Commit
•
5f2e867
1
Parent(s):
a64a51b
update model card README.md
Browse files
README.md
CHANGED
|
@@ -7,7 +7,6 @@ datasets:
|
|
| 7 |
- financial_phrasebank
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
| 10 |
-
- f1
|
| 11 |
model-index:
|
| 12 |
- name: distilbert-finance
|
| 13 |
results:
|
|
@@ -23,10 +22,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
-
- name: F1
|
| 28 |
-
type: f1
|
| 29 |
-
value: 0.6955188246097337
|
| 30 |
---
|
| 31 |
|
| 32 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -36,9 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 36 |
|
| 37 |
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the financial_phrasebank dataset.
|
| 38 |
It achieves the following results on the evaluation set:
|
| 39 |
-
- Loss: 1.
|
| 40 |
-
- Accuracy: 0.
|
| 41 |
-
- F1: 0.6955
|
| 42 |
|
| 43 |
## Model description
|
| 44 |
|
|
@@ -57,9 +52,9 @@ More information needed
|
|
| 57 |
### Training hyperparameters
|
| 58 |
|
| 59 |
The following hyperparameters were used during training:
|
| 60 |
-
- learning_rate:
|
| 61 |
-
- train_batch_size:
|
| 62 |
-
- eval_batch_size:
|
| 63 |
- seed: 42
|
| 64 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 65 |
- lr_scheduler_type: linear
|
|
@@ -67,11 +62,43 @@ The following hyperparameters were used during training:
|
|
| 67 |
|
| 68 |
### Training results
|
| 69 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 70 |
|
| 71 |
|
| 72 |
### Framework versions
|
| 73 |
|
| 74 |
- Transformers 4.31.0
|
| 75 |
- Pytorch 2.0.1+cu117
|
| 76 |
-
- Datasets 2.14.
|
| 77 |
- Tokenizers 0.13.3
|
|
|
|
| 7 |
- financial_phrasebank
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 10 |
model-index:
|
| 11 |
- name: distilbert-finance
|
| 12 |
results:
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.6993801652892562
|
|
|
|
|
|
|
|
|
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the financial_phrasebank dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 1.5012
|
| 36 |
+
- Accuracy: 0.6994
|
|
|
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
|
|
|
| 52 |
### Training hyperparameters
|
| 53 |
|
| 54 |
The following hyperparameters were used during training:
|
| 55 |
+
- learning_rate: 0.0002
|
| 56 |
+
- train_batch_size: 64
|
| 57 |
+
- eval_batch_size: 64
|
| 58 |
- seed: 42
|
| 59 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 60 |
- lr_scheduler_type: linear
|
|
|
|
| 62 |
|
| 63 |
### Training results
|
| 64 |
|
| 65 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 66 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 67 |
+
| 0.6623 | 0.33 | 20 | 1.4571 | 0.3709 |
|
| 68 |
+
| 0.424 | 0.66 | 40 | 1.0223 | 0.6126 |
|
| 69 |
+
| 0.4282 | 0.98 | 60 | 1.0824 | 0.6343 |
|
| 70 |
+
| 0.3126 | 1.31 | 80 | 0.9320 | 0.6612 |
|
| 71 |
+
| 0.2464 | 1.64 | 100 | 0.8817 | 0.6963 |
|
| 72 |
+
| 0.2677 | 1.97 | 120 | 0.9278 | 0.6994 |
|
| 73 |
+
| 0.1221 | 2.3 | 140 | 1.7929 | 0.6322 |
|
| 74 |
+
| 0.1392 | 2.62 | 160 | 1.0517 | 0.7004 |
|
| 75 |
+
| 0.1982 | 2.95 | 180 | 1.0295 | 0.6684 |
|
| 76 |
+
| 0.1055 | 3.28 | 200 | 0.9028 | 0.7252 |
|
| 77 |
+
| 0.0704 | 3.61 | 220 | 1.6708 | 0.6188 |
|
| 78 |
+
| 0.0962 | 3.93 | 240 | 1.1233 | 0.7200 |
|
| 79 |
+
| 0.0356 | 4.26 | 260 | 1.1614 | 0.7603 |
|
| 80 |
+
| 0.0553 | 4.59 | 280 | 1.1362 | 0.7149 |
|
| 81 |
+
| 0.0665 | 4.92 | 300 | 1.2905 | 0.6529 |
|
| 82 |
+
| 0.0441 | 5.25 | 320 | 1.5180 | 0.6932 |
|
| 83 |
+
| 0.0425 | 5.57 | 340 | 1.3501 | 0.6808 |
|
| 84 |
+
| 0.0267 | 5.9 | 360 | 1.2062 | 0.7159 |
|
| 85 |
+
| 0.0198 | 6.23 | 380 | 1.3289 | 0.7262 |
|
| 86 |
+
| 0.0228 | 6.56 | 400 | 1.6142 | 0.6694 |
|
| 87 |
+
| 0.0209 | 6.89 | 420 | 1.8779 | 0.6147 |
|
| 88 |
+
| 0.0295 | 7.21 | 440 | 1.1260 | 0.6994 |
|
| 89 |
+
| 0.0138 | 7.54 | 460 | 1.4690 | 0.6756 |
|
| 90 |
+
| 0.0091 | 7.87 | 480 | 1.2774 | 0.7035 |
|
| 91 |
+
| 0.0094 | 8.2 | 500 | 1.8177 | 0.6384 |
|
| 92 |
+
| 0.0075 | 8.52 | 520 | 1.3794 | 0.7004 |
|
| 93 |
+
| 0.0079 | 8.85 | 540 | 1.4167 | 0.6994 |
|
| 94 |
+
| 0.0039 | 9.18 | 560 | 1.4824 | 0.6921 |
|
| 95 |
+
| 0.0023 | 9.51 | 580 | 1.5161 | 0.6932 |
|
| 96 |
+
| 0.005 | 9.84 | 600 | 1.5012 | 0.6994 |
|
| 97 |
|
| 98 |
|
| 99 |
### Framework versions
|
| 100 |
|
| 101 |
- Transformers 4.31.0
|
| 102 |
- Pytorch 2.0.1+cu117
|
| 103 |
+
- Datasets 2.14.4
|
| 104 |
- Tokenizers 0.13.3
|