metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:10501
  - loss:CosineSimilarityLoss
base_model: klue/roberta-base
widget:
  - source_sentence: 아침마다 제가 원하는 시간에 맛있는 조식도 먹을 수 있었어요.
    sentences:
      - 매일 아침 내가 원하는 시간에 맛있는 아침식사를 먹을 수 있었습니다.
      - 태풍과 폭염 중 어떤 것이 올까요?
      - 떼르미니 역에서 5분 이내고 주변에 마트 식당 빵집 등등 편의시설도 가득합니다.
  - source_sentence: 아무리 우수한 방역체계도 신뢰 없이는 작동하기 어렵습니다.
    sentences:
      - 좋은 위치와 좋은 숙소와 좋은 호스트가 있습니다.
      - 위치도 룸도 모든 기 완벽한 곳이었다!
      - 콜센터 시설 내외부 방역도 철저히 실시하기로 했다.
  - source_sentence: 굳이 모든 메일을 다 가지고 있을 필요는 없어. 중요하지 않은 학회 홍보 메일은 지워도 돼.
    sentences:
      - 바르셀로나에 가실 거면 시내에 안 계셔도 된다면 이 숙소를 추천해 드릴게요!
      - 학교에서 온 메일 말고 학회 홍보메일만 삭제해줘
      - 사그라다 파밀리아까지는 걸어서 10분거리구요.
  - source_sentence: 더운물로 세탁하자.
    sentences:
      - 네가 시간 떼울 때 보고싶은 오락 프로그램 이름 알려주면 찾아볼께
      - 장인어른과의 약속에 정시에 가지 말고 일찍 나오세요.
      - 안방 취침등 또는 형광등은 어떻게 켜?
  - source_sentence: 또한 숙소는 청결하고 아늑한 장소입니다.
    sentences:
      - 또한, 숙소는 깨끗하고 아늑한 곳입니다.
      - 깜빡하고 백화점 세일 일정 잊어버리면 안된다.
      - 전체적으로 집 내부가 너무 예뻤어요.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
co2_eq_emissions:
  emissions: 6.29574616666927
  energy_consumed: 0.014386922744112848
  source: codecarbon
  training_type: fine-tuning
  on_cloud: false
  cpu_model: Intel(R) Core(TM) i7-14700KF
  ram_total_size: 63.83439254760742
  hours_used: 0.044
  hardware_used: 1 x NVIDIA GeForce RTX 4090
model-index:
  - name: SentenceTransformer based on klue/roberta-base
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: pearson_cosine
            value: 0.3477070403258199
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.35560473197486514
            name: Spearman Cosine
          - type: pearson_cosine
            value: 0.9624051736790307
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.922152297127282
            name: Spearman Cosine

SentenceTransformer based on klue/roberta-base

This is a sentence-transformers model finetuned from klue/roberta-base. 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: klue/roberta-base
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: RobertaModel 
  (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("sentence_transformers_model_id")
# Run inference
sentences = [
    '또한 숙소는 청결하고 아늑한 장소입니다.',
    '또한, 숙소는 깨끗하고 아늑한 곳입니다.',
    '깜빡하고 백화점 세일 일정 잊어버리면 안된다.',
]
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

Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.3477
spearman_cosine	0.3556

Semantic Similarity

Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.9624
spearman_cosine	0.9222

Training Details

Training Dataset

Unnamed Dataset

Size: 10,501 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: 6 tokens
mean: 19.8 tokens
max: 81 tokens

min: 5 tokens
mean: 19.36 tokens
max: 64 tokens

min: 0.0
mean: 0.46
max: 1.0

	sentence_0	sentence_1	label
type	string	string	float
details	min: 6 tokens mean: 19.8 tokens max: 81 tokens	min: 5 tokens mean: 19.36 tokens max: 64 tokens	min: 0.0 mean: 0.46 max: 1.0

Samples:

sentence_0	sentence_1	label
`아울러, 4월 9일부터 5월말까지 EBS 교육사이트를 데이터 걱정 없이 이용할 수 있습니다`	`현장방문 신청 둘째 주인 11월 2일부터 11월 6일까지는 구분없이 신청할 수 있다.`	`0.08`
`내일 오전에 있는 수업 몇 시에 시작하더라?`	`남자친구 생일이 언제야?`	`0.0`
`아무리 우수한 방역체계도 신뢰 없이는 작동하기 어렵습니다.`	`콜센터 시설 내외부 방역도 철저히 실시하기로 했다.`	`0.12`

Loss: CosineSimilarityLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 4
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
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
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: 4
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
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, '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
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	spearman_cosine
0	0	-	0.3556
0.7610	500	0.0279	-
1.0	657	-	0.9086
1.5221	1000	0.0087	0.9158
2.0	1314	-	0.9177
2.2831	1500	0.0046	-
3.0	1971	-	0.9191
3.0441	2000	0.0034	0.9199
3.8052	2500	0.0027	-
4.0	2628	-	0.9222

Environmental Impact

Carbon emissions were measured using CodeCarbon.

Energy Consumed: 0.014 kWh
Carbon Emitted: 0.006 kg of CO2
Hours Used: 0.044 hours

Training Hardware

On Cloud: No
GPU Model: 1 x NVIDIA GeForce RTX 4090
CPU Model: Intel(R) Core(TM) i7-14700KF
RAM Size: 63.83 GB

Framework Versions

Python: 3.12.8
Sentence Transformers: 3.3.1
Transformers: 4.40.1
PyTorch: 2.5.1+cu118
Accelerate: 0.29.3
Datasets: 2.19.1
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",
}