negasibert-ct / README.md
damand2061's picture
Add new SentenceTransformer model.
9c34982 verified
metadata
base_model: indobenchmark/indobert-base-p1
datasets: []
language: []
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:12800
  - loss:ContrastiveTensionLoss
widget:
  - source_sentence: >-
      Makalah ini diterbitkan dalam format online hanya oleh Metro
      International.
    sentences:
      - >-
        Liga ini berkembang dari tahun 1200 hingga 1500, dan terus menjadi
        semakin penting setelahnya.
      - >-
        Ini dirancang oleh orang lain selain WL Bottomley / William Lawrence
        Bottomley.
      - >-
        Lahan tersebut sekarang menjadi Cagar Alam Bentley Priory, sebuah Situs
        Kepentingan Ilmiah Khusus.
  - source_sentence: >-
      Pengadilan menentang keputusan tahun 2010 dan kasus ini dilanjutkan sesuai
      dengan manfaatnya.
    sentences:
      - Gunung itu berada di Front Allegheny.
      - >-
        Stasiun St Albans Abbey adalah stasiun dalam perjalanan jalur ganda dari
        stasiun Watford Junction.
      - >-
        Pada tahun 2011, keluarga Penner tidak lagi menyebut rumah Habitatnya,
        rumah.
  - source_sentence: Aku tidak jahat dalam hal ini.
    sentences:
      - >-
        Awalnya disetujui untuk onchocerciasis dan strongyloidiasis, Ivermectin
        sekarang disetujui oleh FDA untuk pedikulosis.
      - Lagu ini mencapai ARIA Singles Chart Top 100.
      - >-
        Bebaskan diri Anda dari permusuhan dan kemarahan untuk menunjukkan rasa
        hormat terhadap tubuh dan kehidupan Anda.
  - source_sentence: Waktu pengiriman sangat cepat.
    sentences:
      - Dia kemudian bermain untuk South West Ham.
      - >-
        Qatar, bagaimanapun, tidak diminta untuk mengibarkan bendera Trucial
        yang ditentukan.
      - Sepasang pintu ini juga meredam suara dari luar.
  - source_sentence: >-
      Dengan demikian, seorang model penutur harus mengolah representasi warna
      dalam konteks dan menghasilkan ujaran yang dapat membedakan warna sasaran
      dengan ujaran lainnya.
    sentences:
      - Dia bukan bagian dari American Institute of Architects.
      - >-
        Pada tahun 1975 VTL dibeli oleh Greyhound Lines, menjadi anak
        perusahaan.
      - Pada tanggal 24 April 2009, Forum Terbuka IBIS menyetujui versi 2.0.
model-index:
  - name: SentenceTransformer based on indobenchmark/indobert-base-p1
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: str dev
          type: str-dev
        metrics:
          - type: pearson_cosine
            value: 0.47668991144701395
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.48495339068233534
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.5041035764250676
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.49270037559673846
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.5059182139447496
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.4915516775931335
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.2991963739133043
            name: Pearson Dot
          - type: spearman_dot
            value: 0.2630042391245101
            name: Spearman Dot
          - type: pearson_max
            value: 0.5059182139447496
            name: Pearson Max
          - type: spearman_max
            value: 0.49270037559673846
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: str test
          type: str-test
        metrics:
          - type: pearson_cosine
            value: 0.47374249981827143
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.5083479438750005
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.49828227586252527
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.4962152495999787
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.5006486050380166
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.49701891829828837
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.2573207350736585
            name: Pearson Dot
          - type: spearman_dot
            value: 0.24350607759185028
            name: Spearman Dot
          - type: pearson_max
            value: 0.5006486050380166
            name: Pearson Max
          - type: spearman_max
            value: 0.5083479438750005
            name: Spearman Max

SentenceTransformer based on indobenchmark/indobert-base-p1

This is a sentence-transformers model finetuned from indobenchmark/indobert-base-p1. 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-p1
  • Maximum Sequence Length: 32 tokens
  • Output Dimensionality: 768 tokens
  • Similarity Function: Cosine Similarity

Model Sources

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 32, '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("damand2061/negasibert-ct")
# Run inference
sentences = [
    'Dengan demikian, seorang model penutur harus mengolah representasi warna dalam konteks dan menghasilkan ujaran yang dapat membedakan warna sasaran dengan ujaran lainnya.',
    'Pada tahun 1975 VTL dibeli oleh Greyhound Lines, menjadi anak perusahaan.',
    'Pada tanggal 24 April 2009, Forum Terbuka IBIS menyetujui versi 2.0.',
]
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

Metric Value
pearson_cosine 0.4767
spearman_cosine 0.485
pearson_manhattan 0.5041
spearman_manhattan 0.4927
pearson_euclidean 0.5059
spearman_euclidean 0.4916
pearson_dot 0.2992
spearman_dot 0.263
pearson_max 0.5059
spearman_max 0.4927

Semantic Similarity

Metric Value
pearson_cosine 0.4737
spearman_cosine 0.5083
pearson_manhattan 0.4983
spearman_manhattan 0.4962
pearson_euclidean 0.5006
spearman_euclidean 0.497
pearson_dot 0.2573
spearman_dot 0.2435
pearson_max 0.5006
spearman_max 0.5083

Training Details

Training Dataset

Unnamed Dataset

  • Size: 12,800 training samples
  • Columns: sentence_0, sentence_1, and label
  • Approximate statistics based on the first 1000 samples:
    sentence_0 sentence_1 label
    type string string int
    details
    • min: 5 tokens
    • mean: 14.81 tokens
    • max: 32 tokens
    • min: 5 tokens
    • mean: 14.92 tokens
    • max: 32 tokens
    • 0: ~87.50%
    • 1: ~12.50%
  • Samples:
    sentence_0 sentence_1 label
    Warnanya tercermin pada corak dan lambang universitas kota tersebut. Warnanya tercermin pada corak dan lambang universitas kota tersebut. 1
    Pada awal tahun 2008, Ikerbasque menolak menugaskan Enrique Zuazua. Oh, ayolah, itu adil. 0
    Pada tahun 2006, sebuah studi diselesaikan tentang prospek jalur Scarborough. Jurnal Pendidikan Modern didirikan olehnya. 0
  • Loss: ContrastiveTensionLoss

Training Hyperparameters

Non-Default Hyperparameters

  • per_device_train_batch_size: 64
  • per_device_eval_batch_size: 64
  • num_train_epochs: 5
  • multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: no
  • prediction_loss_only: True
  • per_device_train_batch_size: 64
  • per_device_eval_batch_size: 64
  • 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
  • num_train_epochs: 5
  • max_steps: -1
  • lr_scheduler_type: linear
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.0
  • 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: False
  • 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: 0
  • dataloader_prefetch_factor: None
  • past_index: -1
  • disable_tqdm: False
  • remove_unused_columns: True
  • label_names: None
  • load_best_model_at_end: False
  • 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: 0.0
  • 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: False
  • hub_always_push: False
  • gradient_checkpointing: False
  • gradient_checkpointing_kwargs: None
  • include_inputs_for_metrics: False
  • 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: False
  • eval_use_gather_object: False
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: round_robin

Training Logs

Epoch Step Training Loss str-dev_spearman_max str-test_spearman_max
1.0 200 - 0.5009 0.5084
2.0 400 - 0.4926 0.5025
2.5 500 2328.8573 - -
3.0 600 - 0.4909 0.5058
4.0 800 - 0.4909 0.5064
5.0 1000 0.5625 0.4927 0.5083

Framework Versions

  • Python: 3.10.14
  • Sentence Transformers: 3.0.1
  • Transformers: 4.44.0
  • PyTorch: 2.4.0
  • Accelerate: 0.33.0
  • Datasets: 2.21.0
  • Tokenizers: 0.19.1

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",
}

ContrastiveTensionLoss

@inproceedings{carlsson2021semantic,
    title={Semantic Re-tuning with Contrastive Tension},
    author={Fredrik Carlsson and Amaru Cuba Gyllensten and Evangelia Gogoulou and Erik Ylip{"a}{"a} Hellqvist and Magnus Sahlgren},
    booktitle={International Conference on Learning Representations},
    year={2021},
    url={https://openreview.net/forum?id=Ov_sMNau-PF}
}