Add new SentenceTransformer model

df3619b verified over 1 year ago

19.9 kB

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:90000
  - loss:CosineSimilarityLoss
base_model: sentence-transformers/msmarco-distilbert-base-v4
widget:
  - source_sentence: roast lamb internal temperature celsius
    sentences:
      - >-
        For a moderated subform of flagellation, described as bastinado, the
        soles of a person's bare feet are used as a target for beating (see foot
        whipping). In some circumstances the word flogging is used loosely to
        include any sort of corporal punishment, including birching and caning.
      - >-
        The periodic table provides a useful framework for analyzing chemical
        behaviour, and is widely used in chemistry and other sciences. Dmitri
        Mendeleev published in 1869 the first widely recognized periodic table.
        He developed his table to illustrate periodic trends in the properties
        of the then-known elements.
      - >-
        Service & Support. Key to COROB's success, is the extensive direct and
        partner network. The highly appreciated network sets COROB apart from
        competition as its main target is clearly to deliver the highest
        standards of customer satisfaction.
  - source_sentence: ups danville va phone number
    sentences:
      - >-
        A polysaccharide is a type of sugar composed of a long chain of 
        monosaccharides. Examples are cellulose, starch, glycogen, and  chitin.
      - >-
        $7.99. Get an INSTANT estimate of the cost to Refinish Countertops! Our
        free calculator uses recent, trusted data to estimate costs for your
        Countertop Refinishing project. For a basic 120 square feet project in
        zip code 47474, the benchmark cost to Refinish Countertops ranges
        between $3.60 - $7.99 per square foot* .To estimate costs for your
        project: 1.or a basic 120 square feet project in zip code 47474, the
        benchmark cost to Refinish Countertops ranges between $3.60 - $7.99 per
        square foot* .
      - >-
        UPS Drop Box - Danville 370 Mt Cross Rd, Danville VA 24541 Phone Number:
        (800) 742-5877
  - source_sentence: incubation period of mallard duck eggs
    sentences:
      - >-
        Havana Climate & Temperature. 1  Havana, Cuba is at 2258'N, 8224'W, 75 m
        (246 ft). 2  Havana has a tropical wet and dry/ savanna climate
        (Kppen-Geiger classification: Aw) with a pronounced dry season in the
        low-sun months , no cold season, wet season is in the high-sun months.
      - >-
        compression injuries such as carpal tunnel syndrome and compression
        neuropathy. A nerve conduction velocity test is often done at the same
        time as an electromyogram (EMG). Pinched Nerve Overview A pinched nerve
        causes pain, numbness, or tingling in the affected area due to pressure
        on a nerve.
      - >-
        1 57% - How long is mail from kamloops to ontario? 2  62% - How long
        does mail take to go from ontario to montreal? 3  74% - How long does it
        take for mail from kentucky to ontario? 4  64% - How long would it take
        to mail a package from ontario to georgia? 71% - How long does it take
        to get mail from vermont to ontario canada? 2  62% - How long does mail
        take to travel from ontario to new brunswick? 3  66% - How long does it
        take for mail to get from ontario to washington sate?
  - source_sentence: how to cook a rack of lamb well done
    sentences:
      - >-
        As the 20th anniversary of Freddie's death looms — he died on November
        24, 1991 — his friends have been celebrating what would have been his
        65th birthday this week with an AIDS charity dinner for the Mercury
        Phoenix Trust at London's Savoy Hotel, marking the start of a
        fund-raising year.
      - >-
        Medication. Medication for ulcerative colitis can suppress the
        inflammation of the colon and allow for tissues to heal. Symptoms
        including diarrhea, bleeding, and abdominal pain can also be reduced and
        controlled with effective medication.
      - >-
        Rack Opinions. To roast a lamb rack with the traditional method, you
        place your seasoned racks in a ripping hot oven (around 450°F), and
        basically hope that the center reaches the perfect medium rare (around
        125 to 130°F), just as the exterior hits that
        deep-golden-brown-but-not-yet-black sweet spot.ack Opinions. To roast a
        lamb rack with the traditional method, you place your seasoned racks in
        a ripping hot oven (around 450°F), and basically hope that the center
        reaches the perfect medium rare (around 125 to 130°F), just as the
        exterior hits that deep-golden-brown-but-not-yet-black sweet spot.
  - source_sentence: when is best time of year to visit vegas
    sentences:
      - >-
        Extended Description of this Name (viewable to Premium Members only).
        You enjoy freedom from monotony and are stimulated by unexpected
        opportunities for meeting people, entertaining, or pursuing activities
        of a carefree nature. In your work, you find it difficult to be neat and
        orderly.viewable to Premium Members only). Below is a brief analysis of
        the first name only. For an analysis of your full name and destiny, see
        our full free Name and Birth Date Report service for further details.
      - >-
        Create anything you can imagine. Anywhere you are. The world's best
        imaging and design app is at the core of almost every creative project.
        Work across desktop and mobile devices to create and enhance your
        photographs, web and mobile app designs, 3D artwork, videos, and more.
      - >-
        Of all the Tiger subspecies, P. t. corbetti showed the highest genetic
        diversity in microsatellite alleles, indicating a fairly stable
        evolutionary history, and alleles found in other subspecies were almost
        always a subset of those found in P. t. corbetti. Luo et al.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: >-
      SentenceTransformer based on
      sentence-transformers/msmarco-distilbert-base-v4
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: val eval
          type: val-eval
        metrics:
          - type: pearson_cosine
            value: 0.9685934106430019
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8659063123134304
            name: Spearman Cosine

SentenceTransformer based on sentence-transformers/msmarco-distilbert-base-v4

This is a sentence-transformers model finetuned from sentence-transformers/msmarco-distilbert-base-v4. 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/msmarco-distilbert-base-v4
Maximum Sequence Length: 512 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': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel 
  (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("C-KAI/sbert-academic-group44")
# Run inference
sentences = [
    'when is best time of year to visit vegas',
    'Of all the Tiger subspecies, P. t. corbetti showed the highest genetic diversity in microsatellite alleles, indicating a fairly stable evolutionary history, and alleles found in other subspecies were almost always a subset of those found in P. t. corbetti. Luo et al.',
    'Extended Description of this Name (viewable to Premium Members only). You enjoy freedom from monotony and are stimulated by unexpected opportunities for meeting people, entertaining, or pursuing activities of a carefree nature. In your work, you find it difficult to be neat and orderly.viewable to Premium Members only). Below is a brief analysis of the first name only. For an analysis of your full name and destiny, see our full free Name and Birth Date Report service for further details.',
]
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: val-eval
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.9686
spearman_cosine	0.8659

Training Details

Training Dataset

Unnamed Dataset

Size: 90,000 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: 4 tokens
mean: 9.03 tokens
max: 32 tokens

min: 18 tokens
mean: 78.5 tokens
max: 201 tokens

min: 0.0
mean: 0.49
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 4 tokens mean: 9.03 tokens max: 32 tokens	min: 18 tokens mean: 78.5 tokens max: 201 tokens	min: 0.0 mean: 0.49 max: 1.0

Samples:

sentence_0	sentence_1	label
`where is sky zone located canada`	`In the 1950s, a dryer cost the equivalent of $1600 today and only ten percent of households used them. By 1997 seventy-eight percent of households owned a dry as the cost had fallen to an average $370. The report also includes a table of Brief Chronology of Clothes Dryer Improvements.`	`0.0`
`what is syracuse new york population`	`East Syracuse, New York. East Syracuse is an incorporated village and a suburb of the City of Syracuse in eastern Onondaga County, New York. United States. As of the 2010 U.S. census, the village had a population of 3,084. It is located immediately east of Syracuse, in the town of DeWitt.`	`1.0`
`what are the tiny blood vessels that surround alveoli`	`What is 'people-first language'? People-first or person-first language is a way of describing disability that involves putting the word person or people before the word disability or the name of a disability, rather than placing the disability first and using it as an adjective.`	`0.0`

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: 32
per_device_eval_batch_size: 32
num_train_epochs: 2
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: 32
per_device_eval_batch_size: 32
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: 2
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}
tp_size: 0
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	Training Loss	val-eval_spearman_cosine
0.0355	100	-	0.8657
0.0711	200	-	0.8657
0.1066	300	-	0.8658
0.1422	400	-	0.8658
0.1777	500	0.06	0.8659
0.2133	600	-	0.8659
0.2488	700	-	0.8659
0.2844	800	-	0.8659

Framework Versions

Python: 3.11.12
Sentence Transformers: 4.0.2
Transformers: 4.50.3
PyTorch: 2.6.0+cu124
Accelerate: 1.6.0
Datasets: 3.5.0
Tokenizers: 0.21.1

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",
}