metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:79621
  - loss:CosineSimilarityLoss
base_model: sentence-transformers/paraphrase-multilingual-mpnet-base-v2
widget:
  - source_sentence: Data demografi Indonesia 2021 perempuan dan lakilaki
    sentences:
      - >-
        Buletin Statistik Perdagangan Luar Negeri Ekspor Menurut Komoditi HS,
        Februari 2015
      - Statistik Potensi Desa Provinsi Jawa Barat 2014
      - Pengeluaran untuk Konsumsi Penduduk Indonesia, September 2017
  - source_sentence: Data analisis tematik kependudukan Indonesia migrasi dan ketenagakerjaan
    sentences:
      - >-
        Direktori Perusahaan Industri Penggilingan Padi Tahun 2012 Provinsi
        Bengkulu
      - Buletin Statistik Perdagangan Luar Negeri Ekspor Menurut HS, Juni 2023
      - Luas Panen dan Produksi Padi 2022
  - source_sentence: Daftar perusahaan penggilingan padi Kalimantan
    sentences:
      - Ringkasan Neraca Arus Dana, Triwulan II, 2011*), (Miliar Rupiah)
      - Klasifikasi Baku Komoditas Indonesia 2012 Buku 1
      - >-
        Statistik Penduduk Lanjut Usia Provinsi Nusa Tenggara Barat 2010-Hasil
        Sensus Penduduk 2010
  - source_sentence: Perdagangan luar negeri impor Januari 2010
    sentences:
      - Buletin Statistik Perdagangan Luar Negeri Impor Januari 2010
      - Statistik Tanaman Sayuran dan Buah-buahan Semusim Indonesia 2012
      - Klasifikasi Baku Komoditas Indonesia (KBKI) 2012 Buku 4
  - source_sentence: Biaya hidup kelompok perumahan Indonesia 2017
    sentences:
      - Indeks Harga Perdagangan Besar 2007
      - Statistik Upah 2013
      - >-
        Survei Biaya Hidup (SBH) 2018 Bulukumba, Watampone, Makassar, Pare-Pare,
        dan Palopo
datasets:
  - yahyaabd/allstats-search-pairs-dataset
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: >-
      SentenceTransformer based on
      sentence-transformers/paraphrase-multilingual-mpnet-base-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: allstats semantic mpnet eval
          type: allstats-semantic-mpnet-eval
        metrics:
          - type: pearson_cosine
            value: 0.9830659002996571
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8410118780087822
            name: Spearman Cosine
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: allstats semantic mpnet test
          type: allstats-semantic-mpnet-test
        metrics:
          - type: pearson_cosine
            value: 0.9831682336847792
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8378561244643341
            name: Spearman Cosine

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-mpnet-base-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-mpnet-base-v2 on the allstats-search-pairs-dataset 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-search-pairs-dataset

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.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("yahyaabd/test-model")
# Run inference
sentences = [
    'Biaya hidup kelompok perumahan Indonesia 2017',
    'Statistik Upah 2013',
    'Survei Biaya Hidup (SBH) 2018 Bulukumba, Watampone, Makassar, Pare-Pare, dan Palopo',
]
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	0.9831	0.9832
spearman_cosine	0.841	0.8379

Training Details

Training Dataset

allstats-search-pairs-dataset

Dataset: allstats-search-pairs-dataset at 6712cb1
Size: 79,621 training 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: 10.78 tokens
max: 39 tokens

min: 5 tokens
mean: 13.73 tokens
max: 58 tokens

min: 0.0
mean: 0.44
max: 0.99

	query	doc	label
type	string	string	float
details	min: 5 tokens mean: 10.78 tokens max: 39 tokens	min: 5 tokens mean: 13.73 tokens max: 58 tokens	min: 0.0 mean: 0.44 max: 0.99

Samples:

query	doc	label
`Produksi jagung di Indonesia tahun 2009`	`Indeks Unit Value Ekspor Menurut Kode SITC Bulan Februari 2024`	`0.1`
`Data produksi industri manufaktur 2021`	`Perkembangan Indeks Produksi Industri Manufaktur 2021`	`0.96`
`direktori perusahaan industri penggilingan padi tahun 2012 provinsi sulawesi utara dan gorontalo`	`Neraca Pemerintahan Umum Indonesia 2007-2012`	`0.03`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Evaluation Dataset

allstats-search-pairs-dataset

Dataset: allstats-search-pairs-dataset at 6712cb1
Size: 9,952 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: 10.75 tokens
max: 40 tokens

min: 4 tokens
mean: 14.09 tokens
max: 49 tokens

min: 0.01
mean: 0.48
max: 0.99

	query	doc	label
type	string	string	float
details	min: 5 tokens mean: 10.75 tokens max: 40 tokens	min: 4 tokens mean: 14.09 tokens max: 49 tokens	min: 0.01 mean: 0.48 max: 0.99

Samples:

query	doc	label
`Daftar perusahaan industri pengolahan skala kecil 2006`	`Statistik Migrasi Nusa Tenggara Barat Hasil SP 2010`	`0.05`
`Populasi Indonesia per provinsi 2000-2010`	`Indikator Ekonomi Desember 2023`	`0.08`
`Data harga barang desa non-pangan tahun 2022`	`Statistik Kunjungan Tamu Asing 2004`	`0.1`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
num_train_epochs: 6
warmup_ratio: 0.1
fp16: True
dataloader_num_workers: 4
load_best_model_at_end: True
label_smoothing_factor: 0.01
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: 6
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: 0.01
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	-	0.0958	0.6404	-
0.1004	250	0.0482	0.0269	0.7685	-
0.2009	500	0.0249	0.0200	0.7737	-
0.3013	750	0.0196	0.0172	0.7768	-
0.4018	1000	0.0184	0.0172	0.7744	-
0.5022	1250	0.0185	0.0159	0.7751	-
0.6027	1500	0.0155	0.0156	0.7825	-
0.7031	1750	0.0165	0.0149	0.7826	-
0.8035	2000	0.0146	0.0136	0.7791	-
0.9040	2250	0.0132	0.0127	0.7825	-
1.0044	2500	0.0127	0.0129	0.7818	-
1.1049	2750	0.0104	0.0114	0.7849	-
1.2053	3000	0.009	0.0108	0.7870	-
1.3057	3250	0.0091	0.0112	0.7877	-
1.4062	3500	0.009	0.0104	0.7888	-
1.5066	3750	0.009	0.0101	0.7937	-
1.6071	4000	0.0084	0.0099	0.7924	-
1.7075	4250	0.008	0.0097	0.7942	-
1.8080	4500	0.0079	0.0094	0.7946	-
1.9084	4750	0.0078	0.0092	0.7928	-
2.0088	5000	0.0081	0.0088	0.7986	-
2.1093	5250	0.0056	0.0079	0.8025	-
2.2097	5500	0.0052	0.0080	0.8019	-
2.3102	5750	0.0048	0.0079	0.8073	-
2.4106	6000	0.0053	0.0081	0.8058	-
2.5110	6250	0.0049	0.0079	0.8091	-
2.6115	6500	0.0053	0.0077	0.8081	-
2.7119	6750	0.0052	0.0075	0.8075	-
2.8124	7000	0.0049	0.0077	0.8089	-
2.9128	7250	0.0051	0.0076	0.8066	-
3.0133	7500	0.0048	0.0074	0.8127	-
3.1137	7750	0.0034	0.0069	0.8162	-
3.2141	8000	0.0033	0.0070	0.8164	-
3.3146	8250	0.0036	0.0068	0.8194	-
3.4150	8500	0.0032	0.0069	0.8156	-
3.5155	8750	0.0032	0.0068	0.8196	-
3.6159	9000	0.0032	0.0067	0.8197	-
3.7164	9250	0.0034	0.0068	0.8194	-
3.8168	9500	0.0034	0.0066	0.8194	-
3.9172	9750	0.0032	0.0063	0.8239	-
4.0177	10000	0.0032	0.0065	0.8229	-
4.1181	10250	0.0023	0.0063	0.8258	-
4.2186	10500	0.0022	0.0062	0.8293	-
4.3190	10750	0.002	0.0063	0.8283	-
4.4194	11000	0.0023	0.0060	0.8306	-
4.5199	11250	0.0021	0.0061	0.8318	-
4.6203	11500	0.0021	0.0060	0.8290	-
4.7208	11750	0.0022	0.0058	0.8329	-
4.8212	12000	0.0021	0.0058	0.8342	-
4.9217	12250	0.0021	0.0058	0.8360	-
5.0221	12500	0.0017	0.0057	0.8355	-
5.1225	12750	0.0014	0.0057	0.8377	-
5.2230	13000	0.0015	0.0058	0.8363	-
5.3234	13250	0.0015	0.0058	0.8369	-
5.4239	13500	0.0014	0.0057	0.8386	-
5.5243	13750	0.0014	0.0057	0.8387	-
5.6247	14000	0.0015	0.0057	0.8392	-
5.7252	14250	0.0014	0.0056	0.8409	-
5.8256	14500	0.0014	0.0056	0.8410	-
5.9261	14750	0.0014	0.0056	0.8410	-
6.0	14934	-	-	-	0.8379

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.3.1
Transformers: 4.48.0
PyTorch: 2.5.1+cu121
Accelerate: 1.2.1
Datasets: 3.2.0
Tokenizers: 0.21.0

Citation

BibTeX

Sentence Transformers

@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",
}