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, 'architecture': '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("juanpprim/finetuned-bge-base-en")
# Run inference
sentences = [
    '\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 : TechSavvy Solutions\nCategory: Software Services, Online Subscription\nDepartment: Engineering\nLocation: Austin, TX\nAmount: 1200.0\nCard: Annual Engineering Tools Budget\nTrip Name: unknown\n',
    '\nName : Vitality Systems\nCategory: Facility Management, Health Services\nDepartment: Office Administration\nLocation: Chicago, IL\nAmount: 347.29\nCard: Office Wellness Initiative\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)
# tensor([[nan, nan, nan],
#         [nan, nan, nan],
#         [nan, nan, nan]])

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	0.8531	0.0

Training Details

Training Dataset

Unnamed Dataset

Size: 211 training samples
Columns: sentence and label

Approximate statistics based on the first 211 samples:

	sentence	label
type	string	int
details	min: 33 tokens mean: 39.82 tokens max: 49 tokens	0: ~3.79% 1: ~3.32% 2: ~3.79% 3: ~2.37% 4: ~5.21% 5: ~5.21% 6: ~4.27% 7: ~3.32% 8: ~3.79% 9: ~4.27% 10: ~3.32% 11: ~3.79% 12: ~2.84% 13: ~5.21% 14: ~3.32% 15: ~5.69% 16: ~4.27% 17: ~2.37% 18: ~2.84% 19: ~3.32% 20: ~3.79% 21: ~4.27% 22: ~2.84% 23: ~4.27% 24: ~4.74% 25: ~1.90% 26: ~1.90%

Samples:

sentence	label
`[object Object]Name : Transcend[object Object]Category: Upskilling[object Object]Department: Human Resource[object Object]Location: London, UK[object Object]Amount: 859.47[object Object]Card: Technology Skills Enhancement[object Object]Trip Name: unknown[object Object]`	`0`
`[object Object]Name : Ayden[object Object]Category: Financial Software[object Object]Department: Finance[object Object]Location: Berlin, DE[object Object]Amount: 1273.45[object Object]Card: Enterprise Technology Services[object Object]Trip Name: unknown[object Object]`	`1`
`[object Object]Name : Urban Sphere[object Object]Category: Utilities Management, Facility Services[object Object]Department: Office Administration[object Object]Location: New York, NY[object Object]Amount: 937.32[object Object]Card: Monthly Operations Budget[object Object]Trip Name: unknown[object Object]`	`2`

Loss: BatchSemiHardTripletLoss

Evaluation Dataset

Unnamed Dataset

Size: 52 evaluation samples
Columns: sentence and label

Approximate statistics based on the first 52 samples:

	sentence	label
type	string	int
details	min: 32 tokens mean: 38.37 tokens max: 43 tokens	0: ~1.92% 4: ~1.92% 5: ~11.54% 7: ~5.77% 8: ~5.77% 10: ~7.69% 11: ~3.85% 12: ~3.85% 13: ~1.92% 16: ~3.85% 17: ~1.92% 18: ~13.46% 19: ~5.77% 20: ~3.85% 21: ~3.85% 22: ~7.69% 23: ~3.85% 24: ~5.77% 25: ~5.77%

Samples:

sentence	label
`[object Object]Name : Tooly[object Object]Category: Survey Software, SaaS[object Object]Department: Marketing[object Object]Location: San Francisco, CA[object Object]Amount: 2000.0[object Object]Card: Annual Marketing Technology Budget[object Object]Trip Name: unknown[object Object]`	`10`
`[object Object]Name : CloudFlare Inc.[object Object]Category: Internet & Network Services, SaaS[object Object]Department: IT Operations[object Object]Location: New York, NY[object Object]Amount: 2000.0[object Object]Card: Annual Cloud Services Budget[object Object]Trip Name: unknown[object Object]`	`21`
`[object Object]Name : Gartner & Associates[object Object]Category: Consulting, Business Services[object Object]Department: Legal[object Object]Location: San Francisco, CA[object Object]Amount: 5000.0[object Object]Card: Legal Consultation Fund[object Object]Trip Name: unknown[object Object]`	`5`

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
bf16: 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: 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: True
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
prompts: None
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	bge-base-en-train_cosine_accuracy	bge-base-en-eval_cosine_accuracy
-1	-1	0.8531	0.0

Framework Versions

Python: 3.11.13
Sentence Transformers: 5.0.0
Transformers: 4.45.2
PyTorch: 2.7.1+cu126
Accelerate: 1.3.0
Datasets: 3.6.0
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}
}