final/README.md

metadata

language:
  - en
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dataset_size:100K<n<1M
  - loss:CachedMultipleNegativesRankingLoss
base_model: FacebookAI/xlm-roberta-large
metrics:
  - cosine_accuracy
  - dot_accuracy
  - manhattan_accuracy
  - euclidean_accuracy
  - max_accuracy
widget:
  - source_sentence: The boy scowls
    sentences:
      - The boy is outside.
      - The man is in a city.
      - A woman at home.
  - source_sentence: A woman sings.
    sentences:
      - The woman is singing.
      - a man is wearing blue
      - The boys are eating.
  - source_sentence: A bird flying.
    sentences:
      - A butterfly flys freely.
      - She checks her phone.
      - A man is sleeping.
  - source_sentence: an eagle flies
    sentences:
      - A butterfly flys freely.
      - The men are together.
      - A man is sleeping.
  - source_sentence: There's a dock
    sentences:
      - There are people outdoors
      - Boy playing baseball.
      - A man is sleeping.
pipeline_tag: sentence-similarity
model-index:
  - name: SentenceTransformer based on FacebookAI/xlm-roberta-large
    results:
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: all nli dev
          type: all-nli-dev
        metrics:
          - type: cosine_accuracy
            value: 0.941
            name: Cosine Accuracy
          - type: dot_accuracy
            value: 0.062
            name: Dot Accuracy
          - type: manhattan_accuracy
            value: 0.937
            name: Manhattan Accuracy
          - type: euclidean_accuracy
            value: 0.938
            name: Euclidean Accuracy
          - type: max_accuracy
            value: 0.941
            name: Max Accuracy
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: all nli test
          type: all-nli-test
        metrics:
          - type: cosine_accuracy
            value: 0.943
            name: Cosine Accuracy
          - type: dot_accuracy
            value: 0.057
            name: Dot Accuracy
          - type: manhattan_accuracy
            value: 0.947
            name: Manhattan Accuracy
          - type: euclidean_accuracy
            value: 0.947
            name: Euclidean Accuracy
          - type: max_accuracy
            value: 0.947
            name: Max Accuracy

SentenceTransformer based on FacebookAI/xlm-roberta-large

This is a sentence-transformers model finetuned from FacebookAI/xlm-roberta-large on the sentence-transformers/all-nli dataset. 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: FacebookAI/xlm-roberta-large
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • sentence-transformers/all-nli
• Language: en

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: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 1024, '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("sentence_transformers_model_id")
# Run inference
sentences = [
    "There's a dock",
    'There are people outdoors',
    'Boy playing baseball.',
]
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

Triplet

• Dataset: all-nli-dev
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.941
dot_accuracy	0.062
manhattan_accuracy	0.937
euclidean_accuracy	0.938
max_accuracy	0.941

Triplet

• Dataset: all-nli-test
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.943
dot_accuracy	0.057
manhattan_accuracy	0.947
euclidean_accuracy	0.947
max_accuracy	0.947

Training Details

Training Dataset

sentence-transformers/all-nli

• Dataset: sentence-transformers/all-nli at d482672
• Size: 100,000 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 7 tokens mean: 10.9 tokens max: 52 tokens	min: 6 tokens mean: 13.62 tokens max: 42 tokens	min: 5 tokens mean: 14.76 tokens max: 55 tokens

• Samples:

  anchor	positive	negative
  A person on a horse jumps over a broken down airplane.	A person is outdoors, on a horse.	A person is at a diner, ordering an omelette.
  Children smiling and waving at camera	There are children present	The kids are frowning
  A boy is jumping on skateboard in the middle of a red bridge.	The boy does a skateboarding trick.	The boy skates down the sidewalk.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

sentence-transformers/all-nli

• Dataset: sentence-transformers/all-nli at d482672
• Size: 1,000 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 6 tokens mean: 20.31 tokens max: 83 tokens	min: 5 tokens mean: 10.71 tokens max: 35 tokens	min: 5 tokens mean: 11.39 tokens max: 32 tokens

• Samples:

  anchor	positive	negative
  Two women are embracing while holding to go packages.	Two woman are holding packages.	The men are fighting outside a deli.
  Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.	Two kids in numbered jerseys wash their hands.	Two kids in jackets walk to school.
  A man selling donuts to a customer during a world exhibition event held in the city of Angeles	A man selling donuts to a customer.	A woman drinks her coffee in a small cafe.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 1
• warmup_ratio: 0.1
• bf16: True
• batch_sampler: no_duplicates

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: 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: 1
• 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: 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: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	all-nli-dev_max_accuracy	all-nli-test_max_accuracy
0	0	-	-	0.613	-
0.016	100	3.4639	3.4199	0.621	-
0.032	200	3.4496	3.1967	0.841	-
0.048	300	2.2928	1.0476	0.864	-
0.064	400	1.2217	0.9993	0.871	-
0.08	500	1.1075	1.2674	0.85	-
0.096	600	1.2113	1.2565	0.866	-
0.112	700	1.0326	1.3313	0.855	-
0.128	800	1.2326	1.3698	0.851	-
0.144	900	1.2897	1.2690	0.855	-
0.16	1000	1.275	1.1231	0.863	-
0.176	1100	1.0823	1.2453	0.853	-
0.192	1200	1.1933	1.1119	0.868	-
0.208	1300	1.0102	0.9491	0.86	-
0.224	1400	0.8738	1.0682	0.87	-
0.24	1500	0.9482	0.8546	0.89	-
0.256	1600	0.6985	0.9136	0.88	-
0.272	1700	0.9908	0.9539	0.873	-
0.288	1800	1.0166	0.9277	0.878	-
0.304	1900	0.9441	0.9000	0.886	-
0.32	2000	0.8911	0.8364	0.891	-
0.336	2100	0.6746	0.8585	0.883	-
0.352	2200	0.7379	0.8332	0.888	-
0.368	2300	0.896	0.7617	0.89	-
0.384	2400	0.7901	0.7351	0.887	-
0.4	2500	0.811	0.7855	0.89	-
0.416	2600	0.6723	0.6756	0.899	-
0.432	2700	0.8839	0.7839	0.894	-
0.448	2800	0.9027	0.7319	0.903	-
0.464	2900	0.9276	0.7038	0.893	-
0.48	3000	0.7692	0.6653	0.903	-
0.496	3100	0.8044	0.6466	0.901	-
0.512	3200	0.6433	0.6145	0.906	-
0.528	3300	0.6642	0.5774	0.912	-
0.544	3400	0.5904	0.6054	0.907	-
0.56	3500	0.6378	0.5841	0.91	-
0.576	3600	0.5602	0.5444	0.921	-
0.592	3700	0.6436	0.5563	0.917	-
0.608	3800	0.588	0.5108	0.927	-
0.624	3900	0.5834	0.5059	0.925	-
0.64	4000	0.842	0.5217	0.929	-
0.656	4100	1.0995	0.5060	0.933	-
0.672	4200	0.9605	0.4842	0.928	-
0.688	4300	0.7811	0.4756	0.93	-
0.704	4400	0.7288	0.4650	0.938	-
0.72	4500	0.6636	0.4576	0.94	-
0.736	4600	0.7445	0.4552	0.934	-
0.752	4700	0.7687	0.4511	0.934	-
0.768	4800	0.7101	0.4446	0.936	-
0.784	4900	0.6586	0.4378	0.937	-
0.8	5000	0.789	0.4368	0.938	-
0.816	5100	0.6227	0.4344	0.941	-
0.832	5200	0.6994	0.4349	0.939	-
0.848	5300	0.687	0.4327	0.943	-
0.864	5400	0.76	0.4319	0.943	-
0.88	5500	0.6644	0.4323	0.941	-
0.896	5600	0.6535	0.4306	0.941	-
0.912	5700	0.7622	0.4289	0.941	-
0.928	5800	0.7053	0.4288	0.94	-
0.944	5900	0.8093	0.4289	0.94	-
0.96	6000	0.8658	0.4284	0.941	-
0.976	6100	0.7624	0.4283	0.941	-
0.992	6200	0.0003	0.4286	0.941	-
1.0	6250	-	-	-	0.947

Framework Versions

• Python: 3.9.10
• Sentence Transformers: 3.0.0
• Transformers: 4.41.2
• PyTorch: 2.3.0+cu121
• Accelerate: 0.26.1
• Datasets: 2.16.1
• 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",
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup}, 
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}