File size: 2,408 Bytes
88ba91c |
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 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 |
---
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-256
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: 0.8234544620559604
- name: F1
type: f1
value: 0.8176243580045963
- name: Recall
type: recall
value: 0.8234544620559604
- name: Precision
type: precision
value: 0.8171438106054644
---
<!-- 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-256
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: 0.5453
- Accuracy: 0.8235
- F1: 0.8176
- Recall: 0.8235
- Precision: 0.8171
## 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: 4.097565552226687e-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: 256
- num_epochs: 2
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:---------:|
| 0.6691 | 0.61 | 1500 | 0.6475 | 0.7962 | 0.7818 | 0.7962 | 0.7875 |
| 0.5361 | 1.22 | 3000 | 0.5794 | 0.8161 | 0.8080 | 0.8161 | 0.8112 |
| 0.4659 | 1.83 | 4500 | 0.5453 | 0.8235 | 0.8176 | 0.8235 | 0.8171 |
### Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0+cu102
- Datasets 2.3.2
- Tokenizers 0.12.1
|