This is a sentence-transformers model finetuned from BAAI/bge-base-en. 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: BAAI/bge-base-en
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 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': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("lzwcv/finetuned-bge-base-en")
# Run inference
sentences = [
    '\nName : Apex Innovations Group\nCategory: Business Consulting, Training Services\nDepartment: Executive\nLocation: Sydney, Australia\nAmount: 1575.34\nCard: Leadership Development Program\nTrip Name: unknown\n',
    '\nName : CloudFlare Inc.\nCategory: Internet & Network Services, SaaS\nDepartment: IT Operations\nLocation: New York, NY\nAmount: 2000.0\nCard: Annual Cloud Services Budget\nTrip Name: unknown\n',
    '\nName : EcoClean Systems\nCategory: Environmental Services, Industrial Equipment Care\nDepartment: Office Administration\nLocation: San Francisco, CA\nAmount: 952.63\nCard: Essential Facility Sustainability\nTrip Name: unknown\n',
]
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

Triplet

Dataset: bge-base-en-train
Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.8371
dot_accuracy	0.1629
manhattan_accuracy	0.8281
euclidean_accuracy	0.8371
max_accuracy	0.8371

Triplet

Dataset: bge-base-en-eval
Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	1.0
dot_accuracy	0.0
manhattan_accuracy	0.9714
euclidean_accuracy	1.0
max_accuracy	1.0

Training Details

Training Dataset

Unnamed Dataset

Size: 221 training samples
Columns: sentence and label

Approximate statistics based on the first 221 samples:

	sentence	label
type	string	int
details	min: 33 tokens mean: 39.6 tokens max: 49 tokens	0: ~4.52% 1: ~4.52% 2: ~5.43% 3: ~2.26% 4: ~2.26% 5: ~2.71% 6: ~3.17% 7: ~3.62% 8: ~2.71% 9: ~5.43% 10: ~2.71% 11: ~4.07% 12: ~1.81% 13: ~4.52% 14: ~4.98% 15: ~3.62% 16: ~4.52% 17: ~4.98% 18: ~4.52% 19: ~2.71% 20: ~2.71% 21: ~4.52% 22: ~3.62% 23: ~4.07% 24: ~3.17% 25: ~4.98% 26: ~1.81%

Samples:

sentence	label
`[object Object]Name : Quantifire Insights[object Object]Category: Predictive Analytics Solutions[object Object]Department: Marketing[object Object]Location: Zurich, Switzerland[object Object]Amount: 1275.58[object Object]Card: Customer Engagement Enhancement[object Object]Trip Name: unknown[object Object]`	`0`
`[object Object]Name : ElevateLearning Solutions[object Object]Category: E-Learning Platforms, Collaborative Software[object Object]Department: Engineering[object Object]Location: Toronto, Canada[object Object]Amount: 1523.89[object Object]Card: Dev Team Skill Boosting Initiative[object Object]Trip Name: unknown[object Object]`	`1`
`[object Object]Name : Innovative Patents Co.[object Object]Category: Intellectual Property Services, Legal Services[object Object]Department: Legal[object Object]Location: New York, NY[object Object]Amount: 3250.0[object Object]Card: Patent Acquisition Fund[object Object]Trip Name: unknown[object Object]`	`2`

Loss: BatchSemiHardTripletLoss

Evaluation Dataset

Unnamed Dataset

Size: 55 evaluation samples
Columns: sentence and label

Approximate statistics based on the first 55 samples:

	sentence	label
type	string	int
details	min: 32 tokens mean: 39.73 tokens max: 47 tokens	0: ~1.82% 1: ~5.45% 2: ~9.09% 3: ~3.64% 4: ~5.45% 5: ~1.82% 6: ~1.82% 7: ~5.45% 10: ~5.45% 11: ~5.45% 12: ~3.64% 13: ~1.82% 14: ~3.64% 15: ~3.64% 16: ~7.27% 17: ~1.82% 18: ~5.45% 19: ~5.45% 20: ~1.82% 21: ~1.82% 22: ~3.64% 23: ~1.82% 24: ~7.27% 25: ~3.64% 26: ~1.82%

Samples:

sentence	label
`[object Object]Name : CyberGuard Provisions[object Object]Category: Security Software Solutions, Data Protection Services[object Object]Department: Information Security[object Object]Location: San Francisco, CA[object Object]Amount: 879.92[object Object]Card: Digital Fortress Action Plan[object Object]Trip Name: unknown[object Object]`	`17`
`[object Object]Name : Sphere Financial Systems[object Object]Category: Financial Management Services, International Billing Solutions[object Object]Department: Finance[object Object]Location: London, United Kingdom[object Object]Amount: 856.47[object Object]Card: Cross-Border Transaction Reconciliation[object Object]Trip Name: unknown[object Object]`	`7`
`[object Object]Name : RBC[object Object]Category: Transaction Processing, Financial Services[object Object]Department: Finance[object Object]Location: Limassol, Cyprus[object Object]Amount: 843.56[object Object]Card: Quarterly Financial Management[object Object]Trip Name: unknown[object Object]`	`7`

Loss: BatchSemiHardTripletLoss

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 2e-05
num_train_epochs: 5
warmup_ratio: 0.1
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: 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: 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: 5
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: 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: 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
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	bge-base-en-eval_max_accuracy	bge-base-en-train_max_accuracy
0	0	-	0.8371
5.0	35	1.0	-

Framework Versions

Python: 3.10.0
Sentence Transformers: 3.1.1
Transformers: 4.45.2
PyTorch: 2.6.0+cu124
Accelerate: 1.3.0
Datasets: 3.2.0
Tokenizers: 0.20.3

Citation

BibTeX

Sentence Transformers

bibtex

@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",
}

BatchSemiHardTripletLoss

bibtex

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}