metadata

language: []
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:844
  - loss:CoSENTLoss
base_model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
datasets: []
metrics:
  - pearson_cosine
  - spearman_cosine
  - pearson_manhattan
  - spearman_manhattan
  - pearson_euclidean
  - spearman_euclidean
  - pearson_dot
  - spearman_dot
  - pearson_max
  - spearman_max
widget:
  - source_sentence: Hilf mir, das Software-Update durchzuführen
    sentences:
      - order query
      - support query
      - faq query
  - source_sentence: 马上给我提供这个商品的跟踪信息
    sentences:
      - payment query
      - technical support query
      - support query
  - source_sentence: Downgrade my subscription plan
    sentences:
      - support query
      - product query
      - product query
  - source_sentence: Help resolve issues with my operating system
    sentences:
      - technical support query
      - product query
      - product query
  - source_sentence: Ayúdame a solucionar problemas de red
    sentences:
      - product query
      - support query
      - product query
pipeline_tag: sentence-similarity
model-index:
  - name: >-
      SentenceTransformer based on
      sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: MiniLM dev
          type: MiniLM-dev
        metrics:
          - type: pearson_cosine
            value: 0.7960441122484267
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.8189711310679958
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.6824455970208276
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.701004701178111
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.6821384996384094
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.7065633287645454
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.7871337514786776
            name: Pearson Dot
          - type: spearman_dot
            value: 0.7979718712970215
            name: Spearman Dot
          - type: pearson_max
            value: 0.7960441122484267
            name: Pearson Max
          - type: spearman_max
            value: 0.8189711310679958
            name: Spearman Max
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: MiniLM test
          type: MiniLM-test
        metrics:
          - type: pearson_cosine
            value: 0.7614418952584415
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.7585961676423125
            name: Spearman Cosine
          - type: pearson_manhattan
            value: 0.620319727073133
            name: Pearson Manhattan
          - type: spearman_manhattan
            value: 0.6192118311486844
            name: Spearman Manhattan
          - type: pearson_euclidean
            value: 0.6116132687052156
            name: Pearson Euclidean
          - type: spearman_euclidean
            value: 0.6124276377795256
            name: Spearman Euclidean
          - type: pearson_dot
            value: 0.7670292333817905
            name: Pearson Dot
          - type: spearman_dot
            value: 0.7764817683428225
            name: Spearman Dot
          - type: pearson_max
            value: 0.7670292333817905
            name: Pearson Max
          - type: spearman_max
            value: 0.7764817683428225
            name: Spearman Max

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2. It maps sentences & paragraphs to a 384-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/paraphrase-multilingual-MiniLM-L12-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 384 tokens
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': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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("philipp-zettl/MiniLM-similarity-small")
# Run inference
sentences = [
    'Ayúdame a solucionar problemas de red',
    'support query',
    'product query',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

Dataset: MiniLM-dev
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.796
spearman_cosine	0.819
pearson_manhattan	0.6824
spearman_manhattan	0.701
pearson_euclidean	0.6821
spearman_euclidean	0.7066
pearson_dot	0.7871
spearman_dot	0.798
pearson_max	0.796
spearman_max	0.819

Semantic Similarity

Dataset: MiniLM-test
Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7614
spearman_cosine	0.7586
pearson_manhattan	0.6203
spearman_manhattan	0.6192
pearson_euclidean	0.6116
spearman_euclidean	0.6124
pearson_dot	0.767
spearman_dot	0.7765
pearson_max	0.767
spearman_max	0.7765

Training Details

Training Dataset

Unnamed Dataset

Size: 844 training samples
Columns: sentence1, sentence2, and score
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 score
type string string float
details
min: 6 tokens
mean: 10.83 tokens
max: 19 tokens

min: 4 tokens
mean: 5.34 tokens
max: 6 tokens

min: 0.0
mean: 0.5
max: 1.0

	sentence1	sentence2	score
type	string	string	float
details	min: 6 tokens mean: 10.83 tokens max: 19 tokens	min: 4 tokens mean: 5.34 tokens max: 6 tokens	min: 0.0 mean: 0.5 max: 1.0

Samples:

sentence1	sentence2	score
`Покажите мне доступные гостиницы в Москве`	`product query`	`1.0`
`أرني العروض المتاحة على الهواتف الذكية`	`product query`	`1.0`
`Tengo problemas con el micrófono, ¿puedes ayudarme?`	`product query`	`0.0`

Loss: CoSENTLoss with these parameters:

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

Evaluation Dataset

Unnamed Dataset

Size: 106 evaluation samples
Columns: sentence1, sentence2, and score
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 score
type string string float
details
min: 7 tokens
mean: 10.63 tokens
max: 17 tokens

min: 4 tokens
mean: 5.32 tokens
max: 6 tokens

min: 0.0
mean: 0.47
max: 1.0

	sentence1	sentence2	score
type	string	string	float
details	min: 7 tokens mean: 10.63 tokens max: 17 tokens	min: 4 tokens mean: 5.32 tokens max: 6 tokens	min: 0.0 mean: 0.47 max: 1.0

Samples:

sentence1	sentence2	score
`Help me with device driver installation`	`product query`	`0.0`
`Check the status of my account verification`	`product query`	`0.0`
`我怎样重置我的密码？`	`product query`	`0.0`

Loss: CoSENTLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
learning_rate: 2e-05
num_train_epochs: 8
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: 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
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: 8
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	MiniLM-dev_spearman_cosine	MiniLM-test_spearman_cosine
0.3704	10	5.613	1.4994	0.2761	-
0.7407	20	4.8872	1.3690	0.3483	-
1.1111	30	3.2993	1.0579	0.4657	-
1.4815	40	2.1968	0.6858	0.5935	-
1.8519	50	0.7306	0.5191	0.6528	-
2.2222	60	0.9746	0.3735	0.6998	-
2.5926	70	0.3889	0.3532	0.7393	-
2.9630	80	0.1857	0.3598	0.7554	-
3.3333	90	0.2923	0.2795	0.7714	-
3.7037	100	0.6776	0.2881	0.7825	-
4.0741	110	0.2404	0.2679	0.7887	-
4.4444	120	0.0168	0.2583	0.7918	-
4.8148	130	0.0179	0.2273	0.7980	-
5.1852	140	0.0006	0.2196	0.8023	-
5.5556	150	0.0276	0.2068	0.8066	-
5.9259	160	0.061	0.2063	0.8103	-
6.2963	170	0.0265	0.2259	0.8103	-
6.6667	180	0.0105	0.2236	0.8165	-
7.0370	190	0.0008	0.2208	0.8177	-
7.4074	200	0.361	0.2340	0.8171	-
7.7778	210	0.0	0.2345	0.8190	-
8.0	216	-	-	-	0.7586

Framework Versions

Python: 3.10.14
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.1+cu121
Accelerate: 0.33.0
Datasets: 2.21.0
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",
}

CoSENTLoss

@online{kexuefm-8847,
    title={CoSENT: A more efficient sentence vector scheme than Sentence-BERT},
    author={Su Jianlin},
    year={2022},
    month={Jan},
    url={https://kexue.fm/archives/8847},
}