tomaarsen/reranker-distilroberta-base-stsb

CrossEncoder based on distilbert/distilroberta-base

This is a Cross Encoder model finetuned from distilbert/distilroberta-base on the stsb dataset using the sentence-transformers library. It computes scores for pairs of texts, which can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

• Model Type: Cross Encoder
• Base model: distilbert/distilroberta-base
• Maximum Sequence Length: 514 tokens
• Training Dataset:
  • stsb
• Language: en

Model Sources

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

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("tomaarsen/reranker-distilroberta-base-stsb")
# Get scores for pairs...
pairs = [
    ['A man with a hard hat is dancing.', 'A man wearing a hard hat is dancing.'],
    ['A young child is riding a horse.', 'A child is riding a horse.'],
    ['A man is feeding a mouse to a snake.', 'The man is feeding a mouse to the snake.'],
    ['A woman is playing the guitar.', 'A man is playing guitar.'],
    ['A woman is playing the flute.', 'A man is playing a flute.'],
]
scores = model.predict(pairs)
print(scores.shape)
# [5]

# ... or rank different texts based on similarity to a single text
ranks = model.rank(
    'A man with a hard hat is dancing.',
    [
        'A man wearing a hard hat is dancing.',
        'A child is riding a horse.',
        'The man is feeding a mouse to the snake.',
        'A man is playing guitar.',
        'A man is playing a flute.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Correlation

• Datasets: stsb-validation and stsb-test
• Evaluated with CECorrelationEvaluator

Metric	stsb-validation	stsb-test
pearson	0.8773	0.8503
spearman	0.8754	0.8389

Training Details

Training Dataset

stsb

• Dataset: stsb at ab7a5ac
• Size: 5,749 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 16 characters mean: 31.92 characters max: 113 characters	min: 16 characters mean: 31.51 characters max: 94 characters	min: 0.0 mean: 0.45 max: 1.0

• Samples:

  sentence1	sentence2	score
  A plane is taking off.	An air plane is taking off.	1.0
  A man is playing a large flute.	A man is playing a flute.	0.76
  A man is spreading shreded cheese on a pizza.	A man is spreading shredded cheese on an uncooked pizza.	0.76

• Loss: BinaryCrossEntropyLoss

Evaluation Dataset

stsb

• Dataset: stsb at ab7a5ac
• Size: 1,500 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	float
  details	min: 12 characters mean: 57.37 characters max: 144 characters	min: 17 characters mean: 56.84 characters max: 141 characters	min: 0.0 mean: 0.42 max: 1.0

• Samples:

  sentence1	sentence2	score
  A man with a hard hat is dancing.	A man wearing a hard hat is dancing.	1.0
  A young child is riding a horse.	A child is riding a horse.	0.95
  A man is feeding a mouse to a snake.	The man is feeding a mouse to the snake.	1.0

• Loss: BinaryCrossEntropyLoss

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• num_train_epochs: 4
• warmup_ratio: 0.1
• bf16: 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: 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: 1.0
• num_train_epochs: 4
• 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: True
• 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: None
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• include_for_metrics: []
• 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
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	stsb-validation_spearman	stsb-test_spearman
-1	-1	-	-	-0.0150	-
0.2222	20	0.6905	-	-	-
0.4444	40	0.6548	-	-	-
0.6667	60	0.5906	-	-	-
0.8889	80	0.5631	0.5475	0.8589	-
1.1111	100	0.5517	-	-	-
1.3333	120	0.5473	-	-	-
1.5556	140	0.5454	-	-	-
1.7778	160	0.5402	0.5346	0.8760	-
2.0	180	0.542	-	-	-
2.2222	200	0.5229	-	-	-
2.4444	220	0.524	-	-	-
2.6667	240	0.5286	0.5373	0.8744	-
2.8889	260	0.5236	-	-	-
3.1111	280	0.5269	-	-	-
3.3333	300	0.5209	-	-	-
3.5556	320	0.5115	0.5409	0.8754	-
3.7778	340	0.5149	-	-	-
4.0	360	0.5084	-	-	-
-1	-1	-	-	-	0.8389

Environmental Impact

Carbon emissions were measured using CodeCarbon.

• Energy Consumed: 0.007 kWh
• Carbon Emitted: 0.003 kg of CO2
• Hours Used: 0.031 hours

Training Hardware

• On Cloud: No
• GPU Model: 1 x NVIDIA GeForce RTX 3090
• CPU Model: 13th Gen Intel(R) Core(TM) i7-13700K
• RAM Size: 31.78 GB

Framework Versions

• Python: 3.11.6
• Sentence Transformers: 3.5.0.dev0
• Transformers: 4.49.0.dev0
• PyTorch: 2.5.0+cu121
• Accelerate: 1.3.0
• Datasets: 2.20.0
• Tokenizers: 0.21.0

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",
}