This is a sentence-transformers model finetuned from indobenchmark/indobert-base-p2. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: indobenchmark/indobert-base-p2
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("cassador/indobert-snli-v1")
# Run inference
sentences = [
    'Wanita profesional di meja pendaftaran acara sementara pria berjas melihat.',
    'Orang-orang berkumpul untuk sebuah acara.',
    'Ada seorang anak yang tersenyum untuk difoto.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

Dataset: sts-dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.2315
spearman_cosine	0.2318
pearson_manhattan	0.1985
spearman_manhattan	0.2038
pearson_euclidean	0.1987
spearman_euclidean	0.2039
pearson_dot	0.2562
spearman_dot	0.251
pearson_max	0.2562
spearman_max	0.251

Semantic Similarity

Dataset: sts-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.5915
spearman_cosine	0.5979
pearson_manhattan	0.5132
spearman_manhattan	0.5147
pearson_euclidean	0.5943
spearman_euclidean	0.6002
pearson_dot	0.588
spearman_dot	0.5934
pearson_max	0.5943
spearman_max	0.6002

Training Details

Training Dataset

Unnamed Dataset

Size: 133,472 training samples
Columns: label, kalimat1, and kalimat2
Approximate statistics based on the first 1000 samples:
label kalimat1 kalimat2
type int string string
details
0: ~50.00%
1: ~50.00%
min: 5 tokens
mean: 16.47 tokens
max: 48 tokens
min: 4 tokens
mean: 9.62 tokens
max: 22 tokens

Samples:

label	kalimat1	kalimat2
`0`	`Seseorang di atas kuda melompati pesawat yang rusak.`	`Seseorang sedang makan malam, memesan telur dadar.`
`1`	`Seseorang di atas kuda melompati pesawat yang rusak.`	`Seseorang berada di luar ruangan, di atas kuda.`
`1`	`Anak-anak tersenyum dan melambai ke kamera`	`Ada anak-anak yang hadir`

Loss: SoftmaxLoss

Evaluation Dataset

Unnamed Dataset

Size: 6,607 evaluation samples
Columns: label, kalimat1, and kalimat2
Approximate statistics based on the first 1000 samples:
label kalimat1 kalimat2
type int string string
details
0: ~50.10%
1: ~49.90%
min: 5 tokens
mean: 16.87 tokens
max: 49 tokens
min: 3 tokens
mean: 9.45 tokens
max: 27 tokens

Samples:

label	kalimat1	kalimat2
`1`	`Dua wanita berpelukan sambil memegang paket untuk pergi.`	`Dua wanita memegang paket.`
`0`	`Dua wanita berpelukan sambil memegang paket untuk pergi.`	`Orang-orang berkelahi di luar toko makanan.`
`1`	`Dua anak kecil berbaju biru, satu dengan nomor 9 dan satu dengan nomor 2 berdiri di tangga kayu di kamar mandi dan mencuci tangan di wastafel.`	`Dua anak dengan kaus bernomor mencuci tangan mereka.`

Loss: SoftmaxLoss

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
learning_rate: 2e-05
num_train_epochs: 2
warmup_ratio: 0.1
fp16: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
learning_rate: 2e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 2
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: False
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
dispatch_batches: None
split_batches: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	sts-dev_spearman_cosine	sts-test_spearman_cosine
0	0	0.2318	-
2.0	8342	-	0.5979

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.0+cu121
Accelerate: 0.31.0
Datasets: 2.20.0
Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers and SoftmaxLoss

bibtex

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

	label	kalimat1	kalimat2
type	int	string	string
details	0: ~50.00% 1: ~50.00%	min: 5 tokens mean: 16.47 tokens max: 48 tokens	min: 4 tokens mean: 9.62 tokens max: 22 tokens