tomaarsen/mpnet-base-all-nli-triplet

MPNet base trained on AllNLI triplets

This is a sentence-transformers model finetuned from microsoft/mpnet-base on the sentence-transformers/all-nli dataset. 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: microsoft/mpnet-base
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • sentence-transformers/all-nli
• Language: en
• License: apache-2.0

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: MPNetModel 
  (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("tomaarsen/mpnet-base-all-nli-triplet")
# Run inference
sentences = [
    'Then he ran.',
    'The people are running.',
    'The man is on his bike.',
]
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

Triplet

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

Metric	Value
cosine_accuracy	0.9004
dot_accuracy	0.0971
manhattan_accuracy	0.8969
euclidean_accuracy	0.8975
max_accuracy	0.9004

Triplet

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

Metric	Value
cosine_accuracy	0.915
dot_accuracy	0.0856
manhattan_accuracy	0.9115
euclidean_accuracy	0.9135
max_accuracy	0.915

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: 10.46 tokens max: 46 tokens	min: 6 tokens mean: 12.81 tokens max: 40 tokens	min: 5 tokens mean: 13.4 tokens max: 50 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: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

sentence-transformers/all-nli

• Dataset: sentence-transformers/all-nli at d482672
• Size: 6,584 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: 17.95 tokens max: 63 tokens	min: 4 tokens mean: 9.78 tokens max: 29 tokens	min: 5 tokens mean: 10.35 tokens max: 29 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: MultipleNegativesRankingLoss 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
• fp16: 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: 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: 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	-	-	0.6832	-
0.016	100	2.6355	1.0725	0.7924	-
0.032	200	0.9206	0.8342	0.8080	-
0.048	300	1.2567	0.7855	0.8133	-
0.064	400	0.7949	0.8857	0.7974	-
0.08	500	0.7583	0.9487	0.7872	-
0.096	600	1.0022	1.1312	0.7848	-
0.112	700	0.8178	1.2282	0.7895	-
0.128	800	0.9997	1.5132	0.7488	-
0.144	900	1.1173	1.4605	0.7473	-
0.16	1000	1.0089	1.3794	0.7543	-
0.176	1100	1.0235	1.4188	0.7640	-
0.192	1200	1.0031	1.2465	0.7570	-
0.208	1300	0.8286	1.4176	0.7426	-
0.224	1400	0.8411	1.1914	0.7600	-
0.24	1500	0.8389	1.1719	0.7820	-
0.256	1600	0.7144	1.1167	0.7691	-
0.272	1700	0.881	1.0747	0.7902	-
0.288	1800	0.8657	1.1576	0.7966	-
0.304	1900	0.7323	1.0122	0.8322	-
0.32	2000	0.6578	1.1248	0.8273	-
0.336	2100	0.6037	1.1194	0.8269	-
0.352	2200	0.641	1.1410	0.8341	-
0.368	2300	0.7843	1.0600	0.8328	-
0.384	2400	0.8222	0.9988	0.8161	-
0.4	2500	0.7287	1.2026	0.8395	-
0.416	2600	0.6035	0.8802	0.8273	-
0.432	2700	0.8275	1.1631	0.8458	-
0.448	2800	0.8483	0.9218	0.8316	-
0.464	2900	0.8813	1.1187	0.8147	-
0.48	3000	0.7408	0.9582	0.8246	-
0.496	3100	0.7886	0.9364	0.8261	-
0.512	3200	0.6064	0.8338	0.8302	-
0.528	3300	0.6415	0.7895	0.8650	-
0.544	3400	0.5766	0.7525	0.8571	-
0.56	3500	0.6212	0.8605	0.8572	-
0.576	3600	0.5773	0.7460	0.8419	-
0.592	3700	0.6104	0.7480	0.8580	-
0.608	3800	0.5754	0.7215	0.8657	-
0.624	3900	0.5525	0.7900	0.8630	-
0.64	4000	0.7802	0.7443	0.8612	-
0.656	4100	0.9796	0.7756	0.8748	-
0.672	4200	0.9355	0.6917	0.8796	-
0.688	4300	0.7081	0.6442	0.8832	-
0.704	4400	0.6868	0.6395	0.8891	-
0.72	4500	0.5964	0.5983	0.8820	-
0.736	4600	0.6618	0.5754	0.8861	-
0.752	4700	0.6957	0.6177	0.8803	-
0.768	4800	0.6375	0.5577	0.8881	-
0.784	4900	0.5481	0.5496	0.8835	-
0.8	5000	0.6626	0.5728	0.8949	-
0.816	5100	0.5192	0.5329	0.8935	-
0.832	5200	0.5856	0.5188	0.8935	-
0.848	5300	0.5142	0.5252	0.8920	-
0.864	5400	0.6404	0.5641	0.8885	-
0.88	5500	0.5466	0.5209	0.8929	-
0.896	5600	0.575	0.5170	0.8961	-
0.912	5700	0.626	0.5095	0.9001	-
0.928	5800	0.5631	0.4817	0.8984	-
0.944	5900	0.7301	0.4996	0.8984	-
0.96	6000	0.7712	0.5160	0.9014	-
0.976	6100	0.6203	0.5000	0.9007	-
0.992	6200	0.0005	0.4996	0.9004	-
1.0	6250	-	-	-	0.9150

Environmental Impact

Carbon emissions were measured using CodeCarbon.

• Energy Consumed: 0.306 kWh
• Carbon Emitted: 0.119 kg of CO2
• Hours Used: 1.661 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: 3.0.0.dev0
• Transformers: 4.41.1
• PyTorch: 2.3.0+cu121
• Accelerate: 0.30.1
• Datasets: 2.19.1
• 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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}