t5 base indonesian summarization cased

Providercahya
Categorysummarization
LicenseApache-2.0
Downloads10.9K
Stars0

Overview

The t5-base-indonesian-summarization-cased model is a specialized encoder-decoder transformer fine-tuned specifically for condensing Indonesian text. Unlike general-purpose LLMs, this model is optimized for extractive and abstractive summarization tasks where maintaining the nuance of cased Indonesian text is critical for accuracy. For developers, this means a smaller footprint and faster inference times compared to massive frontier models, making it ideal for deployment in local pipelines or edge environments. It integrates seamlessly into Hugging Face workflows, allowing you to automate news aggregation, document shortening, or meeting note synthesis without the latency or cost of a general API. It provides a stable, task-specific alternative for those needing reliable Indonesian NLP without the overhead of prompt engineering for general models.

Highlights

  • Optimized for high-accuracy Indonesian text summarization
  • Efficient T5 base architecture ensures low inference latency
  • Cased tokenization preserves proper nouns and context
  • Apache-2.0 license allows flexible commercial integration
  • Seamless deployment via standard Hugging Face transformers

Usage

Install
# Install Hugging Face transformers
pip install transformers torch
SDK Usage
# Load model with transformers
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("cahya/t5-base-indonesian-summarization-cased")
tokenizer = AutoTokenizer.from_pretrained("cahya/t5-base-indonesian-summarization-cased")

Hugging Face Download

We recommend downloading the model via the Hugging Face CLI or Hub SDK.

Guidance：Before downloading, install huggingface_hub with:

Guidance 
pip install -U huggingface_hub

CLI Download

Download the full repository

Download the full repository 
huggingface-cli download cahya/t5-base-indonesian-summarization-cased

Download a single file to a local folder (e.g. config.json into ./dir)

Download a single file to a local folder (e.g. config.json into ./dir) 
huggingface-cli download cahya/t5-base-indonesian-summarization-cased config.json --local-dir ./dir

See the official docs for more CLI options

SDK Download

SDK Download 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('cahya/t5-base-indonesian-summarization-cased')

Git Download

Make sure git-lfs is installed first

Git Download 
git lfs install
git clone https://huggingface.co/cahya/t5-base-indonesian-summarization-cased

To skip LFS large-file downloads, use:

Skip LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/cahya/t5-base-indonesian-summarization-cased

Model files are hosted on the Hugging Face Hub — download directly via HF CLI / SDK / Git, not through this site.

PyTorch / Transformers Usage

Install Transformers

Install Transformers 
pip install -U transformers torch

Load the model and run inference

Load the model and run inference 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('cahya/t5-base-indonesian-summarization-cased')
tokenizer = AutoTokenizer.from_pretrained('cahya/t5-base-indonesian-summarization-cased')

Full Documentation

来源: HuggingFace

---
language: id
tags:

  • pipeline:summarization

  • summarization

  • t5

datasets:
  • id_liputan6

---

Indonesian T5 Summarization Base Model

Finetuned T5 base summarization model for Indonesian.

Finetuning Corpus

t5-base-indonesian-summarization-cased model is based on t5-base-bahasa-summarization-cased by huseinzol05, finetuned using id_liputan6 dataset.

Load Finetuned Model

python
from transformers import T5Tokenizer, T5Model, T5ForConditionalGeneration

tokenizer = T5Tokenizer.from_pretrained("cahya/t5-base-indonesian-summarization-cased")
model = T5ForConditionalGeneration.from_pretrained("cahya/t5-base-indonesian-summarization-cased")

Code Sample

python
from transformers import T5Tokenizer, T5ForConditionalGeneration

tokenizer = T5Tokenizer.from_pretrained("cahya/t5-base-indonesian-summarization-cased")
model = T5ForConditionalGeneration.from_pretrained("cahya/t5-base-indonesian-summarization-cased")


# 
ARTICLE_TO_SUMMARIZE = ""


generate summary

input_ids = tokenizer.encode(ARTICLE_TO_SUMMARIZE, return_tensors='pt')
summary_ids = model.generate(input_ids,
            min_length=20,
            max_length=80,
            num_beams=10,
            repetition_penalty=2.5,
            length_penalty=1.0,
            early_stopping=True,
            no_repeat_ngram_size=2,
            use_cache=True,
            do_sample = True,
            temperature = 0.8,
            top_k = 50,
            top_p = 0.95)

summary_text = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
print(summary_text)

Output:

code
View official source
Join our Telegram