---
library_name: transformers
license: mit
base_model: deepset/gbert-large
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
model-index:
- name: gbert_synset_classifier_pair
  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. -->

# gbert_synset_classifier_pair

This model is a fine-tuned version of [deepset/gbert-large](https://huggingface.co/deepset/gbert-large) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.4826
- Accuracy: 0.8551
- F1: 0.8501
- Precision: 0.8590
- Recall: 0.8551
- F1 Macro: 0.7419
- Precision Macro: 0.7334
- Recall Macro: 0.7673
- F1 Micro: 0.8551
- Precision Micro: 0.8551
- Recall Micro: 0.8551

## 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: 2e-05
- train_batch_size: 20
- eval_batch_size: 20
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 80
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 5
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1     | Precision | Recall | F1 Macro | Precision Macro | Recall Macro | F1 Micro | Precision Micro | Recall Micro |
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|
| 2.7743        | 0.3891 | 100  | 1.0270          | 0.7995   | 0.7723 | 0.7757    | 0.7995 | 0.5034   | 0.5166          | 0.5282       | 0.7995   | 0.7995          | 0.7995       |
| 0.8957        | 0.7782 | 200  | 0.6587          | 0.8291   | 0.8176 | 0.8179    | 0.8291 | 0.5871   | 0.5863          | 0.6064       | 0.8291   | 0.8291          | 0.8291       |
| 0.6278        | 1.1673 | 300  | 0.5440          | 0.8457   | 0.8363 | 0.8395    | 0.8457 | 0.6407   | 0.6380          | 0.6622       | 0.8457   | 0.8457          | 0.8457       |
| 0.5132        | 1.5564 | 400  | 0.5295          | 0.8362   | 0.8275 | 0.8355    | 0.8362 | 0.6593   | 0.6525          | 0.6857       | 0.8362   | 0.8362          | 0.8362       |
| 0.4843        | 1.9455 | 500  | 0.4930          | 0.8511   | 0.8439 | 0.8500    | 0.8511 | 0.6777   | 0.6675          | 0.7028       | 0.8511   | 0.8511          | 0.8511       |
| 0.39          | 2.3346 | 600  | 0.4827          | 0.8564   | 0.8521 | 0.8555    | 0.8564 | 0.7073   | 0.6989          | 0.7287       | 0.8564   | 0.8564          | 0.8564       |
| 0.3536        | 2.7237 | 700  | 0.4818          | 0.8551   | 0.8492 | 0.8576    | 0.8551 | 0.7314   | 0.7421          | 0.7476       | 0.8551   | 0.8551          | 0.8551       |
| 0.3462        | 3.1128 | 800  | 0.4826          | 0.8551   | 0.8501 | 0.8590    | 0.8551 | 0.7419   | 0.7334          | 0.7673       | 0.8551   | 0.8551          | 0.8551       |


### Framework versions

- Transformers 4.45.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.20.3