amazon-sagemaker-community/encoder_decoder_es

1---2tags:3- generated_from_trainer4datasets:5- cc_news_es_titles6model-index:7- name: encoder_decoder_es8  results: []9---10 11<!-- This model card has been generated automatically according to the information the Trainer had access to. You12should probably proofread and complete it, then remove this comment. -->13 14# encoder_decoder_es15 16This model is a fine-tuned version of [](https://huggingface.co/) on the cc_news_es_titles dataset.17It achieves the following results on the evaluation set:18- Loss: 7.877319- Rouge2 Precision: 0.00220- Rouge2 Recall: 0.011621- Rouge2 Fmeasure: 0.003422 23## Model description24 25More information needed26 27## Intended uses & limitations28 29More information needed30 31## Training and evaluation data32 33More information needed34 35## Training procedure36 37### Training hyperparameters38 39The following hyperparameters were used during training:40- learning_rate: 0.00341- train_batch_size: 3242- eval_batch_size: 843- seed: 4244- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-0845- lr_scheduler_type: linear46- lr_scheduler_warmup_steps: 50047- num_epochs: 448- mixed_precision_training: Native AMP49 50### Training results51 52| Training Loss | Epoch | Step  | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |53|:-------------:|:-----:|:-----:|:---------------:|:----------------:|:-------------:|:---------------:|54| 7.8807        | 1.0   | 5784  | 7.8976          | 0.0023           | 0.012         | 0.0038          |55| 7.8771        | 2.0   | 11568 | 7.8873          | 0.0018           | 0.0099        | 0.003           |56| 7.8588        | 3.0   | 17352 | 7.8819          | 0.0015           | 0.0085        | 0.0025          |57| 7.8507        | 4.0   | 23136 | 7.8773          | 0.002            | 0.0116        | 0.0034          |58 59 60### Framework versions61 62- Transformers 4.12.363- Pytorch 1.9.164- Datasets 1.15.165- Tokenizers 0.10.366