ragilbuaj/sentiment-analysis-TWS-reviews

Sentiment Analysis for TWS Reviews

This model is a fine-tuned version of w11wo/indonesian-roberta-base-sentiment-classifier on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.2671
• Accuracy: 0.9051
• Precision: 0.9004
• Recall: 0.9051
• F1: 0.9007

How to Use

from transformers import pipeline

# Buat pipeline klasifikasi menggunakan model dari Hugging Face Hub
model_name = "ragilbuaj/sentiment-analysis-TWS-reviews"
classifier = pipeline("sentiment-analysis", model=model_name, tokenizer=model_name)

# Gunakan pipeline untuk mengklasifikasikan teks
result = classifier("I love using Hugging Face models!")
print(result)

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.2786	0.9992	589	0.2370	0.9109	0.9053	0.9109	0.9060
0.2205	2.0	1179	0.2626	0.9143	0.9175	0.9143	0.9154
0.1252	2.9992	1768	0.3468	0.9169	0.9160	0.9169	0.9164
0.0869	4.0	2358	0.4000	0.9220	0.9200	0.9220	0.9208
0.0147	4.9958	2945	0.4424	0.9220	0.9180	0.9220	0.9194

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1