This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-mpnet-base-v2 on the allstats-semantic-dataset-v4 dataset. 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: sentence-transformers/paraphrase-multilingual-mpnet-base-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- allstats-semantic-dataset-v4

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': 128, 'do_lower_case': False}) with Transformer model: XLMRobertaModel 
  (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("yahyaabd/allstats-semantic-mpnet")
# Run inference
sentences = [
    'Pernikahan usia anak di Indonesia periode 2013-2015',
    'Jumlah penduduk Indonesia 2013-2015',
    'Indeks Tendensi Bisnis dan Indeks Tendensi Konsumen 2013',
]
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

Datasets: allstats-semantic-mpnet-eval and allstats-semantic-mpnet-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	allstats-semantic-mpnet-eval	allstats-semantic-mpnet-test
pearson_cosine	0.9714	0.9723
spearman_cosine	0.8934	0.8932

Training Details

Training Dataset

allstats-semantic-dataset-v4

Dataset: allstats-semantic-dataset-v4 at 06c3cf8
Size: 88,250 training samples
Columns: query, doc, and label
Approximate statistics based on the first 1000 samples:
query doc label
type string string float
details
min: 4 tokens
mean: 11.38 tokens
max: 46 tokens
min: 4 tokens
mean: 14.48 tokens
max: 67 tokens
min: 0.0
mean: 0.51
max: 1.0

Samples:

query	doc	label
`Industri teh Indonesia tahun 2021`	`Statistik Transportasi Laut 2014`	`0.1`
`Tahun berapa data pertumbuhan ekonomi Indonesia tersebut?`	`Nilai Tukar Petani (NTP) November 2023 sebesar 116,73 atau naik 0,82 persen. Harga Gabah Kering Panen di Tingkat Petani turun 1,94 persen dan Harga Beras Premium di Penggilingan turun 0,91 persen.`	`0.0`
`Kemiskinan di Indonesia Maret`	`2018 Feb Tenaga Kerja`	`0.1`

Loss: CosineSimilarityLoss with these parameters:
json
1{ 2 "loss_fct": "torch.nn.modules.loss.MSELoss" 3}

Evaluation Dataset

allstats-semantic-dataset-v4

Dataset: allstats-semantic-dataset-v4 at 06c3cf8
Size: 18,910 evaluation samples
Columns: query, doc, and label
Approximate statistics based on the first 1000 samples:
query doc label
type string string float
details
min: 5 tokens
mean: 11.35 tokens
max: 33 tokens
min: 4 tokens
mean: 14.25 tokens
max: 52 tokens
min: 0.0
mean: 0.49
max: 1.0

Samples:

query	doc	label
`nAalisis keuangam deas tshun 019`	`Statistik Migrasi Nusa Tenggara Barat Hasil Survei Penduduk Antar Sensus 2015`	`0.1`
`Data tanaman buah dan sayur Indonesia tahun 2016`	`Statistik Penduduk Lanjut Usia 2010`	`0.1`
`Pasar beras di Indonesia tahun 2018`	`Buletin Statistik Perdagangan Luar Negeri Ekspor Menurut Kelompok Komoditi dan Negara, April 2021`	`0.2`

Loss: CosineSimilarityLoss with these parameters:
json
1{ 2 "loss_fct": "torch.nn.modules.loss.MSELoss" 3}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
num_train_epochs: 8
warmup_ratio: 0.1
fp16: True
dataloader_num_workers: 4
load_best_model_at_end: True
label_smoothing_factor: 0.05
eval_on_start: 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
torch_empty_cache_steps: None
learning_rate: 5e-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: 8
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: 4
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: True
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: 0.05
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: None
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
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
eval_on_start: True
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	allstats-semantic-mpnet-eval_spearman_cosine	allstats-semantic-mpnet-test_spearman_cosine
0	0	-	0.0979	0.6119	-
0.0906	250	0.0646	0.0427	0.7249	-
0.1813	500	0.039	0.0324	0.7596	-
0.2719	750	0.032	0.0271	0.7860	-
0.3626	1000	0.0276	0.0255	0.7920	-
0.4532	1250	0.0264	0.0230	0.8072	-
0.5439	1500	0.0249	0.0222	0.8197	-
0.6345	1750	0.0226	0.0210	0.8200	-
0.7252	2000	0.0218	0.0209	0.8202	-
0.8158	2250	0.0208	0.0201	0.8346	-
0.9065	2500	0.0209	0.0211	0.8240	-
0.9971	2750	0.0211	0.0190	0.8170	-
1.0877	3000	0.0161	0.0182	0.8332	-
1.1784	3250	0.0158	0.0179	0.8393	-
1.2690	3500	0.0167	0.0189	0.8341	-
1.3597	3750	0.0152	0.0168	0.8371	-
1.4503	4000	0.0151	0.0165	0.8435	-
1.5410	4250	0.0143	0.0156	0.8365	-
1.6316	4500	0.0147	0.0157	0.8467	-
1.7223	4750	0.0138	0.0155	0.8501	-
1.8129	5000	0.0147	0.0154	0.8457	-
1.9036	5250	0.0137	0.0152	0.8498	-
1.9942	5500	0.0144	0.0143	0.8485	-
2.0848	5750	0.0108	0.0139	0.8439	-
2.1755	6000	0.01	0.0146	0.8563	-
2.2661	6250	0.011	0.0141	0.8558	-
2.3568	6500	0.0107	0.0144	0.8497	-
2.4474	6750	0.01	0.0138	0.8577	-
2.5381	7000	0.0097	0.0136	0.8585	-
2.6287	7250	0.0102	0.0135	0.8521	-
2.7194	7500	0.0106	0.0133	0.8537	-
2.8100	7750	0.0098	0.0133	0.8643	-
2.9007	8000	0.0105	0.0138	0.8543	-
2.9913	8250	0.009	0.0129	0.8555	-
3.0819	8500	0.0071	0.0121	0.8692	-
3.1726	8750	0.006	0.0120	0.8709	-
3.2632	9000	0.0078	0.0120	0.8660	-
3.3539	9250	0.0072	0.0122	0.8656	-
3.4445	9500	0.007	0.0123	0.8696	-
3.5352	9750	0.0075	0.0117	0.8707	-
3.6258	10000	0.0081	0.0115	0.8682	-
3.7165	10250	0.0083	0.0116	0.8617	-
3.8071	10500	0.0075	0.0116	0.8665	-
3.8978	10750	0.0077	0.0119	0.8733	-
3.9884	11000	0.008	0.0113	0.8678	-
4.0790	11250	0.0051	0.0110	0.8760	-
4.1697	11500	0.0052	0.0108	0.8729	-
4.2603	11750	0.0056	0.0108	0.8771	-
4.3510	12000	0.0052	0.0109	0.8793	-
4.4416	12250	0.0049	0.0109	0.8766	-
4.5323	12500	0.0055	0.0114	0.8742	-
4.6229	12750	0.0061	0.0108	0.8749	-
4.7136	13000	0.0058	0.0109	0.8833	-
4.8042	13250	0.0049	0.0108	0.8767	-
4.8949	13500	0.0046	0.0108	0.8839	-
4.9855	13750	0.0052	0.0104	0.8790	-
5.0761	14000	0.0041	0.0102	0.8826	-
5.1668	14250	0.004	0.0103	0.8775	-
5.2574	14500	0.0036	0.0102	0.8855	-
5.3481	14750	0.0037	0.0104	0.8841	-
5.4387	15000	0.0036	0.0101	0.8860	-
5.5294	15250	0.0043	0.0104	0.8852	-
5.6200	15500	0.004	0.0100	0.8856	-
5.7107	15750	0.0043	0.0101	0.8842	-
5.8013	16000	0.0043	0.0099	0.8835	-
5.8920	16250	0.0041	0.0099	0.8852	-
5.9826	16500	0.0036	0.0101	0.8866	-
6.0732	16750	0.0031	0.0100	0.8881	-
6.1639	17000	0.0031	0.0098	0.8880	-
6.2545	17250	0.0027	0.0098	0.8886	-
6.3452	17500	0.0032	0.0097	0.8868	-
6.4358	17750	0.0027	0.0097	0.8876	-
6.5265	18000	0.0031	0.0097	0.8893	-
6.6171	18250	0.0032	0.0096	0.8903	-
6.7078	18500	0.003	0.0096	0.8898	-
6.7984	18750	0.0029	0.0098	0.8907	-
6.8891	19000	0.003	0.0096	0.8896	-
6.9797	19250	0.0026	0.0096	0.8913	-
7.0703	19500	0.0024	0.0096	0.8921	-
7.1610	19750	0.0021	0.0097	0.8920	-
7.2516	20000	0.0023	0.0096	0.8910	-
7.3423	20250	0.002	0.0096	0.8920	-
7.4329	20500	0.0022	0.0096	0.8924	-
7.5236	20750	0.002	0.0097	0.8917	-
7.6142	21000	0.0024	0.0096	0.8923	-
7.7049	21250	0.0025	0.0095	0.8928	-
7.7955	21500	0.0022	0.0095	0.8931	-
7.8861	21750	0.0023	0.0095	0.8932	-
7.9768	22000	0.0022	0.0095	0.8934	-
8.0	22064	-	-	-	0.8932

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.3.1
Transformers: 4.48.0
PyTorch: 2.4.1+cu121
Accelerate: 0.34.2
Datasets: 3.2.0
Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers

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",
}

	query	doc	label
type	string	string	float
details	min: 4 tokens mean: 11.38 tokens max: 46 tokens	min: 4 tokens mean: 14.48 tokens max: 67 tokens	min: 0.0 mean: 0.51 max: 1.0