|
--- |
|
language: es |
|
tags: |
|
- sagemaker |
|
- roberta |
|
- ruperta |
|
- TextClassification |
|
license: apache-2.0 |
|
datasets: |
|
- IMDbreviews_es |
|
model-index: |
|
- name: RuPERTa_base_sentiment_analysis_es |
|
results: |
|
- task: |
|
name: Sentiment Analysis |
|
type: sentiment-analysis |
|
dataset: |
|
name: "IMDb Reviews in Spanish" |
|
type: IMDbreviews_es |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.881866 |
|
- name: F1 Score |
|
type: f1 |
|
value: 0.008272 |
|
- name: Precision |
|
type: precision |
|
value: 0.858605 |
|
- name: Recall |
|
type: recall |
|
value: 0.920062 |
|
## `RuPERTa_base_sentiment_analysis_es` |
|
|
|
This model was trained using Amazon SageMaker and the new Hugging Face Deep Learning container. |
|
|
|
The base model is RuPERTa-base (uncased) which is a RoBERTa model trained on a uncased version of big Spanish corpus. |
|
It was trained by mrm8488, Manuel Romero. |
|
|
|
## Hyperparameters |
|
|
|
{ |
|
"epochs": "4", |
|
"eval_batch_size": "8", |
|
"fp16": "true", |
|
"learning_rate": "3e-05", |
|
"model_name": "\"mrm8488/RuPERTa-base\"", |
|
"sagemaker_container_log_level": "20", |
|
"sagemaker_job_name": "\"ruperta-sentiment-analysis-full-p2-2021-12-06-20-32-27\"", |
|
"sagemaker_program": "\"train.py\"", |
|
"sagemaker_region": "\"us-east-1\"", |
|
"sagemaker_submit_directory": "\"s3://edumunozsala-ml-sagemaker/ruperta-sentiment/ruperta-sentiment-analysis-full-p2-2021-12-06-20-32-27/source/sourcedir.tar.gz\"", |
|
"train_batch_size": "32", |
|
"train_filename": "\"train_data.pt\"", |
|
"val_filename": "\"val_data.pt\"" |
|
} |
|
|
|
|
|
## Usage |
|
|
|
## Results |
|
|
|
epoch = 1.0 |
|
eval_accuracy = 0.8629333333333333 |
|
eval_f1 = 0.8648790746582545 |
|
eval_loss = 0.3160930573940277 |
|
eval_mem_cpu_alloc_delta = 0 |
|
eval_mem_cpu_peaked_delta = 0 |
|
eval_mem_gpu_alloc_delta = 0 |
|
eval_mem_gpu_peaked_delta = 94507520 |
|
eval_precision = 0.8479381443298969 |
|
eval_recall = 0.8825107296137339 |
|
eval_runtime = 114.4994 |
|
eval_samples_per_second = 32.751 |
|
|
|
|