update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,113 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
datasets:
|
| 6 |
+
- amazon_reviews_multi
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
model-index:
|
| 10 |
+
- name: bert_reviews
|
| 11 |
+
results:
|
| 12 |
+
- task:
|
| 13 |
+
name: Text Classification
|
| 14 |
+
type: text-classification
|
| 15 |
+
dataset:
|
| 16 |
+
name: amazon_reviews_multi
|
| 17 |
+
type: amazon_reviews_multi
|
| 18 |
+
config: en
|
| 19 |
+
split: test
|
| 20 |
+
args: en
|
| 21 |
+
metrics:
|
| 22 |
+
- name: Accuracy
|
| 23 |
+
type: accuracy
|
| 24 |
+
value: 0.6408
|
| 25 |
+
---
|
| 26 |
+
|
| 27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 28 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 29 |
+
|
| 30 |
+
# bert_reviews
|
| 31 |
+
|
| 32 |
+
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the amazon_reviews_multi dataset.
|
| 33 |
+
It achieves the following results on the evaluation set:
|
| 34 |
+
- Loss: 0.8312
|
| 35 |
+
- Accuracy: 0.6408
|
| 36 |
+
|
| 37 |
+
## Model description
|
| 38 |
+
|
| 39 |
+
More information needed
|
| 40 |
+
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
+
|
| 43 |
+
More information needed
|
| 44 |
+
|
| 45 |
+
## Training and evaluation data
|
| 46 |
+
|
| 47 |
+
More information needed
|
| 48 |
+
|
| 49 |
+
## Training procedure
|
| 50 |
+
|
| 51 |
+
### Training hyperparameters
|
| 52 |
+
|
| 53 |
+
The following hyperparameters were used during training:
|
| 54 |
+
- learning_rate: 0.0001
|
| 55 |
+
- train_batch_size: 16
|
| 56 |
+
- eval_batch_size: 16
|
| 57 |
+
- seed: 42
|
| 58 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 59 |
+
- lr_scheduler_type: linear
|
| 60 |
+
- training_steps: 20000
|
| 61 |
+
|
| 62 |
+
### Training results
|
| 63 |
+
|
| 64 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 65 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 66 |
+
| 1.1326 | 0.04 | 500 | 1.0019 | 0.5832 |
|
| 67 |
+
| 0.9786 | 0.08 | 1000 | 0.9387 | 0.6086 |
|
| 68 |
+
| 0.9481 | 0.12 | 1500 | 0.9117 | 0.6132 |
|
| 69 |
+
| 0.9334 | 0.16 | 2000 | 0.9440 | 0.5744 |
|
| 70 |
+
| 0.9036 | 0.2 | 2500 | 0.9085 | 0.6034 |
|
| 71 |
+
| 0.9065 | 0.24 | 3000 | 0.9250 | 0.5982 |
|
| 72 |
+
| 0.8821 | 0.28 | 3500 | 0.8917 | 0.6232 |
|
| 73 |
+
| 0.9047 | 0.32 | 4000 | 0.8850 | 0.6258 |
|
| 74 |
+
| 0.8838 | 0.36 | 4500 | 0.8814 | 0.6236 |
|
| 75 |
+
| 0.8732 | 0.4 | 5000 | 0.8874 | 0.6198 |
|
| 76 |
+
| 0.8845 | 0.44 | 5500 | 0.8886 | 0.6164 |
|
| 77 |
+
| 0.874 | 0.48 | 6000 | 0.8665 | 0.634 |
|
| 78 |
+
| 0.8693 | 0.52 | 6500 | 0.8985 | 0.6126 |
|
| 79 |
+
| 0.8502 | 0.56 | 7000 | 0.8992 | 0.6248 |
|
| 80 |
+
| 0.8752 | 0.6 | 7500 | 0.8620 | 0.6326 |
|
| 81 |
+
| 0.8477 | 0.64 | 8000 | 0.8586 | 0.6382 |
|
| 82 |
+
| 0.8456 | 0.68 | 8500 | 0.8603 | 0.631 |
|
| 83 |
+
| 0.861 | 0.72 | 9000 | 0.8536 | 0.628 |
|
| 84 |
+
| 0.8605 | 0.76 | 9500 | 0.8478 | 0.6338 |
|
| 85 |
+
| 0.8159 | 0.8 | 10000 | 0.8569 | 0.6324 |
|
| 86 |
+
| 0.8397 | 0.84 | 10500 | 0.8519 | 0.626 |
|
| 87 |
+
| 0.8424 | 0.88 | 11000 | 0.8753 | 0.6302 |
|
| 88 |
+
| 0.8332 | 0.92 | 11500 | 0.8453 | 0.6326 |
|
| 89 |
+
| 0.8286 | 0.96 | 12000 | 0.8334 | 0.6414 |
|
| 90 |
+
| 0.8166 | 1.0 | 12500 | 0.8508 | 0.633 |
|
| 91 |
+
| 0.7656 | 1.04 | 13000 | 0.8393 | 0.646 |
|
| 92 |
+
| 0.749 | 1.08 | 13500 | 0.8339 | 0.643 |
|
| 93 |
+
| 0.7554 | 1.12 | 14000 | 0.8325 | 0.6486 |
|
| 94 |
+
| 0.734 | 1.16 | 14500 | 0.8467 | 0.6524 |
|
| 95 |
+
| 0.7581 | 1.2 | 15000 | 0.8228 | 0.6434 |
|
| 96 |
+
| 0.7413 | 1.24 | 15500 | 0.8339 | 0.6446 |
|
| 97 |
+
| 0.7429 | 1.28 | 16000 | 0.8331 | 0.6448 |
|
| 98 |
+
| 0.7436 | 1.32 | 16500 | 0.8285 | 0.6472 |
|
| 99 |
+
| 0.7343 | 1.36 | 17000 | 0.8381 | 0.6532 |
|
| 100 |
+
| 0.7225 | 1.4 | 17500 | 0.8327 | 0.6476 |
|
| 101 |
+
| 0.7311 | 1.44 | 18000 | 0.8281 | 0.6506 |
|
| 102 |
+
| 0.7298 | 1.48 | 18500 | 0.8324 | 0.6468 |
|
| 103 |
+
| 0.7409 | 1.52 | 19000 | 0.8180 | 0.648 |
|
| 104 |
+
| 0.732 | 1.56 | 19500 | 0.8209 | 0.6464 |
|
| 105 |
+
| 0.7352 | 1.6 | 20000 | 0.8195 | 0.6468 |
|
| 106 |
+
|
| 107 |
+
|
| 108 |
+
### Framework versions
|
| 109 |
+
|
| 110 |
+
- Transformers 4.28.1
|
| 111 |
+
- Pytorch 2.0.0+cu118
|
| 112 |
+
- Datasets 2.11.0
|
| 113 |
+
- Tokenizers 0.13.3
|