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-v2")
# Run inference
sentences = [
    'Direktori perusahaan pengelola hutan 2015',
    'Direktori Perusahaan Kehutanan 2015',
    'Indeks Pembangunan Manusia (IPM) Indonesia tahun 2024 mencapai 75,02, meningkat 0,63 poin atau 0,85 persen dibandingkan tahun sebelumnya yang sebesar 74,39.',
]
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-v2-eval and allstats-semantic-mpnet-v2-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	allstats-semantic-mpnet-v2-eval	allstats-semantic-mpnet-v2-test
pearson_cosine	0.9438	0.9434
spearman_cosine	0.7866	0.787

Training Details

Training Dataset

allstats-semantic-dataset-v4

Dataset: allstats-semantic-dataset-v4 at 0e15dee
Size: 73,392 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: 11.28 tokens
max: 34 tokens
min: 5 tokens
mean: 14.71 tokens
max: 58 tokens
min: 0.0
mean: 0.48
max: 1.0

Samples:

query	doc	label
`Data bisnis Kalbar sensus 2016`	`Indikator Ekonomi Oktober 2012`	`0.1`
`Informasi tentang pola pengeluaran masyarakat Bengkulu berdasarkan kelompok pendapatan?`	`Rata-rata Konsumsi dan Pengeluaran Perkapita Seminggu Menurut Komoditi Makanan dan Golongan Pengeluaran per Kapita Seminggu di Provinsi Bengkulu, 2018-2023`	`0.88`
`Laopran keuagnan lmebaga non proft 20112-013`	`Neraca Lembaga Non Profit yang Melayani Rumah Tangga 2011-2013`	`0.93`

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 0e15dee
Size: 15,726 evaluation 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: 11.52 tokens
max: 37 tokens
min: 5 tokens
mean: 14.38 tokens
max: 61 tokens
min: 0.0
mean: 0.49
max: 1.0

Samples:

query	doc	label
`Data transportasi bulan Februari 2021`	`Tenaga Kerja Februari 2023`	`0.08`
`Sebear berspa prrsen eknaikan Inseks Hraga Predagangan eBsar (IHB) Umym Nasiona di aMret 202?`	`Maret 2020, Indeks Harga Perdagangan Besar (IHPB) Umum Nasional naik 0,10 persen`	`1.0`
`Data ekspor dan moda transportasi tahun 2018-2019`	`Indikator Pasar Tenaga Kerja Indonesia Agustus 2012`	`0.08`

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: 0.5
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: 0.5
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-v2-eval_spearman_cosine	allstats-semantic-mpnet-v2-test_spearman_cosine
0	0	-	0.1031	0.6244	-
0.1090	250	0.0442	0.0321	0.7393	-
0.2180	500	0.0295	0.0248	0.7641	-
0.3269	750	0.0259	0.0224	0.7733	-
0.4359	1000	0.0208	0.0199	0.7866	-
0.5	1147	-	-	-	0.7870

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: 5 tokens mean: 11.28 tokens max: 34 tokens	min: 5 tokens mean: 14.71 tokens max: 58 tokens	min: 0.0 mean: 0.48 max: 1.0