Marco127/Base_T

SentenceTransformer based on sentence-transformers/multi-qa-mpnet-base-dot-v1

This is a sentence-transformers model finetuned from sentence-transformers/multi-qa-mpnet-base-dot-v1. 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: sentence-transformers/multi-qa-mpnet-base-dot-v1
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Dot Product

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': 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("Marco127/Base_T")
# Run inference
sentences = [
    '\nGuests are responsible for damages caused to hotel property according to the valid legal\nprescriptions of Hungary.',
    '\nGuests are responsible for damages caused to hotel property according to the valid legal\nprescriptions of Hungary.',
    '\nWe shall be happy to listen to any suggestions for improvement of the accommodation\nand catering services in the hotel. In case of any complaints we shall purposefully arrange\nthe rectification of any insufficiencies.',
]
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

Binary Classification

• Evaluated with BinaryClassificationEvaluator

Metric	Value
dot_accuracy	0.655
dot_accuracy_threshold	48.3617
dot_f1	0.5143
dot_f1_threshold	40.0116
dot_precision	0.36
dot_recall	0.9
dot_ap	0.3571
dot_mcc	0.0388

Training Details

Training Dataset

Unnamed Dataset

• Size: 684 training samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 684 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 17 tokens mean: 42.77 tokens max: 71 tokens	min: 17 tokens mean: 42.77 tokens max: 71 tokens	0: ~67.11% 1: ~32.89%

• Samples:

  sentence1	sentence2	label
  If a guest fails to vacate the room within the designated time, reception shall charge this guest for the following night's accommodation fee.	If a guest fails to vacate the room within the designated time, reception shall charge this guest for the following night's accommodation fee.	0
  If you do not want someone to enter your room, please hang the "do not disturb” card on your room’s outside door handle. It can be found in the entrance area of your room.	If you do not want someone to enter your room, please hang the "do not disturb” card on your room’s outside door handle. It can be found in the entrance area of your room.	0
  Owners are responsible for ensuring that animals are kept quiet between the hours of 10:00 pm and 06:00 am. In the case of failure to abide by this regulation the guest may be asked to leave the hotel without a refund of the price of the night's accommodation.	Owners are responsible for ensuring that animals are kept quiet between the hours of 10:00 pm and 06:00 am. In the case of failure to abide by this regulation the guest may be asked to leave the hotel without a refund of the price of the night's accommodation.	0

• Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

• Size: 171 evaluation samples
• Columns: sentence1, sentence2, and label
• Approximate statistics based on the first 171 samples:

  	sentence1	sentence2	label
  type	string	string	int
  details	min: 17 tokens mean: 42.01 tokens max: 71 tokens	min: 17 tokens mean: 42.01 tokens max: 71 tokens	0: ~64.91% 1: ~35.09%

• Samples:

  sentence1	sentence2	label
  We shall be happy to listen to any suggestions for improvement of the accommodation and catering services in the hotel. In case of any complaints we shall purposefully arrange the rectification of any insufficiencies.	We shall be happy to listen to any suggestions for improvement of the accommodation and catering services in the hotel. In case of any complaints we shall purposefully arrange the rectification of any insufficiencies.	0
  Between the hours of 10:00 pm and 06:00 am guests are obliged to maintain low noise levels.	Between the hours of 10:00 pm and 06:00 am guests are obliged to maintain low noise levels.	0
  The hotel’s inner courtyard parking facility may be used only upon availability of parking slots. Slots marked as ’Private’ are to be left free for their owners. For parking fees please consult the reception or see the website of the hotel.	The hotel’s inner courtyard parking facility may be used only upon availability of parking slots. Slots marked as ’Private’ are to be left free for their owners. For parking fees please consult the reception or see the website of the hotel.	1

• 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
• learning_rate: 2e-05
• num_train_epochs: 5
• 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
• torch_empty_cache_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: 5
• 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: 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: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	dot_ap
-1	-1	-	-	0.3571
2.2791	100	0.0011	0.0000	-
4.5581	200	0.0	0.0000	-

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.4.1
• Transformers: 4.48.3
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.2.0
• 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",
}

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}
}