|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- amazon_reviews_multi |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: test-xlm-roberta-base-amzaon-reviews-mlm |
|
results: |
|
- task: |
|
name: Masked Language Modeling |
|
type: fill-mask |
|
dataset: |
|
name: amazon_reviews_multi all_languages |
|
type: amazon_reviews_multi |
|
args: all_languages |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.5032103794889962 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# test-xlm-roberta-base-amzaon-reviews-mlm |
|
|
|
This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co./xlm-roberta-base) on the amazon_reviews_multi all_languages dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.1091 |
|
- Accuracy: 0.5032 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0.dev0 |
|
- Pytorch 1.10.0 |
|
- Datasets 1.18.3 |
|
- Tokenizers 0.11.0 |
