metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1602
  - loss:CosineSimilarityLoss
base_model: sentence-transformers/all-mpnet-base-v2
widget:
  - source_sentence: >-
      Has there been any recent discussion on the trend of women choosing to
      become mothers later in life?
    sentences:
      - >-
        British Columbia, Ontario, New Brunswick, Nova Scotia, and Prince Edward
        Island have fully implemented universal hearing screening programs.
      - >-
        If the first readings exceed the maximum allowable difference,
        measurements are taken for a second and, if necessary, a third time.
      - >-
        In recent years, practices have shifted and these professionals are now
        able to observe, assess, and consult on the child’s program at the
        centre rather than in an office visit.
  - source_sentence: >-
      Where can I find more information on facilitating extra-provincial ward
      adoptions in British Columbia?
    sentences:
      - >-
        You can refer to Practice Directive #2021-01 for more information on
        facilitating extra-provincial ward adoptions in British Columbia.
      - >-
        No, a Care Plan is not required if the child/youth has no special
        service needs.
      - >-
        Licensed ECEC programs may fall under the responsibility of one or more
        ministries and departments, including education, health, family, and/or
        social services.
  - source_sentence: What should be done if there are minor differences in openness requests?
    sentences:
      - >-
        If there are minor differences, it is advised to try to reach an
        acceptable compromise in a meeting.
      - >-
        Yes, the adoption can be completed in B.C. even if the child is from
        another province. However, the originating provincial or territorial
        child welfare authority is responsible for finalizing the adoption.
      - >-
        The new standards establish the breastfed child as the normative model
        for child growth and development.
  - source_sentence: >-
      Does the federal government in Canada manage Early Childhood Education and
      Care (ECEC)?
    sentences:
      - You can call a friend or relative to ask for help.
      - >-
        You can start introducing common food allergens to your baby as they
        begin eating solid foods. It's best to introduce them one at a time.
      - >-
        A search of the Parents' Registry should be requested at the time the
        child or youth is registered with the Adoption and Permanency Branch.
  - source_sentence: How can I order a birth certificate in British Columbia?
    sentences:
      - >-
        The Hague Convention is an international treaty that sets standards to
        ensure that the best interests of children and youth are protected.
      - >-
        Only the consents of the Director of Adoption and the child/youth aged
        12 or over are required.
      - >-
        The L value is -0.4488, the M value is 15.2759, and the S value is
        0.08380.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: SentenceTransformer based on sentence-transformers/all-mpnet-base-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: pregnancy val
          type: pregnancy_val
        metrics:
          - type: pearson_cosine
            value: 0.9454219117248748
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8647267521805166
            name: Spearman Cosine

SentenceTransformer based on sentence-transformers/all-mpnet-base-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-mpnet-base-v2. 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/all-mpnet-base-v2
Maximum Sequence Length: 384 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity

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': 384, '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})
  (2): Normalize()
)

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 = [
    'How can I order a birth certificate in British Columbia?',
    'Only the consents of the Director of Adoption and the child/youth aged 12 or over are required.',
    'The L value is -0.4488, the M value is 15.2759, and the S value is 0.08380.',
]
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

Semantic Similarity

Dataset: pregnancy_val
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.9454
spearman_cosine	0.8647

Training Details

Training Dataset

Unnamed Dataset

Size: 1,602 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: 16.16 tokens
max: 34 tokens

min: 9 tokens
mean: 28.61 tokens
max: 75 tokens

min: 0.0
mean: 0.51
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 7 tokens mean: 16.16 tokens max: 34 tokens	min: 9 tokens mean: 28.61 tokens max: 75 tokens	min: 0.0 mean: 0.51 max: 1.0

Samples:

sentence_0	sentence_1	label
`What kind of hearing screening programs do other provinces and territories in Canada have?`	`Unless an adoption placement has already been secured and a brief interim placement with a caregiver is required, the child should be 6 months of age or younger.`	`0.0`
`Are there resources available for children with learning disabilities in early childhood programs?`	`Yes, most PTs dedicate resources, programs or staff to support children with learning disabilities and other special needs.`	`1.0`
`What is parental leave?`	`Parental leave is a type of benefit that allows parents to take time off work after the birth or adoption of a child. The text mentions it but does not provide specific details about the duration or requirements in Canada.`	`1.0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
multi_dataset_batch_sampler: round_robin

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: 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: 3
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
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: 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: round_robin

Training Logs

Epoch	Step	pregnancy_val_spearman_cosine
1.0	101	0.8647

Framework Versions

Python: 3.13.1
Sentence Transformers: 3.4.1
Transformers: 4.49.0
PyTorch: 2.6.0
Accelerate: 1.4.0
Datasets: 3.3.2
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",
}