README.md

metadata

tags:
  - sentence-transformers
  - cross-encoder
  - reranker
  - generated_from_trainer
  - dataset_size:7419
  - loss:BinaryCrossEntropyLoss
base_model: cross-encoder/nli-deberta-v3-base
pipeline_tag: text-ranking
library_name: sentence-transformers
metrics:
  - accuracy
  - accuracy_threshold
  - f1
  - f1_threshold
  - precision
  - recall
  - average_precision
model-index:
  - name: CrossEncoder based on cross-encoder/nli-deberta-v3-base
    results:
      - task:
          type: cross-encoder-classification
          name: Cross Encoder Classification
        dataset:
          name: ce val
          type: ce-val
        metrics:
          - type: accuracy
            value: 0.6036363636363636
            name: Accuracy
          - type: accuracy_threshold
            value: 0.5116937160491943
            name: Accuracy Threshold
          - type: f1
            value: 0.6751269035532994
            name: F1
          - type: f1_threshold
            value: 0.4685322642326355
            name: F1 Threshold
          - type: precision
            value: 0.5188556566970091
            name: Precision
          - type: recall
            value: 0.9661016949152542
            name: Recall
          - type: average_precision
            value: 0.5805485796313634
            name: Average Precision

CrossEncoder based on cross-encoder/nli-deberta-v3-base

This is a Cross Encoder model finetuned from cross-encoder/nli-deberta-v3-base using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

• Model Type: Cross Encoder
• Base model: cross-encoder/nli-deberta-v3-base
• Maximum Sequence Length: 256 tokens
• Number of Output Labels: 1 label
• Supported Modality: Text

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

Full Model Architecture

CrossEncoder(
  (0): Transformer({'transformer_task': 'sequence-classification', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'logits'}}, 'module_output_name': 'scores', 'architecture': 'DebertaV2ForSequenceClassification'})
)

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 CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("cross_encoder_model_id")
# Get scores for pairs of inputs
pairs = [
    ['The last time the planet was even four degrees warmer, Peter Brannen points out in The Ends of the World, his new history of the planet’s major extinction events, the oceans were hundreds of feet higher.', 'Almost all scientists acknowledge that the rate of species loss is greater now than at any time in human history, with extinctions occurring at rates hundreds of times higher than background extinction rates.'],
    ['[S]unspot activity on the surface of our star has dropped to a new low.', 'It has a regular activity cycle of starspots.'],
    ['More money is dedicated within the Department of Homeland Security to climate change than what\'s spent combating "Islamist terrorists radicalizing over the Internet in the United States of America."', 'Homeland security is officially defined by the National Strategy for Homeland Security as "a concerted national effort to prevent terrorist attacks within the United States, reduce America\'s vulnerability to terrorism, and minimize the damage and recover from attacks that do occur".'],
    ['Worst-case global heating scenarios may need to be revised upwards in light of a better understanding of the role of clouds, scientists have said.', 'Results from the CERES and other NASA missions, such as the Earth Radiation Budget Experiment (ERBE), could lead to a better understanding of the role of clouds and the energy cycle in global climate change.'],
    ['Prof Adam Scaife, a climate modelling expert at the UK’s Met Office, said the evidence for a link to shrinking Arctic ice was now good: ‘The consensus points towards that being a real effect.’”', 'Some models of modern climate exhibit Arctic amplification without changes in snow and ice cover.'],
]
scores = model.predict(pairs)
print(scores)
# [0.5664 0.4765 0.5621 0.5187 0.4973]

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'The last time the planet was even four degrees warmer, Peter Brannen points out in The Ends of the World, his new history of the planet’s major extinction events, the oceans were hundreds of feet higher.',
    [
        'Almost all scientists acknowledge that the rate of species loss is greater now than at any time in human history, with extinctions occurring at rates hundreds of times higher than background extinction rates.',
        'It has a regular activity cycle of starspots.',
        'Homeland security is officially defined by the National Strategy for Homeland Security as "a concerted national effort to prevent terrorist attacks within the United States, reduce America\'s vulnerability to terrorism, and minimize the damage and recover from attacks that do occur".',
        'Results from the CERES and other NASA missions, such as the Earth Radiation Budget Experiment (ERBE), could lead to a better understanding of the role of clouds and the energy cycle in global climate change.',
        'Some models of modern climate exhibit Arctic amplification without changes in snow and ice cover.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Classification

• Dataset: ce-val
• Evaluated with CrossEncoderClassificationEvaluator

Metric	Value
accuracy	0.6036
accuracy_threshold	0.5117
f1	0.6751
f1_threshold	0.4685
precision	0.5189
recall	0.9661
average_precision	0.5805

Training Details

Training Dataset

Unnamed Dataset

• Size: 7,419 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	float
  details	min: 7 tokens mean: 26.66 tokens max: 80 tokens	min: 8 tokens mean: 31.95 tokens max: 256 tokens	min: 0.0 mean: 0.52 max: 1.0

• Samples:

  sentence_0	sentence_1	label
  The last time the planet was even four degrees warmer, Peter Brannen points out in The Ends of the World, his new history of the planet’s major extinction events, the oceans were hundreds of feet higher.	Almost all scientists acknowledge that the rate of species loss is greater now than at any time in human history, with extinctions occurring at rates hundreds of times higher than background extinction rates.	0.0
  [S]unspot activity on the surface of our star has dropped to a new low.	It has a regular activity cycle of starspots.	1.0
  More money is dedicated within the Department of Homeland Security to climate change than what's spent combating "Islamist terrorists radicalizing over the Internet in the United States of America."	Homeland security is officially defined by the National Strategy for Homeland Security as "a concerted national effort to prevent terrorist attacks within the United States, reduce America's vulnerability to terrorism, and minimize the damage and recover from attacks that do occur".	1.0

• Loss: BinaryCrossEntropyLoss with these parameters:

  {
      "activation_fn": "torch.nn.modules.linear.Identity",
      "pos_weight": null
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 1
• fp16: True

All Hyperparameters

Click to expand

• do_predict: False
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• 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: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_ratio: None
• warmup_steps: 0
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• enable_jit_checkpoint: False
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• use_cpu: False
• seed: 42
• data_seed: None
• bf16: False
• fp16: True
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: -1
• ddp_backend: None
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: False
• ignore_data_skip: False
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': 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
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• 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
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• hub_always_push: False
• hub_revision: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_for_metrics: []
• eval_do_concat_batches: True
• auto_find_batch_size: False
• full_determinism: False
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• include_num_input_tokens_seen: no
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: True
• use_cache: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	ce-val_average_precision
-1	-1	0.5805

Training Time

• Training: 2.2 minutes

Framework Versions

• Python: 3.12.13
• Sentence Transformers: 5.4.1
• Transformers: 5.0.0
• PyTorch: 2.10.0+cu128
• Accelerate: 1.13.0
• Datasets: 4.0.0
• Tokenizers: 0.22.2

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",
}