metadata
license: mit
tags:
- generated_from_trainer
datasets:
- dutch_social
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: robbert-twitter-sentiment-custom
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: dutch_social
type: dutch_social
args: dutch_social
metrics:
- name: Accuracy
type: accuracy
value: 0.788
- name: F1
type: f1
value: 0.7878005279207152
- name: Precision
type: precision
value: 0.7877102066609215
- name: Recall
type: recall
value: 0.788
robbert-twitter-sentiment-custom
This model is a fine-tuned version of pdelobelle/robbert-v2-dutch-base on the dutch_social dataset. It achieves the following results on the evaluation set:
- Loss: 0.6656
- Accuracy: 0.788
- F1: 0.7878
- Precision: 0.7877
- Recall: 0.788
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: 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| 0.8287 | 1.0 | 282 | 0.7178 | 0.7007 | 0.6958 | 0.6973 | 0.7007 |
| 0.4339 | 2.0 | 564 | 0.5873 | 0.7667 | 0.7668 | 0.7681 | 0.7667 |
| 0.2045 | 3.0 | 846 | 0.6656 | 0.788 | 0.7878 | 0.7877 | 0.788 |
Framework versions
- Transformers 4.17.0
- Pytorch 1.11.0+cpu
- Datasets 2.0.0
- Tokenizers 0.11.6