metadata

language:
  - cs
license: mit
tags:
  - text-classification
  - emotion-detection
  - machine-learning
  - czech
model_name: poltextlab/xlm-roberta-large-pooled-czech-emotions9-v2
metrics:
  - precision
  - recall
  - f1-score
  - accuracy
extra_gated_fields:
  Name: text
  Country: country
  Institution: text
  Institution Email: text
  Please specify your academic use case: text
extra_gated_prompt: >-
  Our models are intended for academic use only. If you are not affiliated with
  an academic institution, please provide a rationale for using our models.
  Please allow us a few business days to manually review subscriptions.

poltextlab/xlm-roberta-large-pooled-czech-emotions9-v2

Proposal 2B: German to Czech Emotion Labeling (v2)

Model Description

This model is designed for emotion classification in Czech texts that have been translated from German.
It was fine-tuned to recognize nine emotion categories and trained on a dataset with labeled examples.

Labels and Their Meanings

Label	Emotion
0	Anger
1	Fear
2	Disgust
3	Sadness
4	Joy
5	None of them
6	Enthusiasm
7	Hope
8	Pride

Evaluation Metrics

The model was evaluated using precision, recall, f1-score, and accuracy.

Classification Report

Label	Precision	Recall	F1-score	Support
Anger (0)	0.49	0.61	0.54	777
Fear (1)	0.88	0.71	0.79	776
Disgust (2)	0.94	0.94	0.94	776
Sadness (3)	0.84	0.84	0.84	775
Joy (4)	0.82	0.81	0.81	736
None of them (5)	0.67	0.64	0.66	1594
Enthusiasm (6)	0.62	0.61	0.62	776
Hope (7)	0.51	0.53	0.52	777
Pride (8)	0.76	0.79	0.78	776

Overall Performance:

Accuracy: 71%
Macro Avg: Precision: 0.73, Recall: 0.72, F1-score: 0.72
Weighted Avg: Precision: 0.72, Recall: 0.71, F1-score: 0.71

How to Use

To use this model for text classification in Python:

from transformers import pipeline

classifier = pipeline("text-classification", model="poltextlab/xlm-roberta-large-pooled-czech-emotions9-v2", use_auth_token="<HF_TOKEN>")

text = "Jsem tak ráda, že jste tady!" 
result = classifier(text)
print(result)