---
library_name: transformers
license: mit
base_model: lazymonster/20k-combined-raw-scratch
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: adaptation_raw
  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. -->

# adaptation_raw

This model is a fine-tuned version of [lazymonster/20k-combined-raw-scratch](https://huggingface.co/lazymonster/20k-combined-raw-scratch) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0900
- Accuracy: 0.6965
- F1 Macro: 0.6952
- Composite Score: 0.6952

## 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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Composite Score |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:---------------:|
| No log        | 1.0   | 333  | 0.9958          | 0.6995   | 0.6983   | 0.6983          |
| 0.4563        | 2.0   | 666  | 1.0290          | 0.7104   | 0.7096   | 0.7096          |
| 0.4563        | 3.0   | 999  | 1.0838          | 0.6859   | 0.6840   | 0.6840          |
| 0.4419        | 4.0   | 1332 | 1.0559          | 0.7006   | 0.6987   | 0.6987          |
| 0.4108        | 5.0   | 1665 | 1.1011          | 0.6946   | 0.6929   | 0.6929          |
| 0.4108        | 6.0   | 1998 | 1.0900          | 0.6965   | 0.6952   | 0.6952          |


### Framework versions

- Transformers 4.53.3
- Pytorch 2.6.0+cu124
- Datasets 4.4.1
- Tokenizers 0.21.2