Yenreh/xlm-roberta-large-tass-sentiment-bs8

XLM-RoBERTa Large Fine-tuned for Spanish Sentiment Analysis (TASS)

Model Description

This model is a fine-tuned version of FacebookAI/xlm-roberta-large for sentiment analysis on Spanish Twitter data (TASS dataset).

Training Details

• Base Model: xlm-roberta-large
• Task: Multi-class Sentiment Classification (Negative/Neutral/Positive)
• Dataset: TASS (Twitter Analysis Sentiment Seminar)
• Training Samples: 4,636
• Validation Samples: 1,159
• Test Samples: 1,449
• Batch Size: 8
• Epochs: 10
• Learning Rate: 2e-05
• Weight Decay: 0.01
• Max Sequence Length: 256
• Class Balancing: Weighted Cross-Entropy Loss
• Early Stopping: Enabled (patience=3)

Performance (Test Set)

Metric	Score
F1 Score	0.6876
Accuracy	0.6888
Precision	0.6874
Recall	0.6888

Training History (Validation Set)

Metrics per epoch during training:

Epoch	Loss	Accuracy	F1 Score	Precision	Recall
1	1.0415	0.3796	0.2859	0.2294	0.3796
2	0.8871	0.5949	0.5811	0.6149	0.5949
3	0.9549	0.5528	0.4932	0.6216	0.5528
4	0.9607	0.6059	0.5713	0.6096	0.6059
5	0.9084	0.6335	0.6301	0.6315	0.6335
6	0.8614	0.6722	0.6702	0.6696	0.6722
7	0.8683	0.6798	0.6774	0.6762	0.6798
8	1.1485	0.6798	0.6786	0.6815	0.6798
9	1.3851	0.6888	0.6876	0.6874	0.6888
10	1.5043	0.6770	0.6762	0.6759	0.6770

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Cargar modelo y tokenizer
tokenizer = AutoTokenizer.from_pretrained("tu-usuario/xlm-roberta-large-tass-sentiment-bs8")
model = AutoModelForSequenceClassification.from_pretrained("tu-usuario/xlm-roberta-large-tass-sentiment-bs8")

# Ejemplo de uso
text = "Me encanta este producto, es excelente"
inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=256)

with torch.no_grad():
    outputs = model(**inputs)
    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
    predicted_class = torch.argmax(predictions, dim=-1).item()

labels = {0: "Negativo", 1: "Neutral", 2: "Positivo"}
print(f"Sentimiento: {labels[predicted_class]}")
print(f"Confianza: {predictions[0][predicted_class].item():.4f}")

Como Pipeline

from transformers import pipeline

# Usar como pipeline
classifier = pipeline('sentiment-analysis', model='tu-usuario/xlm-roberta-large-tass-sentiment-bs8')

result = classifier("Me encanta este producto, es excelente")
print(result)
# Output: [{'label': 'LABEL_2', 'score': 0.95}]
# LABEL_0 = Negativo, LABEL_1 = Neutral, LABEL_2 = Positivo

Labels

• 0 (LABEL_0): Negative sentiment
• 1 (LABEL_1): Neutral sentiment
• 2 (LABEL_2): Positive sentiment

Training Configuration

The model was trained with weighted loss to handle class imbalance.

Distribution in training set (estimated):

• Negative samples: 1854 (40%)
• Neutral samples: 1391 (30%)
• Positive samples: 1391 (30%)

Limitations and Bias

• This model is specifically trained on Spanish Twitter data
• Performance may vary on other Spanish text domains
• The model classifies sentiment into three categories (negative, neutral, positive)
• May reflect biases present in the TASS Twitter dataset