metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:416
  - loss:BatchSemiHardTripletLoss
base_model: BAAI/bge-base-en
widget:
  - source_sentence: |

      Name : CloudMetric Solutions
      Category: Data Analytics, Virtual Infrastructure Management
      Department: Engineering
      Location: Toronto, Canada
      Amount: 1644.75
      Card: Real-Time Resource Monitoring
      Trip Name: unknown
    sentences:
      - |

        Name : Nimbus Networks Inc.
        Category: Cloud Services, Application Hosting
        Department: Research & Development
        Location: Austin, TX
        Amount: 1134.67
        Card: NextGen Application Deployment
        Trip Name: unknown
      - |

        Name : Allianz
        Category: Insurance Services, Financial Services
        Department: Finance
        Location: New York, NY
        Amount: 2547.39
        Card: Quarterly Coverage Evaluation
        Trip Name: unknown
      - |

        Name : Connexis Group
        Category: Venue Logistics Services, Corporate Membership Consultancy
        Department: Sales
        Location: Berlin, Germany
        Amount: 1478.55
        Card: International Trade Show Engagement
        Trip Name: unknown
  - source_sentence: |

      Name : BuroPro Services
      Category: Facilities Management, Maintenance Solutions
      Department: Office Administration
      Location: Berlin, Germany
      Amount: 879.99
      Card: Monthly Equipment Oversight
      Trip Name: unknown
    sentences:
      - |

        Name : SynthioSolutions Global
        Category: Technology Consulting, Research Services
        Department: Research & Development
        Location: Singapore
        Amount: 1342.67
        Card: Advanced Data Integration Project
        Trip Name: unknown
      - |

        Name : Papyrus Solutions Inc.
        Category: Workspace Solutions, Office Technology Rentals
        Department: Office Administration
        Location: Dublin, Ireland
        Amount: 1348.56
        Card: Enhanced Work Efficiency Initiative
        Trip Name: unknown
      - |

        Name : City Shuttle Services
        Category: Transportation, Logistics
        Department: Sales
        Location: San Francisco, CA
        Amount: 85.0
        Card: Sales Team Travel Fund
        Trip Name: Client Meeting in Bay Area
  - source_sentence: |

      Name : SkillAdvance Academy
      Category: Online Learning Platform, Professional Development
      Department: Engineering
      Location: Austin, TX
      Amount: 1875.67
      Card: Continuous Improvement Initiative
      Trip Name: unknown
    sentences:
      - |

        Name : ComplyTech Solutions
        Category: Regulatory Software, Consultancy Services
        Department: Compliance
        Location: Brussels, Belgium
        Amount: 1095.45
        Card: Regulatory Compliance Optimization Plan
        Trip Name: unknown
      - |

        Name : AlphaTech Solutions
        Category: Computer & Electronics Retail
        Department: Research & Development
        Location: Toronto, Canada
        Amount: 1599.99
        Card: Innovative Hardware Acquisition
        Trip Name: unknown
      - |

        Name : Craft Gate Systems
        Category: Payment Processing Gateway, Data Analytics Software
        Department: Finance
        Location: Austin, TX
        Amount: 1132.58
        Card: Quarterly Revenue Analysis
        Trip Name: unknown
  - source_sentence: |

      Name : Rising Tide Solutions
      Category: IT Resource Management
      Department: Engineering
      Location: Amsterdam, Netherlands
      Amount: 1423.57
      Card: Cloud Transition Project
      Trip Name: unknown
    sentences:
      - |

        Name : GigaTrend
        Category: Data Services, Cloud Software Solutions
        Department: Research & Development
        Location: London, UK
        Amount: 1345.67
        Card: Data-Driven Innovation Project
        Trip Name: unknown
      - |

        Name : Apex Innovations Group
        Category: Business Consulting, Training Services
        Department: Executive
        Location: Sydney, Australia
        Amount: 1575.34
        Card: Leadership Development Program
        Trip Name: unknown
      - |

        Name : Aegis Risk Consultants
        Category: Executive Risk Management, Enterprise Solutions
        Department: Legal
        Location: London, UK
        Amount: 1743.56
        Card: Leadership Liability Initiative
        Trip Name: unknown
  - source_sentence: |

      Name : Allegro Integrations
      Category: Payment Processing Solutions, Financial Technology Services
      Department: Finance
      Location: Dublin, Ireland
      Amount: 1298.75
      Card: Bi-annual Financial Systems Audit
      Trip Name: unknown
    sentences:
      - |

        Name : Banyan Tree Pte Ltd
        Category: General Contractors - Residential and Commercial
        Department: Office Administration
        Location: Houston, TX
        Amount: 987.65
        Card: Operational Infrastructure Management
        Trip Name: unknown
      - >

        Name : InsightWave Research

        Category: Business Intelligence Consultations, Market Expansion Strategy
        Services

        Department: Marketing

        Location: Tokyo, Japan

        Amount: 2034.67

        Card: Global Market Insights Program

        Trip Name: unknown
      - |

        Name : ComplyTech Solutions
        Category: Regulatory Software, Consultancy Services
        Department: Compliance
        Location: Brussels, Belgium
        Amount: 1095.45
        Card: Regulatory Compliance Optimization Plan
        Trip Name: unknown
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy
model-index:
  - name: SentenceTransformer based on BAAI/bge-base-en
    results:
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: bge base en train
          type: bge-base-en-train
        metrics:
          - type: cosine_accuracy
            value: 0.4759615361690521
            name: Cosine Accuracy
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: bge base en eval
          type: bge-base-en-eval
        metrics:
          - type: cosine_accuracy
            value: 0
            name: Cosine Accuracy

SentenceTransformer based on BAAI/bge-base-en

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 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': 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.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("ppuva1/finetuned-bge-base-en")
# Run inference
sentences = [
    '\nName : Allegro Integrations\nCategory: Payment Processing Solutions, Financial Technology Services\nDepartment: Finance\nLocation: Dublin, Ireland\nAmount: 1298.75\nCard: Bi-annual Financial Systems Audit\nTrip Name: unknown\n',
    '\nName : Banyan Tree Pte Ltd\nCategory: General Contractors - Residential and Commercial\nDepartment: Office Administration\nLocation: Houston, TX\nAmount: 987.65\nCard: Operational Infrastructure Management\nTrip Name: unknown\n',
    '\nName : ComplyTech Solutions\nCategory: Regulatory Software, Consultancy Services\nDepartment: Compliance\nLocation: Brussels, Belgium\nAmount: 1095.45\nCard: Regulatory Compliance Optimization Plan\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

Datasets: bge-base-en-train and bge-base-en-eval
Evaluated with TripletEvaluator

Metric	bge-base-en-train	bge-base-en-eval
cosine_accuracy	0.476	0.0

Training Details

Training Dataset

Unnamed Dataset

Size: 416 training samples
Columns: sentence and label

Approximate statistics based on the first 416 samples:

	sentence	label
type	string	int
details	min: 32 tokens mean: 39.99 tokens max: 49 tokens	0: ~3.12% 1: ~3.12% 2: ~3.85% 3: ~4.81% 4: ~2.16% 5: ~4.33% 6: ~4.57% 7: ~3.85% 8: ~5.05% 9: ~4.09% 10: ~2.88% 11: ~4.33% 12: ~2.16% 13: ~4.09% 14: ~3.61% 15: ~5.77% 16: ~3.12% 17: ~6.01% 18: ~5.05% 19: ~2.64% 20: ~3.37% 21: ~2.88% 22: ~4.57% 23: ~2.64% 24: ~2.64% 25: ~3.85% 26: ~1.44%

Samples:

sentence	label
`Name : InnovaThink Global Category: Management Consultancy, Technical Training Services Department: HR Location: Zurich, Switzerland Amount: 1675.32 Card: Innovation and Efficiency Program Trip Name: unknown`	`0`
`Name : Global Wellness Network Category: Corporate Wellness Programs, Employee Engagement Department: HR Location: Berlin, Germany Amount: 1285.75 Card: Wellness and Engagement Program Trip Name: unknown`	`1`
`Name : Wong & Lim Category: Technical Equipment Services, Facility Services Department: Office Administration Location: Berlin, Germany Amount: 458.29 Card: Monthly Equipment Care Program Trip Name: unknown`	`2`

Loss: BatchSemiHardTripletLoss

Evaluation Dataset

Unnamed Dataset

Size: 104 evaluation samples
Columns: sentence and label

Approximate statistics based on the first 104 samples:

	sentence	label
type	string	int
details	min: 32 tokens mean: 39.19 tokens max: 46 tokens	0: ~1.92% 1: ~0.96% 2: ~4.81% 3: ~1.92% 5: ~5.77% 6: ~7.69% 7: ~4.81% 8: ~3.85% 9: ~5.77% 10: ~2.88% 11: ~4.81% 12: ~2.88% 13: ~1.92% 14: ~2.88% 15: ~0.96% 16: ~1.92% 17: ~3.85% 18: ~4.81% 19: ~3.85% 20: ~1.92% 21: ~0.96% 22: ~5.77% 23: ~7.69% 24: ~7.69% 25: ~4.81% 26: ~2.88%

Samples:

sentence	label
`Name : Aegis Risk Consultants Category: Executive Risk Management, Enterprise Solutions Department: Legal Location: London, UK Amount: 1743.56 Card: Leadership Liability Initiative Trip Name: unknown`	`11`
`Name : Vinobia Lounge Category: Culinary Experiences, Networking Venues Department: Marketing Location: Dallas, TX Amount: 651.58 Card: Innovative Marketing Strategies Trip Name: Annual Marketing Event`	`8`
`Name : Freenet AG Category: Telecommunication Services Department: IT Operations Location: Zurich, Switzerland Amount: 2794.37 Card: Infrastructure Support Services Trip Name: unknown`	`25`

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: 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: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	Validation Loss	bge-base-en-train_cosine_accuracy	bge-base-en-eval_cosine_accuracy
-1	-1	-	-	0.8510	-
3.8462	100	4.9979	5.0174	0.4760	-
-1	-1	-	-	-	0.0

Framework Versions

Python: 3.11.8
Sentence Transformers: 3.4.1
Transformers: 4.48.2
PyTorch: 2.6.0
Accelerate: 1.3.0
Datasets: 3.2.0
Tokenizers: 0.21.0

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",
}

BatchSemiHardTripletLoss

@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}
}