|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- consumer-finance-complaints |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- recall |
|
- precision |
|
model-index: |
|
- name: distilbert-base-uncased-wandb-week-3-complaints-classifier-1500 |
|
results: |
|
- task: |
|
name: Text Classification |
|
type: text-classification |
|
dataset: |
|
name: consumer-finance-complaints |
|
type: consumer-finance-complaints |
|
args: default |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.8219254879975536 |
|
- name: F1 |
|
type: f1 |
|
value: 0.8151998307079064 |
|
- name: Recall |
|
type: recall |
|
value: 0.8219254879975536 |
|
- name: Precision |
|
type: precision |
|
value: 0.8165753119578384 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# distilbert-base-uncased-wandb-week-3-complaints-classifier-1500 |
|
|
|
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co./distilbert-base-uncased) on the consumer-finance-complaints dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5451 |
|
- Accuracy: 0.8219 |
|
- F1: 0.8152 |
|
- Recall: 0.8219 |
|
- Precision: 0.8166 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 1500 |
|
- num_epochs: 2 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:---------:| |
|
| 1.0678 | 0.2 | 500 | 0.9935 | 0.7193 | 0.6715 | 0.7193 | 0.6348 | |
|
| 0.8447 | 0.41 | 1000 | 0.8331 | 0.7468 | 0.7108 | 0.7468 | 0.6990 | |
|
| 0.7913 | 0.61 | 1500 | 0.7022 | 0.7770 | 0.7457 | 0.7770 | 0.7685 | |
|
| 0.6973 | 0.82 | 2000 | 0.6584 | 0.7922 | 0.7710 | 0.7922 | 0.7849 | |
|
| 0.5572 | 1.02 | 2500 | 0.6034 | 0.8076 | 0.7986 | 0.8076 | 0.7994 | |
|
| 0.5528 | 1.22 | 3000 | 0.6017 | 0.8085 | 0.7986 | 0.8085 | 0.8063 | |
|
| 0.5435 | 1.43 | 3500 | 0.5721 | 0.8147 | 0.8085 | 0.8147 | 0.8107 | |
|
| 0.4995 | 1.63 | 4000 | 0.5598 | 0.8161 | 0.8125 | 0.8161 | 0.8144 | |
|
| 0.4854 | 1.83 | 4500 | 0.5451 | 0.8219 | 0.8152 | 0.8219 | 0.8166 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.11.0+cu102 |
|
- Datasets 2.3.2 |
|
- Tokenizers 0.12.1 |
|
|
