall: 0.
|
| 52 |
-
- Overall F1: 0.
|
| 53 |
-
- Overall Accuracy: 0.
|
| 54 |
|
| 55 |
## Model description
|
| 56 |
|
|
@@ -69,25 +69,24 @@ More information needed
|
|
| 69 |
### Training hyperparameters
|
| 70 |
|
| 71 |
The following hyperparameters were used during training:
|
| 72 |
-
- learning_rate: 0.
|
| 73 |
-
- train_batch_size:
|
| 74 |
- eval_batch_size: 1
|
| 75 |
-
- seed:
|
|
|
|
|
|
|
| 76 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 77 |
- lr_scheduler_type: linear
|
| 78 |
-
- lr_scheduler_warmup_ratio: 0.
|
| 79 |
-
- num_epochs:
|
| 80 |
- mixed_precision_training: Native AMP
|
| 81 |
|
| 82 |
### Training results
|
| 83 |
|
| 84 |
-
| Training Loss | Epoch | Step
|
| 85 |
-
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.124 | 3.0 | 14592 | 0.2990 | 1.0 | 0.5 | 0.6667 | 2 | 1.0 | 1.0 | 1.0 | 12 | 0.0 | 0.0 | 0.0 | 5 | 0.75 | 0.6667 | 0.7059 | 9 | 1.0 | 0.5 | 0.6667 | 4 | 0.7386 | 0.8228 | 0.7784 | 79 | 0.8 | 0.6980 | 0.7455 | 149 | 0.9885 | 0.9885 | 0.9885 | 260 | 0.8904 | 0.8596 | 0.8748 | 0.9435 |
|
| 89 |
-
| 0.104 | 4.0 | 19456 | 0.2852 | 1.0 | 0.5 | 0.6667 | 2 | 0.9231 | 1.0 | 0.9600 | 12 | 0.4286 | 0.6 | 0.5 | 5 | 0.6 | 0.6667 | 0.6316 | 9 | 1.0 | 0.75 | 0.8571 | 4 | 0.7253 | 0.8354 | 0.7765 | 79 | 0.7044 | 0.7517 | 0.7273 | 149 | 0.9847 | 0.9885 | 0.9866 | 260 | 0.8440 | 0.8846 | 0.8638 | 0.9359 |
|
| 90 |
-
| 0.0918 | 5.0 | 24320 | 0.3357 | 1.0 | 0.5 | 0.6667 | 2 | 1.0 | 1.0 | 1.0 | 12 | 0.4 | 0.4 | 0.4000 | 5 | 0.75 | 0.6667 | 0.7059 | 9 | 1.0 | 1.0 | 1.0 | 4 | 0.7442 | 0.8101 | 0.7758 | 79 | 0.7551 | 0.7450 | 0.7500 | 149 | 0.9809 | 0.9885 | 0.9847 | 260 | 0.8705 | 0.8788 | 0.8746 | 0.9411 |
|
| 91 |
|
| 92 |
|
| 93 |
### Framework versions
|
|
|
|
| 14 |
|
| 15 |
This model is a fine-tuned version of [cointegrated/rubert-tiny2](https://huggingface.co/cointegrated/rubert-tiny2) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
+
- Loss: 0.2006
|
| 18 |
+
- Benefactive Precision: 0.0
|
| 19 |
+
- Benefactive Recall: 0.0
|
| 20 |
+
- Benefactive F1: 0.0
|
| 21 |
- Benefactive Number: 2
|
| 22 |
+
- Causator Precision: 0.8571
|
| 23 |
- Causator Recall: 1.0
|
| 24 |
+
- Causator F1: 0.9231
|
| 25 |
- Causator Number: 12
|
| 26 |
+
- Cause Precision: 1.0
|
| 27 |
+
- Cause Recall: 0.2
|
| 28 |
+
- Cause F1: 0.3333
|
| 29 |
- Cause Number: 5
|
| 30 |
+
- Contrsubject Precision: 0.6
|
| 31 |
+
- Contrsubject Recall: 0.3333
|
| 32 |
+
- Contrsubject F1: 0.4286
|
| 33 |
- Contrsubject Number: 9
|
| 34 |
+
- Deliberative Precision: 0.8
|
| 35 |
- Deliberative Recall: 1.0
|
| 36 |
+
- Deliberative F1: 0.8889
|
| 37 |
- Deliberative Number: 4
|
| 38 |
+
- Experiencer Precision: 0.7160
|
| 39 |
+
- Experiencer Recall: 0.7342
|
| 40 |
+
- Experiencer F1: 0.7250
|
| 41 |
- Experiencer Number: 79
|
| 42 |
+
- Object Precision: 0.7203
|
| 43 |
+
- Object Recall: 0.6913
|
| 44 |
+
- Object F1: 0.7055
|
| 45 |
- Object Number: 149
|
| 46 |
+
- Predicate Precision: 0.9847
|
| 47 |
+
- Predicate Recall: 0.9923
|
| 48 |
+
- Predicate F1: 0.9885
|
| 49 |
- Predicate Number: 260
|
| 50 |
+
- Overall Precision: 0.8591
|
| 51 |
+
- Overall Recall: 0.8442
|
| 52 |
+
- Overall F1: 0.8516
|
| 53 |
+
- Overall Accuracy: 0.9331
|
| 54 |
|
| 55 |
## Model description
|
| 56 |
|
|
|
|
| 69 |
### Training hyperparameters
|
| 70 |
|
| 71 |
The following hyperparameters were used during training:
|
| 72 |
+
- learning_rate: 0.00018632464179881193
|
| 73 |
+
- train_batch_size: 4
|
| 74 |
- eval_batch_size: 1
|
| 75 |
+
- seed: 755657
|
| 76 |
+
- gradient_accumulation_steps: 4
|
| 77 |
+
- total_train_batch_size: 16
|
| 78 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 79 |
- lr_scheduler_type: linear
|
| 80 |
+
- lr_scheduler_warmup_ratio: 0.02
|
| 81 |
+
- num_epochs: 2
|
| 82 |
- mixed_precision_training: Native AMP
|
| 83 |
|
| 84 |
### Training results
|
| 85 |
|
| 86 |
+
| Training Loss | Epoch | Step | Validation Loss | Benefactive Precision | Benefactive Recall | Benefactive F1 | Benefactive Number | Causator Precision | Causator Recall | Causator F1 | Causator Number | Cause Precision | Cause Recall | Cause F1 | Cause Number | Contrsubject Precision | Contrsubject Recall | Contrsubject F1 | Contrsubject Number | Deliberative Precision | Deliberative Recall | Deliberative F1 | Deliberative Number | Experiencer Precision | Experiencer Recall | Experiencer F1 | Experiencer Number | Object Precision | Object Recall | Object F1 | Object Number | Predicate Precision | Predicate Recall | Predicate F1 | Predicate Number | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
|
| 87 |
+
|:-------------:|:-----:|:----:|:---------------:|:---------------------:|:------------------:|:--------------:|:------------------:|:------------------:|:---------------:|:-----------:|:---------------:|:---------------:|:------------:|:--------:|:------------:|:----------------------:|:-------------------:|:---------------:|:-------------------:|:----------------------:|:-------------------:|:---------------:|:-------------------:|:---------------------:|:------------------:|:--------------:|:------------------:|:----------------:|:-------------:|:---------:|:-------------:|:-------------------:|:----------------:|:------------:|:----------------:|:-----------------:|:--------------:|:----------:|:----------------:|
|
| 88 |
+
| 0.2606 | 1.0 | 304 | 0.2313 | 0.0 | 0.0 | 0.0 | 2 | 0.7143 | 0.8333 | 0.7692 | 12 | 0.5 | 0.2 | 0.2857 | 5 | 1.0 | 0.2222 | 0.3636 | 9 | 1.0 | 0.25 | 0.4 | 4 | 0.8372 | 0.4557 | 0.5902 | 79 | 0.8 | 0.5101 | 0.6230 | 149 | 0.9846 | 0.9846 | 0.9846 | 260 | 0.9161 | 0.7346 | 0.8154 | 0.9217 |
|
| 89 |
+
| 0.1565 | 2.0 | 608 | 0.2006 | 0.0 | 0.0 | 0.0 | 2 | 0.8571 | 1.0 | 0.9231 | 12 | 1.0 | 0.2 | 0.3333 | 5 | 0.6 | 0.3333 | 0.4286 | 9 | 0.8 | 1.0 | 0.8889 | 4 | 0.7160 | 0.7342 | 0.7250 | 79 | 0.7203 | 0.6913 | 0.7055 | 149 | 0.9847 | 0.9923 | 0.9885 | 260 | 0.8591 | 0.8442 | 0.8516 | 0.9331 |
|
|
|
|
|
|
|
|
|
|
| 90 |
|
| 91 |
|
| 92 |
### Framework versions
|