YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

SATA

Sentiment Analyzer Trained by Araf

SATA is a PyTorch-based transformer model fine-tuned for emotion (sentiment) classification on English text. It is built on top of the roberta-base architecture and trained using the Hugging Face emotion dataset.

The goal of this model is to classify short pieces of text into one of several core emotional categories. SATA is designed to be simple to use, reproducible, and easily deployable via the Hugging Face ecosystem.


Model Overview

  • Model name: SATA
  • Meaning: Sentiment Analyzer Trained by Araf
  • Base model: roberta-base
  • Framework: PyTorch
  • Task: Emotion / sentiment classification
  • Language: English

Dataset

The model was fine-tuned on the Emotion dataset from Hugging Face Datasets.

  • Dataset size: ~20k samples
  • Text domain: Short informal English texts
  • Splits: Train / Validation / Test

Emotion Labels

The model predicts one of the following six classes:

Label ID Emotion
0 Sadness
1 Joy
2 Love
3 Anger
4 Fear
5 Surprise

Training Details

  • Optimizer: AdamW
  • Learning rate: 2e-5
  • Batch size: 16
  • Epochs: 3
  • Max sequence length: 128
  • Loss function: Cross-entropy loss

The model was trained using a custom PyTorch training loop (no Hugging Face Trainer abstraction).


Validation Performance

  • Validation Accuracy: 93%

Accuracy was measured on the official validation split of the Emotion dataset.


Usage Example

from transformers import pipeline

classifier = pipeline(
    "text-classification",
    model="captainaraf/sata"
)

classifier("I am extremely happy today")
Downloads last month
41
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support