---
library_name: transformers
license: mit
base_model: xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: xlm-roberta-base-finetuned-panx-de-fr-en-it
  results:
  - task:
      type: token-classification
    dataset:
      name: google/xtreme
      type: google/xtreme
    metrics:
    - name: f1
      type: f1
      value: 0.8380
datasets:
- google/xtreme
language:
- de
- fr
- en
- it
pipeline_tag: token-classification
---

<!-- 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. -->

# xlm-roberta-base-finetuned-panx-de-fr-en-it

This model is a fine-tuned version of [xlm-roberta-base](https://huggingface.co/xlm-roberta-base) on the Xtreme dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2148
- F1: 0.8380

## Model description

This model is fine-tuned for Named Entity Recognition (NER) in German, French, English, and Italian.
It identifies entities like persons, organizations, locations, etc...

## Intended uses & limitations

More information needed

## How to Use
```Python
from transformers import pipeline

# Load the NER pipeline
model_name = "avanishd/xlm-roberta-base-finetuned-panx-de-fr-en-it"
ner_pipeline = pipeline("token-classification", model=model_name, aggregation_strategy="simple")

# Example text (English, but you can use DE/FR/IT as well)
text = "Barack Obama was born in Hawaii and became President of the United States."

# Get NER predictions
entities = ner_pipeline(text)

# Display results
for entity in entities:
    print(f"{entity['word']} → {entity['entity_group']} ({entity['score']:.2f})")

```

## 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: 64
- eval_batch_size: 64
- 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
- num_epochs: 3

### Training results

| Training Loss | Epoch | Step | Validation Loss | F1     |
|:-------------:|:-----:|:----:|:---------------:|:------:|
| No log        | 1.0   | 313  | 0.2470          | 0.7921 |
| No log        | 2.0   | 626  | 0.2170          | 0.8318 |
| No log        | 3.0   | 939  | 0.2148          | 0.8380 |


### Framework versions

- Transformers 4.50.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1