Edit model card

This is a model for text summarization in Spanish. It has been trained on the spanish portion of mlsum. For that, XLM-ProphetNet (a multilingual version of Prophetnet) was used.

For tuning the hyperparameters of the model we used Optuna, with only 10 different trials and 7 initial random trials, as the dataset chosen for training the model was huge. The set of hyperparameters used was the following:

    def hp_space(trial):
        return {
            "learning_rate": trial.suggest_float(
                "learning_rate", 1e-5, 7e-5, log=True
            "num_train_epochs": trial.suggest_categorical(
                "num_train_epochs", [3, 5, 7, 10]
            "per_device_train_batch_size": trial.suggest_categorical(
                "per_device_train_batch_size", [16]),
            "per_device_eval_batch_size": trial.suggest_categorical(
                "per_device_eval_batch_size", [32]),
            "gradient_accumulation_steps": trial.suggest_categorical(
                "gradient_accumulation_steps", [2, 4, 8]),
            "warmup_steps": trial.suggest_categorical(
                "warmup_steps", [50, 100, 500, 1000]
            "weight_decay": trial.suggest_float(
                 "weight_decay", 0.0, 0.1

The reported results are on the test split of mlsum. Complete metrics are:

{"rouge1": 25.1158, "rouge2": 8.4847, "rougeL": 20.6184, "rougeLsum": 20.8948, "gen_len": 19.6496}

This model is really easy to use, and with the following lines of code you can just start summarizing your documents in Spanish:

from transformers import ProphetNetForConditionalGeneration, AutoTokenizer

text = "Hola esto es un ejemplo de texto a resumir. Poco hay que resumir aquí, pero es sólo de muestra."
model_str = "avacaondata/xprophetnet-spanish-mlsum"

tokenizer = AutoTokenizer.from_pretrained(model_str)
model = ProphetNetForConditionalGeneration.from_pretrained(model_str)

input_ids = tokenizer(text, return_tensors="pt").input_ids

output_ids = model.generate(input_ids)[0]
print(tokenizer.decode(output_ids, skip_special_tokens=True))


Thanks to @avacaondata, @alborotis, @albarji, @Dabs, @GuillemGSubies for adding this model.

Downloads last month
Hosted inference API
This model can be loaded on the Inference API on-demand.

Dataset used to train IIC/xprophetnet-spanish-mlsum

Evaluation results