SentenceTransformer based on indobenchmark/indobert-base-p2
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.
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
, andkalimat2
- 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
, andkalimat2
- 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
: stepsper_device_train_batch_size
: 32per_device_eval_batch_size
: 32learning_rate
: 2e-05num_train_epochs
: 2warmup_ratio
: 0.1fp16
: True
All Hyperparameters
Click to expand
overwrite_output_dir
: Falsedo_predict
: Falseeval_strategy
: stepsprediction_loss_only
: Trueper_device_train_batch_size
: 32per_device_eval_batch_size
: 32per_gpu_train_batch_size
: Noneper_gpu_eval_batch_size
: Nonegradient_accumulation_steps
: 1eval_accumulation_steps
: Nonelearning_rate
: 2e-05weight_decay
: 0.0adam_beta1
: 0.9adam_beta2
: 0.999adam_epsilon
: 1e-08max_grad_norm
: 1.0num_train_epochs
: 2max_steps
: -1lr_scheduler_type
: linearlr_scheduler_kwargs
: {}warmup_ratio
: 0.1warmup_steps
: 0log_level
: passivelog_level_replica
: warninglog_on_each_node
: Truelogging_nan_inf_filter
: Truesave_safetensors
: Truesave_on_each_node
: Falsesave_only_model
: Falserestore_callback_states_from_checkpoint
: Falseno_cuda
: Falseuse_cpu
: Falseuse_mps_device
: Falseseed
: 42data_seed
: Nonejit_mode_eval
: Falseuse_ipex
: Falsebf16
: Falsefp16
: Truefp16_opt_level
: O1half_precision_backend
: autobf16_full_eval
: Falsefp16_full_eval
: Falsetf32
: Nonelocal_rank
: 0ddp_backend
: Nonetpu_num_cores
: Nonetpu_metrics_debug
: Falsedebug
: []dataloader_drop_last
: Falsedataloader_num_workers
: 0dataloader_prefetch_factor
: Nonepast_index
: -1disable_tqdm
: Falseremove_unused_columns
: Truelabel_names
: Noneload_best_model_at_end
: Falseignore_data_skip
: Falsefsdp
: []fsdp_min_num_params
: 0fsdp_config
: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap
: Noneaccelerator_config
: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed
: Nonelabel_smoothing_factor
: 0.0optim
: adamw_torchoptim_args
: Noneadafactor
: Falsegroup_by_length
: Falselength_column_name
: lengthddp_find_unused_parameters
: Noneddp_bucket_cap_mb
: Noneddp_broadcast_buffers
: Falsedataloader_pin_memory
: Truedataloader_persistent_workers
: Falseskip_memory_metrics
: Trueuse_legacy_prediction_loop
: Falsepush_to_hub
: Falseresume_from_checkpoint
: Nonehub_model_id
: Nonehub_strategy
: every_savehub_private_repo
: Falsehub_always_push
: Falsegradient_checkpointing
: Falsegradient_checkpointing_kwargs
: Noneinclude_inputs_for_metrics
: Falseeval_do_concat_batches
: Truefp16_backend
: autopush_to_hub_model_id
: Nonepush_to_hub_organization
: Nonemp_parameters
:auto_find_batch_size
: Falsefull_determinism
: Falsetorchdynamo
: Noneray_scope
: lastddp_timeout
: 1800torch_compile
: Falsetorch_compile_backend
: Nonetorch_compile_mode
: Nonedispatch_batches
: Nonesplit_batches
: Noneinclude_tokens_per_second
: Falseinclude_num_input_tokens_seen
: Falseneftune_noise_alpha
: Noneoptim_target_modules
: Nonebatch_eval_metrics
: Falsebatch_sampler
: batch_samplermulti_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
@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",
}
- Downloads last month
- 16
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for cassador/indobert-snli-v1
Base model
indobenchmark/indobert-base-p2Evaluation results
- Pearson Cosine on sts devself-reported0.231
- Spearman Cosine on sts devself-reported0.232
- Pearson Manhattan on sts devself-reported0.198
- Spearman Manhattan on sts devself-reported0.204
- Pearson Euclidean on sts devself-reported0.199
- Spearman Euclidean on sts devself-reported0.204
- Pearson Dot on sts devself-reported0.256
- Spearman Dot on sts devself-reported0.251
- Pearson Max on sts devself-reported0.256
- Spearman Max on sts devself-reported0.251