metadata

language:
  - en
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:314315
  - loss:MultipleNegativesRankingLoss
base_model: microsoft/mpnet-base
widget:
  - source_sentence: A person dressed in red and black outside a cracked wall.
    sentences:
      - A person in red and black near a wall.
      - Two women are in a car with a man.
      - a baby cries while getting their diaper changed
  - source_sentence: >-
      A man with shoulder length dark hair sits near the rocks of a waterfront
      while holding a cigarette in his right hand.
    sentences:
      - A man holding a cigarette.
      - a pair of fencers practice together
      - Four skaters race each other.
  - source_sentence: A man is reading a newspaper in a car dealership.
    sentences:
      - A man is at a car dealership.
      - Guys wearing white shirts play around by the park.
      - People are outside.
  - source_sentence: A woman in black, seen from behind, sits next to a body of water.
    sentences:
      - A woman sits outside.
      - There are families playing in a fountain
      - A player is hoping to score a run.
  - source_sentence: >-
      AN older woman appears to read from a children's book in an indoor
      setting, while a seated gentleman in a service uniform looks on.
    sentences:
      - a man is sitting in a lawn chair
      - A woman reads from a book while a man watches.
      - Others look while two men carve a babecued hog
datasets:
  - sentence-transformers/all-nli
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy
  - cosine_accuracy_threshold
  - cosine_f1
  - cosine_f1_threshold
  - cosine_precision
  - cosine_recall
  - cosine_ap
  - cosine_mcc
model-index:
  - name: SentenceTransformer based on microsoft/mpnet-base
    results:
      - task:
          type: binary-classification
          name: Binary Classification
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: cosine_accuracy
            value: 0.9998531139835488
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: -0.043851763010025024
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.9999265515975029
            name: Cosine F1
          - type: cosine_f1_threshold
            value: -0.043851763010025024
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 1
            name: Cosine Precision
          - type: cosine_recall
            value: 0.9998531139835488
            name: Cosine Recall
          - type: cosine_ap
            value: 1
            name: Cosine Ap
          - type: cosine_mcc
            value: 0
            name: Cosine Mcc
          - type: cosine_accuracy
            value: 0.9998536085492608
            name: Cosine Accuracy
          - type: cosine_accuracy_threshold
            value: 0.09460622072219849
            name: Cosine Accuracy Threshold
          - type: cosine_f1
            value: 0.9999267989166241
            name: Cosine F1
          - type: cosine_f1_threshold
            value: 0.09460622072219849
            name: Cosine F1 Threshold
          - type: cosine_precision
            value: 1
            name: Cosine Precision
          - type: cosine_recall
            value: 0.9998536085492608
            name: Cosine Recall
          - type: cosine_ap
            value: 1
            name: Cosine Ap
          - type: cosine_mcc
            value: 0
            name: Cosine Mcc

SentenceTransformer based on microsoft/mpnet-base

This is a sentence-transformers model finetuned from microsoft/mpnet-base on the 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 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- 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: 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("mrunali496/mpnet-base-all-nli-pair")
# Run inference
sentences = [
    "AN older woman appears to read from a children's book in an indoor setting, while a seated gentleman in a service uniform looks on.",
    'A woman reads from a book while a man watches.',
    'Others look while two men carve a babecued hog',
]
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
cosine_accuracy	0.9999
cosine_accuracy_threshold	-0.0439
cosine_f1	0.9999
cosine_f1_threshold	-0.0439
cosine_precision	1.0
cosine_recall	0.9999
cosine_ap	1.0
cosine_mcc	0.0

Binary Classification

Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.9999
cosine_accuracy_threshold	0.0946
cosine_f1	0.9999
cosine_f1_threshold	0.0946
cosine_precision	1.0
cosine_recall	0.9999
cosine_ap	1.0
cosine_mcc	0.0

Training Details

Training Dataset

all-nli

Dataset: all-nli at d482672
Size: 314,315 training samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 5 tokens
mean: 17.03 tokens
max: 64 tokens

min: 4 tokens
mean: 9.62 tokens
max: 31 tokens

	anchor	positive
type	string	string
details	min: 5 tokens mean: 17.03 tokens max: 64 tokens	min: 4 tokens mean: 9.62 tokens max: 31 tokens

Samples:

anchor	positive
`A person on a horse jumps over a broken down airplane.`	`A person is outdoors, on a horse.`
`Children smiling and waving at camera`	`There are children present`
`A boy is jumping on skateboard in the middle of a red bridge.`	`The boy does a skateboarding trick.`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

all-nli

Dataset: all-nli at d482672
Size: 6,808 evaluation samples
Columns: anchor and positive
Approximate statistics based on the first 1000 samples:
anchor positive
type string string
details
min: 6 tokens
mean: 18.01 tokens
max: 63 tokens

min: 4 tokens
mean: 9.8 tokens
max: 29 tokens

	anchor	positive
type	string	string
details	min: 6 tokens mean: 18.01 tokens max: 63 tokens	min: 4 tokens mean: 9.8 tokens max: 29 tokens

Samples:

anchor	positive
`Two women are embracing while holding to go packages.`	`Two woman are holding packages.`
`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.`
`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.`

Loss: MultipleNegativesRankingLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
warmup_ratio: 0.1
fp16: True
load_best_model_at_end: 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: 8
per_device_eval_batch_size: 8
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: 3
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: True
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
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	cosine_ap
-1	-1	-	-	1.0
0.008	100	2.0126	1.2036	1.0
0.016	200	1.0366	0.3276	1.0
0.024	300	0.4426	0.1492	1.0
0.032	400	0.2518	0.1048	1.0
0.04	500	0.2026	0.0962	1.0
0.048	600	0.1818	0.0821	1.0
0.056	700	0.1797	0.0816	1.0
0.064	800	0.1845	0.0659	1.0
0.072	900	0.1474	0.0675	1.0
0.08	1000	0.1648	0.0750	1.0
-1	-1	-	-	1.0

The bold row denotes the saved checkpoint.

Framework Versions

Python: 3.12.3
Sentence Transformers: 3.4.1
Transformers: 4.41.1
PyTorch: 2.3.0+cu121
Accelerate: 0.30.1
Datasets: 3.2.0
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}
}