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 retrieval.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: indobenchmark/indobert-base-p2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Supported Modality: Text

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 768, 'pooling_mode': 'mean', '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("Folma/IndoBERT-Retail-Search-Model")
# Run inference
sentences = [
    'KATALOG PRODUK ASEPSO SABUN BADAN BATANG',
    'ASEPSO BAR SOAP FRESH 80 GR',
    'JOHNSONS BABY OIL 125 ML',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.8572, -0.1430],
#         [ 0.8572,  1.0000, -0.1383],
#         [-0.1430, -0.1383,  1.0000]])

Evaluation

Metrics

Information Retrieval

Dataset: retail-production-eval
Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.3367
cosine_accuracy@3	0.5963
cosine_accuracy@5	0.7027
cosine_accuracy@10	0.8207
cosine_precision@1	0.3367
cosine_precision@3	0.1988
cosine_precision@5	0.1405
cosine_precision@10	0.0821
cosine_recall@1	0.3367
cosine_recall@3	0.5963
cosine_recall@5	0.7027
cosine_recall@10	0.8207
cosine_ndcg@10	0.5704
cosine_mrr@10	0.4912
cosine_map@100	0.4988

Training Details

Training Dataset

Unnamed Dataset

Size: 54,541 training samples
Columns: anchor, positive, and negative
Approximate statistics based on the first 1000 samples:
anchor positive negative
type string string string
details
min: 4 tokens
mean: 8.5 tokens
max: 20 tokens
min: 5 tokens
mean: 11.18 tokens
max: 21 tokens
min: 4 tokens
mean: 11.17 tokens
max: 21 tokens

Samples:

anchor	positive	negative
`IKAN LAUT MERK SEAFOOD`	`IKAN FRESH KEMBUNG COMO KG`	`MABELL SOSIS KEJU 60GR`
`KATALOG PRODUK BU SANDRA SAUS SAMBAL`	`BU SANDRA SAMBAL UDANG 150g`	`CADBURY LICKABLE DAIRY MILK 20GR`
`PILIHAN BUMBU KOREA DARI O FOOD`	`O'FOOD KIMCHI JJIGAE SAUCE 120g`	`MAMASUKA RUMPUT LAUT RENDANG 4.5X2`

Loss: MultipleNegativesRankingLoss with these parameters:
json
1{ 2 "scale": 20.0, 3 "similarity_fct": "cos_sim", 4 "gather_across_devices": false, 5 "directions": [ 6 "query_to_doc" 7 ], 8 "partition_mode": "joint", 9 "hardness_mode": null, 10 "hardness_strength": 0.0 11}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 64
num_train_epochs: 50
learning_rate: 2e-05
warmup_steps: 0.1
weight_decay: 0.01
fp16: True
eval_strategy: steps
load_best_model_at_end: True

All Hyperparameters

Click to expand

per_device_train_batch_size: 64
num_train_epochs: 50
max_steps: -1
learning_rate: 2e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0.1
optim: adamw_torch_fused
optim_args: None
weight_decay: 0.01
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 1
average_tokens_across_devices: True
max_grad_norm: 1.0
label_smoothing_factor: 0.0
bf16: False
fp16: True
bf16_full_eval: False
fp16_full_eval: False
tf32: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
use_liger_kernel: False
liger_kernel_config: None
use_cache: False
neftune_noise_alpha: None
torch_empty_cache_steps: None
auto_find_batch_size: False
log_on_each_node: True
logging_nan_inf_filter: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: False
project: huggingface
trackio_space_id: trackio
eval_strategy: steps
per_device_eval_batch_size: 8
prediction_loss_only: True
eval_on_start: False
eval_do_concat_batches: True
eval_use_gather_object: False
eval_accumulation_steps: None
include_for_metrics: []
batch_eval_metrics: False
save_only_model: False
save_on_each_node: False
enable_jit_checkpoint: False
push_to_hub: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: True
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
use_cpu: False
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
remove_unused_columns: True
label_names: None
train_sampling_strategy: random
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
ddp_backend: None
ddp_timeout: 1800
fsdp: []
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	retail-production-eval_cosine_ndcg@10
0.5862	500	1.6535	0.4757
1.1723	1000	0.8146	0.5074
1.7585	1500	0.5929	0.5271
2.3447	2000	0.4561	0.5407
2.9308	2500	0.4064	0.5501
3.5170	3000	0.3404	0.5539
4.1032	3500	0.3302	0.5593
4.6893	4000	0.2986	0.5644
5.2755	4500	0.2868	0.5641
5.8617	5000	0.2767	0.5644
6.4478	5500	0.2532	0.571
7.0340	6000	0.2495	0.5684
7.6202	6500	0.2346	0.5676
8.2063	7000	0.2286	0.5704

The bold row denotes the saved checkpoint.

Training Time

Training: 19.3 minutes

Framework Versions

Python: 3.12.13
Sentence Transformers: 5.4.0
Transformers: 5.5.4
PyTorch: 2.10.0+cu128
Accelerate: 1.13.0
Datasets: 4.0.0
Tokenizers: 0.22.2

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",
}

MultipleNegativesRankingLoss

bibtex

@misc{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}

	anchor	positive	negative
type	string	string	string
details	min: 4 tokens mean: 8.5 tokens max: 20 tokens	min: 5 tokens mean: 11.18 tokens max: 21 tokens	min: 4 tokens mean: 11.17 tokens max: 21 tokens