Downloads · 30 days
36
11% of all-time downloads
ubco-mds-2025-capstone-fujitsu-1
ubco-mds-2025-capstone-fujitsu-1/bge-base-en-v1.5-cve-finetuned
bge-base-en-v1.5-cve-finetuned is a sentence similarity model from ubco-mds-2025-capstone-fujitsu-1. Use it when you need a score for how close two texts are. It is set up for sentence-transformers.
This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for retrieval.
Downloads · 30 days
36
11% of all-time downloads
All-time downloads
334
Public
Parameters
109M
438 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors438 MB · 100%
From the Hugging Face model README
This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for retrieval.
SentenceTransformer(
(0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
(1): Pooling({'embedding_dimension': 768, 'pooling_mode': 'cls', 'include_prompt': True})
(2): Normalize({})
)
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
queries = [
'web console not functioning properly',
]
documents = [
'Multiple cross-site scripting (XSS) vulnerabilities in IBM InfoSphere Information Server through 8.5 FP3, 8.7 through FP2, and 9.1 allow remote authenticated users to inject arbitrary web script or HTML via vectors related to the (1) web console and (2) repository management user interfaces.',
'Cross-site scripting (XSS) vulnerability in the Taxonomy Breadcrumb module 6.x before 6.x-1.1 for Drupal allows remote authenticated users, with administer taxonomy permissions, to inject arbitrary web script or HTML via the node title in a Breadcrumb display.',
'The Relish (Verve Connect) VH510 device with firmware before 1.0.1.6L0516 contains undocumented default admin credentials for the web management interface. A remote attacker could exploit this vulnerability to login and execute commands on the device, as well as upgrade the firmware image to a malicious version.',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 768] [3, 768]
# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[ 0.5274, -0.1275, 0.2536]])
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
validation| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.7696 |
| cosine_accuracy@3 | 0.8714 |
| cosine_accuracy@5 | 0.9001 |
| cosine_accuracy@10 | 0.9349 |
| cosine_precision@1 | 0.7696 |
| cosine_precision@3 | 0.2905 |
| cosine_precision@5 | 0.18 |
| cosine_precision@10 | 0.0935 |
| cosine_recall@1 | 0.7696 |
| cosine_recall@3 | 0.8714 |
| cosine_recall@5 | 0.9001 |
| cosine_recall@10 | 0.9349 |
| cosine_ndcg@10 | 0.8527 |
| cosine_mrr@10 | 0.8263 |
| cosine_map@100 | 0.8289 |
| sentence_0 | sentence_1 | |
|---|---|---|
| type | string | string |
| modality | text | text |
| details | <ul><li>min: 7 tokens</li><li>mean: 12.17 tokens</li><li>max: 21 tokens</li></ul> | <ul><li>min: 28 tokens</li><li>mean: 87.76 tokens</li><li>max: 512 tokens</li></ul> |
| sentence_0 | sentence_1 |
|---|---|
| <code>broadcom bcm4355c0 wifi issues</code> | <code>On Broadcom BCM4355C0 Wi-Fi chips 9.44.78.27.0.1.56 and other chips, properly crafted malicious over-the-air Fast Transition frames can potentially trigger internal Wi-Fi firmware heap and/or stack overflows, leading to denial of service or other effects, aka B-V2017061205.</code> |
| <code>Tenda AC10 firmware crashes when changing WiFi settings</code> | <code>Tenda AC10 V1.0 V15.03.06.23 is vulnerable to Buffer Overflow in form_fast_setting_wifi_set via the parameter ssid.</code> |
| <code>issues with apache axis2 administration console</code> | <code>Cross-site scripting (XSS) vulnerability in axis2-admin/axis2-admin/engagingglobally in the administration console in Apache Axis2/Java 1.4.1, 1.5.1, and possibly other versions, as used in SAP Business Objects 12, 3com IMC, and possibly other products, allows remote attackers to inject arbitrary web script or HTML via the modules parameter. NOTE: some of these details are obtained from third party information.</code> |
{
"scale": 20.0,
"similarity_fct": "cos_sim",
"gather_across_devices": false,
"directions": [
"query_to_doc"
],
"partition_mode": "joint",
"hardness_mode": null,
"hardness_strength": 0.0
}
per_device_train_batch_size: 16num_train_epochs: 1fp16: Trueper_device_eval_batch_size: 16multi_dataset_batch_sampler: round_robinper_device_train_batch_size: 16num_train_epochs: 1max_steps: -1learning_rate: 5e-05lr_scheduler_type: linearlr_scheduler_kwargs: Nonewarmup_steps: 0optim: adamw_torch_fusedoptim_args: Noneweight_decay: 0.0adam_beta1: 0.9adam_beta2: 0.999adam_epsilon: 1e-08optim_target_modules: Nonegradient_accumulation_steps: 1average_tokens_across_devices: Truemax_grad_norm: 1label_smoothing_factor: 0.0bf16: Falsefp16: Truebf16_full_eval: Falsefp16_full_eval: Falsetf32: Nonegradient_checkpointing: Falsegradient_checkpointing_kwargs: Nonetorch_compile: Falsetorch_compile_backend: Nonetorch_compile_mode: Noneuse_liger_kernel: Falseliger_kernel_config: Noneuse_cache: Falseneftune_noise_alpha: Nonetorch_empty_cache_steps: Noneauto_find_batch_size: Falselog_on_each_node: Truelogging_nan_inf_filter: Trueinclude_num_input_tokens_seen: nolog_level: passivelog_level_replica: warningdisable_tqdm: Falseproject: huggingfacetrackio_space_id: Nonetrackio_bucket_id: Nonetrackio_static_space_id: Noneper_device_eval_batch_size: 16prediction_loss_only: Trueeval_on_start: Falseeval_do_concat_batches: Trueeval_use_gather_object: Falseeval_accumulation_steps: Noneinclude_for_metrics: []batch_eval_metrics: Falsesave_only_model: Falsesave_on_each_node: Falseenable_jit_checkpoint: Falsepush_to_hub: Falsehub_private_repo: Nonehub_model_id: Nonehub_strategy: every_savehub_always_push: Falsehub_revision: Noneload_best_model_at_end: Falseignore_data_skip: Falserestore_callback_states_from_checkpoint: Falsefull_determinism: Falseseed: 42data_seed: Noneuse_cpu: Falseaccelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}parallelism_config: Nonedataloader_drop_last: Falsedataloader_num_workers: 0dataloader_pin_memory: Truedataloader_persistent_workers: Falsedataloader_prefetch_factor: Noneremove_unused_columns: Truelabel_names: Nonetrain_sampling_strategy: randomlength_column_name: lengthddp_find_unused_parameters: Noneddp_bucket_cap_mb: Noneddp_broadcast_buffers: Falseddp_static_graph: Noneddp_backend: Noneddp_timeout: 1800fsdp: []fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}deepspeed: Nonedebug: []skip_memory_metrics: Truedo_predict: Falseresume_from_checkpoint: Nonewarmup_ratio: Nonelocal_rank: -1prompts: Nonebatch_sampler: batch_samplermulti_dataset_batch_sampler: round_robinrouter_mapping: {}learning_rate_mapping: {}| Epoch | Step | Training Loss | validation_cosine_ndcg@10 |
|---|---|---|---|
| 0.0093 | 500 | 0.3136 | - |
| 0.0186 | 1000 | 0.1300 | - |
| 0.0279 | 1500 | 0.1098 | - |
| 0.0371 | 2000 | 0.1112 | - |
| 0.0464 | 2500 | 0.0935 | - |
| 0.0557 | 3000 | 0.0961 | - |
| 0.0650 | 3500 | 0.0856 | - |
| 0.0743 | 4000 | 0.0872 | - |
| 0.0836 | 4500 | 0.0857 | - |
| 0.0928 | 5000 | 0.0795 | - |
| 0.1021 | 5500 | 0.0787 | - |
| 0.1114 | 6000 | 0.0749 | - |
| 0.1207 | 6500 | 0.0747 | - |
| 0.1300 | 7000 | 0.0720 | - |
| 0.1393 | 7500 | 0.0720 | - |
| 0.1486 | 8000 | 0.0718 | - |
| 0.1578 | 8500 | 0.0651 | - |
| 0.1671 | 9000 | 0.0660 | - |
| 0.1764 | 9500 | 0.0725 | - |
| 0.1857 | 10000 | 0.0638 | 0.8264 |
| 0.1950 | 10500 | 0.0639 | - |
| 0.2043 | 11000 | 0.0698 | - |
| 0.2136 | 11500 | 0.0637 | - |
| 0.2228 | 12000 | 0.0646 | - |
| 0.2321 | 12500 | 0.0746 | - |
| 0.2414 | 13000 | 0.0623 | - |
| 0.2507 | 13500 | 0.0629 | - |
| 0.2600 | 14000 | 0.0650 | - |
| 0.2693 | 14500 | 0.0578 | - |
| 0.2785 | 15000 | 0.0618 | - |
| 0.2878 | 15500 | 0.0572 | - |
| 0.2971 | 16000 | 0.0535 | - |
| 0.3064 | 16500 | 0.0594 | - |
| 0.3157 | 17000 | 0.0553 | - |
| 0.3250 | 17500 | 0.0530 | - |
| 0.3343 | 18000 | 0.0566 | - |
| 0.3435 | 18500 | 0.0537 | - |
| 0.3528 | 19000 | 0.0597 | - |
| 0.3621 | 19500 | 0.0645 | - |
| 0.3714 | 20000 | 0.0508 | 0.8397 |
| 0.3807 | 20500 | 0.0517 | - |
| 0.3900 | 21000 | 0.0544 | - |
| 0.3992 | 21500 | 0.0499 | - |
| 0.4085 | 22000 | 0.0559 | - |
| 0.4178 | 22500 | 0.0528 | - |
| 0.4271 | 23000 | 0.0552 | - |
| 0.4364 | 23500 | 0.0570 | - |
| 0.4457 | 24000 | 0.0541 | - |
| 0.4550 | 24500 | 0.0536 | - |
| 0.4642 | 25000 | 0.0504 | - |
| 0.4735 | 25500 | 0.0493 | - |
| 0.4828 | 26000 | 0.0545 | - |
| 0.4921 | 26500 | 0.0548 | - |
| 0.5014 | 27000 | 0.0525 | - |
| 0.5107 | 27500 | 0.0571 | - |
| 0.5200 | 28000 | 0.0481 | - |
| 0.5292 | 28500 | 0.0523 | - |
| 0.5385 | 29000 | 0.0512 | - |
| 0.5478 | 29500 | 0.0541 | - |
| 0.5571 | 30000 | 0.0513 | 0.8433 |
| 0.5664 | 30500 | 0.0510 | - |
| 0.5757 | 31000 | 0.0548 | - |
| 0.5849 | 31500 | 0.0463 | - |
| 0.5942 | 32000 | 0.0448 | - |
| 0.6035 | 32500 | 0.0544 | - |
| 0.6128 | 33000 | 0.0481 | - |
| 0.6221 | 33500 | 0.0491 | - |
| 0.6314 | 34000 | 0.0485 | - |
| 0.6407 | 34500 | 0.0509 | - |
| 0.6499 | 35000 | 0.0518 | - |
| 0.6592 | 35500 | 0.0513 | - |
| 0.6685 | 36000 | 0.0516 | - |
| 0.6778 | 36500 | 0.0482 | - |
| 0.6871 | 37000 | 0.0526 | - |
| 0.6964 | 37500 | 0.0462 | - |
| 0.7057 | 38000 | 0.0420 | - |
| 0.7149 | 38500 | 0.0445 | - |
| 0.7242 | 39000 | 0.0430 | - |
| 0.7335 | 39500 | 0.0428 | - |
| 0.7428 | 40000 | 0.0412 | 0.8483 |
| 0.7521 | 40500 | 0.0418 | - |
| 0.7614 | 41000 | 0.0456 | - |
| 0.7706 | 41500 | 0.0442 | - |
| 0.7799 | 42000 | 0.0449 | - |
| 0.7892 | 42500 | 0.0463 | - |
| 0.7985 | 43000 | 0.0429 | - |
| 0.8078 | 43500 | 0.0392 | - |
| 0.8171 | 44000 | 0.0404 | - |
| 0.8264 | 44500 | 0.0494 | - |
| 0.8356 | 45000 | 0.0472 | - |
| 0.8449 | 45500 | 0.0456 | - |
| 0.8542 | 46000 | 0.0463 | - |
| 0.8635 | 46500 | 0.0398 | - |
| 0.8728 | 47000 | 0.0448 | - |
| 0.8821 | 47500 | 0.0428 | - |
| 0.8913 | 48000 | 0.0447 | - |
| 0.9006 | 48500 | 0.0408 | - |
| 0.9099 | 49000 | 0.0403 | - |
| 0.9192 | 49500 | 0.0419 | - |
| 0.9285 | 50000 | 0.0398 | 0.8515 |
| 0.9378 | 50500 | 0.0420 | - |
| 0.9471 | 51000 | 0.0393 | - |
| 0.9563 | 51500 | 0.0428 | - |
| 0.9656 | 52000 | 0.0451 | - |
| 0.9749 | 52500 | 0.0448 | - |
| 0.9842 | 53000 | 0.0361 | - |
| 0.9935 | 53500 | 0.0411 | - |
| 1.0 | 53851 | - | 0.8527 |
@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",
}
@misc{oord2019representationlearningcontrastivepredictive,
title={Representation Learning with Contrastive Predictive Coding},
author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
year={2019},
eprint={1807.03748},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/1807.03748},
}
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->