File size: 1,633 Bytes
b3e1ad2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
---
language:
- en
license: apache-2.0
base_model: bert-base-uncased
tags:
- mhs
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: bert_base_uncased
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# bert_base_uncased
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the word-puzzle dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4856
- Accuracy: 0.8500
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
| No log | 1.0 | 23 | 1.0218 | 0.6000 |
| No log | 2.0 | 46 | 0.6112 | 0.7250 |
| No log | 3.0 | 69 | 0.4168 | 0.875 |
| No log | 4.0 | 92 | 0.4505 | 0.8500 |
| No log | 5.0 | 115 | 0.4856 | 0.8500 |
### Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0
|