UAE_Large_V1_nav1 / README.md
Naveen20o1's picture
Add new SentenceTransformer model.
7593fdb verified
metadata
language: []
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:899
  - loss:CoSENTLoss
base_model: WhereIsAI/UAE-Large-V1
datasets: []
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
widget:
  - source_sentence: hr
    sentences:
      - Geographical
      - Quantity
      - Person
  - source_sentence: product
    sentences:
      - Organization
      - Time
      - Artifact
  - source_sentence: council
    sentences:
      - Person
      - Person
      - Quantity
  - source_sentence: salesman
    sentences:
      - Person
      - Time
      - Person
  - source_sentence: joint_venture_name
    sentences:
      - Person
      - Organization
      - Person
pipeline_tag: sentence-similarity
model-index:
  - name: SentenceTransformer based on WhereIsAI/UAE-Large-V1
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts dev
          type: sts-dev
        metrics:
          - type: pearson_cosine
            value: 0.8883347646952768
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8463283813349622
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.8611263810572393
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.838590521848471
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.8622761936152195
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8405249867200939
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.8773449747713008
            name: Pearson Dot
          - type: spearman_dot
            value: 0.8443939164633394
            name: Spearman Dot
          - type: pearson_max
            value: 0.8883347646952768
            name: Pearson Max
          - type: spearman_max
            value: 0.8463283813349622
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts dev test
          type: sts-dev_test
        metrics:
          - type: pearson_cosine
            value: 0.9278166656810813
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8783100656536799
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.954242190347034
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.8783100656536799
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.9519570678729806
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.8783100656536799
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.9258180799496141
            name: Pearson Dot
          - type: spearman_dot
            value: 0.8783100656536799
            name: Spearman Dot
          - type: pearson_max
            value: 0.954242190347034
            name: Pearson Max
          - type: spearman_max
            value: 0.8783100656536799
            name: Spearman Max

SentenceTransformer based on WhereIsAI/UAE-Large-V1

This is a sentence-transformers model finetuned from WhereIsAI/UAE-Large-V1. It maps sentences & paragraphs to a 1024-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: WhereIsAI/UAE-Large-V1
  • Maximum Sequence Length: 512 tokens
  • Output Dimensionality: 1024 tokens
  • Similarity Function: Cosine Similarity

Model Sources

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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("Naveen20o1/UAE_Large_V1_nav1")
# Run inference
sentences = [
    'joint_venture_name',
    'Organization',
    'Person',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# 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 0.8883
spearman_cosine 0.8463
pearson_manhattan 0.8611
spearman_manhattan 0.8386
pearson_euclidean 0.8623
spearman_euclidean 0.8405
pearson_dot 0.8773
spearman_dot 0.8444
pearson_max 0.8883
spearman_max 0.8463

Semantic Similarity

Metric Value
pearson_cosine 0.9278
spearman_cosine 0.8783
pearson_manhattan 0.9542
spearman_manhattan 0.8783
pearson_euclidean 0.952
spearman_euclidean 0.8783
pearson_dot 0.9258
spearman_dot 0.8783
pearson_max 0.9542
spearman_max 0.8783

Training Details

Training Dataset

Unnamed Dataset

  • Size: 899 training samples
  • Columns: sentence1, sentence2, and score
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2 score
    type string string float
    details
    • min: 3 tokens
    • mean: 4.33 tokens
    • max: 10 tokens
    • min: 3 tokens
    • mean: 3.0 tokens
    • max: 3 tokens
    • min: 0.0
    • mean: 0.48
    • max: 1.0
  • Samples:
    sentence1 sentence2 score
    postcode Communication 0.0
    telephone_number Communication 1.0
    vehicle_type Person 0.0
  • Loss: CoSENTLoss with these parameters:
    {
        "scale": 20.0,
        "similarity_fct": "pairwise_cos_sim"
    }
    

Evaluation Dataset

Unnamed Dataset

  • Size: 60 evaluation samples
  • Columns: sentence1, sentence2, and score
  • Approximate statistics based on the first 1000 samples:
    sentence1 sentence2 score
    type string string float
    details
    • min: 3 tokens
    • mean: 4.15 tokens
    • max: 7 tokens
    • min: 3 tokens
    • mean: 3.0 tokens
    • max: 3 tokens
    • min: 0.0
    • mean: 0.55
    • max: 1.0
  • Samples:
    sentence1 sentence2 score
    surgical_history Person 0.0
    count Quantity 1.0
    board Person 0.0
  • Loss: CoSENTLoss with these parameters:
    {
        "scale": 20.0,
        "similarity_fct": "pairwise_cos_sim"
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 16
  • per_device_eval_batch_size: 16
  • learning_rate: 2e-05
  • num_train_epochs: 11
  • warmup_ratio: 0.1
  • fp16: 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: 16
  • per_device_eval_batch_size: 16
  • per_gpu_train_batch_size: None
  • per_gpu_eval_batch_size: None
  • gradient_accumulation_steps: 1
  • eval_accumulation_steps: None
  • learning_rate: 2e-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: 11
  • 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: 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
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss loss sts-dev_spearman_cosine sts-dev_test_spearman_cosine
0.8772 50 2.6697 - - -
1.7544 100 0.5212 2.4196 0.8057 -
2.6316 150 0.3741 - - -
3.5088 200 0.0033 1.7749 0.8115 -
4.3860 250 0.0257 - - -
5.2632 300 0.0159 2.2808 0.8154 -
6.1404 350 0.0057 - - -
7.0175 400 0.0044 1.5027 0.8444 -
7.8947 450 0.0004 - - -
8.7719 500 0.0008 0.9416 0.8483 -
9.6491 550 0.0001 - - -
10.5263 600 0.0002 1.1264 0.8463 -
11.0 627 - - - 0.8783

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

@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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}