Hugging Face's logo

MugheesAwan11
/
bge-base-securiti-dataset-1-v8

Sentence Similarity
sentence-transformers
Safetensors
English
bert
feature-extraction
Generated from Trainer
dataset_size:900
loss:MatryoshkaLoss
loss:MultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Inference Endpoints
Model card Files Community
bge-base-securiti-dataset-1-v8 / README.md
MugheesAwan11's picture
MugheesAwan11
Add new SentenceTransformer model.
4b66b8c verified
preview code
|
raw
history blame contribute delete
No virus
55.8 kB
metadata
language:
  - en
license: apache-2.0
library_name: sentence-transformers
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:900
  - loss:MatryoshkaLoss
  - loss:MultipleNegativesRankingLoss
base_model: BAAI/bge-base-en-v1.5
datasets: []
metrics:
  - cosine_accuracy@1
  - cosine_accuracy@3
  - cosine_accuracy@5
  - cosine_accuracy@10
  - cosine_precision@1
  - cosine_precision@3
  - cosine_precision@5
  - cosine_precision@10
  - cosine_recall@1
  - cosine_recall@3
  - cosine_recall@5
  - cosine_recall@10
  - cosine_ndcg@10
  - cosine_mrr@10
  - cosine_map@100
widget:
  - source_sentence: >-
      ["Vendor Risk Assessment\n\nView\n\nBreach Management\n\nView\n\nPrivacy
      Policy Management\n\nView\n\nPrivacy Center\n\nView\n\nLearn
      more\n\nSecurity\n\nIdentify data risk and enable protection &
      control\n\nData Security Posture Management\n\nView\n\nData Access
      Intelligence & Governance\n\nView\n\nData Risk Management\n\nView\n\nData
      Breach Analysis\n\nView\n\nLearn more\n\nGovernance\n\nOptimize Data
      Governance with granular insights into your data\n\nData
      Catalog\n\nView\n\nData Lineage\n\nView\n\nData Quality\n\nView\n\nData
      Controls Orchestrator\n\nView\n\nSolutions\n\nTechnologies\n\nCovering you
      everywhere with 1000+ integrations across data
      systems.\n\nSnowflake\n\nView\n\nAWS\n\nView\n\nMicrosoft
      365\n\nView\n\nSalesforce\n\nView\n\nWorkday\n\nView\n\nGCP\n\nView\n\nAzure\n\nView\n\nOracle\n\nView\n\nLearn
      more\n\nRegulations\n\nAutomate compliance with global privacy
      regulations.\n\nUS California CCPA\n\nView\n\nUS California
      CPRA\n\nView\n\nEuropean Union GDPR\n\nView\n\nThailand’s
      PDPA\n\nView\n\nChina PIPL\n\nView\n\nCanada PIPEDA\n\nView\n\nBrazil's
      LGPD\n\nView\n\n\\+ More\n\nView\n\nLearn more\n\nRoles\n\nIdentify data
      risk and enable protection &
      control.\n\nPrivacy\n\nView\n\nSecurity\n\nView\n\nGovernance\n\nView\n\nMarketing\n\nView\n\nResources\n\nBlog\n\nRead
      through our articles written by industry experts\n\nCollateral\n\nProduct
      brochures, white papers, infographics, analyst reports and
      more.\n\nKnowledge Center\n\nLearn about the data privacy, security and
      governance landscape.\n\nSecuriti Education\n\nCourses and Certifications
      for data privacy, security and governance
      professionals.\n\nCompany\n\nAbout Us\n\nLearn all about Securiti, our
      mission and history\n\nPartner Program\n\nJoin our Partner
      Program\n\nContact Us\n\nContact us to learn more or schedule a
      demo\n\nNews Coverage\n\nRead about Securiti in the news\n\nPress
      Releases\n\nFind our latest press releases\n\nCareers\n\nJoin the"]
    sentences:
      - >-
        What is the purpose of tracking changes and transformations of data
        throughout its lifecycle?
      - >-
        What is the role of ePD in the European privacy regime and its relation
        to GDPR?
      - How can data governance be optimized using granular insights?
  - source_sentence: >-
      ['Learn more\n\nAsset and Data Discovery\n\nDiscover dark and native data
      assets\n\nLearn more\n\nData Access Intelligence & Governance\n\nIdentify
      which users have access to sensitive data and prevent unauthorized
      access\n\nLearn more\n\nData Privacy Automation\n\nPrivacyCenter.Cloud |
      Data Mapping | DSR Automation | Assessment Automation | Vendor Assessment
      | Breach Management | Privacy Notice\n\nLearn more\n\nSensitive Data
      Intelligence\n\nDiscover & Classify Structured and Unstructured Data |
      People Data Graph\n\nLearn more\n\nData Flow Intelligence &
      Governance\n\nPrevent sensitive data sprawl through real-time streaming
      platforms\n\nLearn more\n\nData Consent Automation\n\nFirst Party Consent
      | Third Party & Cookie Consent\n\nLearn more\n\nData Security Posture
      Management\n\nSecure sensitive data in hybrid multicloud and SaaS
      environments\n\nLearn more\n\nData Breach Impact Analysis &
      Response\n\nAnalyze impact of a data breach and coordinate response per
      global regulatory obligations\n\nLearn more\n\nData
      Catalog\n\nAutomatically catalog datasets and enable users to find,
      understand, trust and access data\n\nLearn more\n\nData Lineage\n\nTrack
      changes and transformations of data throughout its lifecycle\n\nData
      Controls Orchestrator\n\nView\n\nData Command Center\n\nView\n\nSensitive
      Data Intelligence\n\nView\n\nAsset Discovery\n\nData Discovery &
      Classification\n\nSensitive Data Catalog\n\nPeople Data Graph\n\nLearn
      more\n\nPrivacy\n\nAutomate compliance with global privacy
      regulations\n\nData Mapping Automation\n\nView\n\nData Subject Request
      Automation\n\nView\n\nPeople Data Graph\n\nView\n\nAssessment
      Automation\n\nView\n\nCookie Consent\n\nView\n\nUniversal
      Consent\n\nView\n\nVendor Risk Assessment\n\nView\n\nBreach
      Management\n\nView\n\nPrivacy Policy Management\n\nView\n\nPrivacy
      Center\n\nView\n\nLearn more\n\nSecurity\n\nIdentify data risk and enable
      protection & control\n\nData Security Posture Management\n\nView\n\nData
      Access Intelligence & Governance\n\nView\n\nData Risk
      Management\n\nView\n\nData Breach Analysis\n\nView\n\nLearn
      more\n\nGovernance\n\nOptimize Data Governance with granular insights into
      your data\n\nData Catalog\n\nView\n\nData Lineage\n\nView\n\nData
      Quality\n\nView\n\nData Controls Orchestrator\n\n', '\n\nView\n\nLearn
      more\n\nAsset and Data Discovery\n\nDiscover dark and native data
      assets\n\nLearn more\n\nData Access Intelligence & Governance\n\nIdentify
      which users have access to sensitive data and prevent unauthorized
      access\n\nLearn more\n\nData Privacy Automation\n\nPrivacyCenter.Cloud |
      Data Mapping | DSR Automation | Assessment Automation | Vendor Assessment
      | Breach Management | Privacy Notice\n\nLearn more\n\nSensitive Data
      Intelligence\n\nDiscover & Classify Structured and Unstructured Data |
      People Data Graph\n\nLearn more\n\nData Flow Intelligence &
      Governance\n\nPrevent sensitive data sprawl through real-time streaming
      platforms\n\nLearn more\n\nData Consent Automation\n\nFirst Party Consent
      | Third Party & Cookie Consent\n\nLearn more\n\nData Security Posture
      Management\n\nSecure sensitive data in hybrid multicloud and SaaS
      environments\n\nLearn more\n\nData Breach Impact Analysis &
      Response\n\nAnalyze impact of a data breach and coordinate response per
      global regulatory obligations\n\nLearn more\n\nData
      Catalog\n\nAutomatically catalog datasets and enable users to find,
      understand, trust and access data\n\nLearn more\n\nData Lineage\n\nTrack
      changes and transformations of data throughout its lifecycle\n\nData
      Controls Orchestrator\n\nView\n\nData Command Center\n\nView\n\nSensitive
      Data Intelligence\n\nView\n\nAsset Discovery\n\nData Discovery &
      Classification\n\nSensitive Data Catalog\n\nPeople Data Graph\n\nLearn
      more\n\nPrivacy\n\nAutomate compliance with global privacy
      regulations\n\nData Mapping Automation\n\nView\n\nData Subject Request
      Automation\n\nView\n\nPeople Data Graph\n\nView\n\nAssessment
      Automation\n\nView\n\nCookie Consent\n\nView\n\nUniversal
      Consent\n\nView\n\nVendor Risk Assessment\n\nView\n\nBreach
      Management\n\nView\n\nPrivacy Policy Management\n\nView\n\nPrivacy
      Center\n\nView\n\nLearn more\n\nSecurity\n\nIdentify data risk and enable
      protection & control\n\nData Security Posture Management\n\nView\n\nData
      Access Intelligence & Governance\n\nView\n\nData Risk
      Management\n\nView\n\nData Breach Analysis\n\nView\n\nLearn
      more\n\nGovernance\n\nOptimize Data Governance with granular insights into
      your data\n\nData Catalog\n\nView\n\nData Lineage\n\nView\n\nData
      Quality\n\nView\n\nData Controls']
    sentences:
      - >-
        What is the purpose of Asset and Data Discovery in data governance and
        security?
      - Which EU member states have strict cyber laws?
      - >-
        What is the obligation for organizations to provide Data Protection
        Impact Assessments (DPIAs) under the LGPD?
  - source_sentence: >-
      [' which the data is processed.\n\n**Right to Access:** Data subjects have
      the right to obtain confirmation whether or not the controller holds
      personal data about them, access their personal data, and obtain
      descriptions of data recipients.\n\n**Right to Rectification** : Under the
      right to rectification, data subjects can request the correction of their
      data.\n\n**Right to Erasure:** Data subjects have the right to request the
      erasure and destruction of the data that is no longer needed by the
      organization.\n\n**Right to Object:** The data subject has the right to
      prevent the data controller from processing personal data if such
      processing causes or is likely to cause unwarranted damage or distress to
      the data subject.\n\n**Right not to be Subjected to Automated
      Decision-Making** : The data subject has the right to not be subject to
      automated decision-making that significantly affects the individual.\n\n##
      Facts related to Ghana’s Data Protection Act 2012\n\n1\n\nWhile processing
      personal data, organizations must comply with eight privacy principles:
      lawfulness of processing, data quality, security measures, accountability,
      purpose specification, purpose limitation, openness, and data subject
      participation.\n\n2\n\nIn the event of a security breach, the data
      controller shall take measures to prevent the breach and notify the
      Commission and the data subject about the breach as soon as reasonably
      practicable after the discovery of the breach.\n\n3\n\nThe DPA specifies
      lawful grounds for data processing, including data subject’s consent, the
      performance of a contract, the interest of data subject and public
      interest, lawful obligations, and the legitimate interest of the data
      controller.\n\n4\n\nThe DPA requires data controllers to register with the
      Data Protection Commission (DPC).\n\n5\n\nThe DPA provides varying fines
      and terms of imprisonment according to the severity and sensitivity of the
      violation, such as any person who sells personal data may get fined up to
      2500 penalty units or up to five years imprisonment or both.\n\n###
      Forrester Names Securiti a Leader in the Privacy Management Wave Q4,
      2021\n\nRead the Report\n\n### Securiti named a Leader in the IDC
      MarketScape for Data Privacy Compliance Software\n\nRead the Report\n\nAt
      Securiti, our mission is to enable enterprises to safely harness the
      incredible power of data and the cloud by controlling the complex
      security, privacy and compliance risks.\n\nCopyright (C) 2023
      Securiti\n\nSitem']
    sentences:
      - >-
        What information is required for data subjects regarding data transfers
        under the GDPR, including personal data categories, data recipients,
        retention period, and automated decision making?
      - >-
        What privacy principles must organizations follow when processing
        personal data under Ghana's Data Protection Act 2012?
      - What is the purpose of Thailand's PDPA?
  - source_sentence: >-
      [" consumer has the right to have his/her personal data stored or
      processed by the data controller be deleted.\n\n## Portability\n\nThe
      consumer has a right to obtain a copy of his/her personal data in a
      portable, technically feasible and readily usable format that allows the
      consumer to transmit the data to another controller without
      hindrance.\n\n## Opt\n\nout\n\nThe consumer has the right to opt out of
      the processing of the personal data for purposes of targeted advertising,
      the sale of personal data, or profiling in furtherance of decisions that
      produce legal or similarly significant effects concerning the
      consumer.\n\n**Time period to fulfill DSR request:\n\n** All data subject
      rights’ requests (DSR requests) must be fulfilled by the data controller
      within a 45 day period.\n\n**Extension in time period:\n\n** data
      controllers may seek for an extension of 45 days in fulfilling the request
      depending on the complexity and number of the consumer's
      requests.\n\n**Denial of DSR request:\n\n** If a DSR request is to be
      denied, the data controller must inform the consumer of the reasons within
      a 45 days period.\n\n**Appeal against refusal:\n\n** Consumers have a
      right to appeal the decision for refusal of grant of the DSR request. The
      appeal must be decided within 45 days but the time period can be further
      extended by 60 additional days.\n\n**Limitation of DSR requests per
      year:\n\n** Requests for data portability may be made only twice in a
      year.\n\n**Charges:\n\n** DSR requests must be fulfilled free of charge
      once in a year. Any subsequent request within a 12 month period can be
      charged.\n\n**Authentication:\n\n** A data controller is not to respond to
      a consumer request unless it can authenticate the request using reasonably
      commercial means. A data controller can request additional information
      from the consumer for the purposes of authenticating the request.\n\n##
      Who must comply?\n\nCPA applies to all data controllers who conduct
      business in Colorado or produce or deliver commercial products or services
      that are intentionally targeted to residents of Colorado\n\nif they match
      any one or both of these conditions:\n\nIf they control or process the
      personal data of 100,000 consumers or more during a calendar year;
      or\n\nIf they derive revenue or receive a discount on the price of goods
      or services from the sale of personal data and process or control the
      personal data of 25,000"]
    sentences:
      - >-
        What is the US California CCPA and how does it relate to data privacy
        regulations?
      - >-
        What does the People Data Graph serve in terms of privacy, security, and
        governance?
      - >-
        What rights does a consumer have regarding the portability of their
        personal data?
  - source_sentence: >-
      ["PR and Federal Data Protection Act within Germany;\n\nTo promote
      awareness within the public related to the risks, rules, safeguards, and
      rights concerning the processing of personal data;\n\nTo handle all
      complaints raised by data subjects related to data processing in addition
      to carrying out investigations to find out if any data handler has
      breached any provisions of the Act;\n\n## Penalties for
      Non\n\ncompliance\n\nThe GDPR already laid down some stringent penalties
      for companies that would be found in breach of the law's provisions. More
      importantly, as opposed to other data protection laws such as the CCPA and
      CPRA, non-compliance with the law also meant penalties.\n\nGermany's
      Federal Data Protection Act has a slightly more lenient take in this
      regard. Suppose a data handler is found to have fraudulently collected
      data, processed, shared, or sold data without proper consent from the data
      subjects, not responded or responded with delay to a data subject request,
      or failed to inform the data subject of a breach properly. In that case,
      it can be fined up to €50,000.\n\nThis is in addition to the GDPR's €20
      million or 4% of the total worldwide annual turnover of the preceding
      financial year, whichever is higher, that any organisation found in breach
      of the law is subject to.\n\nHowever, for this fine to be applied, either
      the data subject, the Federal Commissioner, or the regulatory authority
      must file an official complaint.\n\n## How an Organization Can
      Operationalize the Law\n\nData handlers processing data inside Germany can
      remain compliant with the country's data protection law if they fulfill
      the following conditions:\n\nHave a comprehensive privacy policy that
      educates all users of their rights and how to contact the relevant
      personnel within the organisation in case of a query\n\nHire a competent
      Data Protection Officer that understands the GDPR and Federal Data
      Protection Act thoroughly and can lead compliance efforts within your
      organisation\n\nEnsure all the company's employees and staff are acutely
      aware of their responsibilities under the law\n\nConduct regular data
      protection impact assessments as well as data mapping exercises to ensure
      maximum efficiency in your compliance efforts\n\nNotify the relevant
      authorities of a data breach as soon as possible\n\n## How can Securiti
      Help\n\nData privacy and compliance have become incredibly vital in
      earning users' trust globally. Most users now expect most businesses to
      take all the relevant measures to ensure the data they collect is properly
      stored, protected, and maintained. Data protection laws have made such
      efforts legally mandatory"]
    sentences:
      - >-
        How does Data Access Intelligence & Governance prevent unauthorized
        access to sensitive data?
      - >-
        What is required for an official complaint to be filed under Germany's
        Federal Data Protection Act?
      - Why is tracking data lineage important for data management and security?
pipeline_tag: sentence-similarity
model-index:
  - name: SentenceTransformer based on BAAI/bge-base-en-v1.5
    results:
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 512
          type: dim_512
        metrics:
          - type: cosine_accuracy@1
            value: 0.07
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.26
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.44
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.63
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.07
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.08666666666666668
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.088
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.06299999999999999
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.07
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.26
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.44
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.63
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.3150525932481703
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.2180119047619047
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.23183767291183585
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 256
          type: dim_256
        metrics:
          - type: cosine_accuracy@1
            value: 0.06
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.24
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.44
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.06
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.07999999999999999
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.088
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.059999999999999984
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.06
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.24
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.44
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.2944478644544164
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.19998809523809516
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.21493741340512212
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 128
          type: dim_128
        metrics:
          - type: cosine_accuracy@1
            value: 0.07
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.21
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.4
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.6
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.07
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.06999999999999999
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.08
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.059999999999999984
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.07
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.21
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.4
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.6
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.29018137407094874
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.19626984126984123
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.21169474427113727
            name: Cosine Map@100
      - task:
          type: information-retrieval
          name: Information Retrieval
        dataset:
          name: dim 64
          type: dim_64
        metrics:
          - type: cosine_accuracy@1
            value: 0.07
            name: Cosine Accuracy@1
          - type: cosine_accuracy@3
            value: 0.17
            name: Cosine Accuracy@3
          - type: cosine_accuracy@5
            value: 0.32
            name: Cosine Accuracy@5
          - type: cosine_accuracy@10
            value: 0.53
            name: Cosine Accuracy@10
          - type: cosine_precision@1
            value: 0.07
            name: Cosine Precision@1
          - type: cosine_precision@3
            value: 0.056666666666666664
            name: Cosine Precision@3
          - type: cosine_precision@5
            value: 0.064
            name: Cosine Precision@5
          - type: cosine_precision@10
            value: 0.05299999999999999
            name: Cosine Precision@10
          - type: cosine_recall@1
            value: 0.07
            name: Cosine Recall@1
          - type: cosine_recall@3
            value: 0.17
            name: Cosine Recall@3
          - type: cosine_recall@5
            value: 0.32
            name: Cosine Recall@5
          - type: cosine_recall@10
            value: 0.53
            name: Cosine Recall@10
          - type: cosine_ndcg@10
            value: 0.2594266732084936
            name: Cosine Ndcg@10
          - type: cosine_mrr@10
            value: 0.17759523809523803
            name: Cosine Mrr@10
          - type: cosine_map@100
            value: 0.194555422694347
            name: Cosine Map@100

SentenceTransformer based on BAAI/bge-base-en-v1.5

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 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-v1.5
  • Maximum Sequence Length: 512 tokens
  • Output Dimensionality: 768 tokens
  • Similarity Function: Cosine Similarity
  • Language: en
  • License: apache-2.0

Model Sources

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("MugheesAwan11/bge-base-securiti-dataset-1-v8")
# Run inference
sentences = [
    '["PR and Federal Data Protection Act within Germany;\\n\\nTo promote awareness within the public related to the risks, rules, safeguards, and rights concerning the processing of personal data;\\n\\nTo handle all complaints raised by data subjects related to data processing in addition to carrying out investigations to find out if any data handler has breached any provisions of the Act;\\n\\n## Penalties for Non\\n\\ncompliance\\n\\nThe GDPR already laid down some stringent penalties for companies that would be found in breach of the law\'s provisions. More importantly, as opposed to other data protection laws such as the CCPA and CPRA, non-compliance with the law also meant penalties.\\n\\nGermany\'s Federal Data Protection Act has a slightly more lenient take in this regard. Suppose a data handler is found to have fraudulently collected data, processed, shared, or sold data without proper consent from the data subjects, not responded or responded with delay to a data subject request, or failed to inform the data subject of a breach properly. In that case, it can be fined up to €50,000.\\n\\nThis is in addition to the GDPR\'s €20 million or 4% of the total worldwide annual turnover of the preceding financial year, whichever is higher, that any organisation found in breach of the law is subject to.\\n\\nHowever, for this fine to be applied, either the data subject, the Federal Commissioner, or the regulatory authority must file an official complaint.\\n\\n## How an Organization Can Operationalize the Law\\n\\nData handlers processing data inside Germany can remain compliant with the country\'s data protection law if they fulfill the following conditions:\\n\\nHave a comprehensive privacy policy that educates all users of their rights and how to contact the relevant personnel within the organisation in case of a query\\n\\nHire a competent Data Protection Officer that understands the GDPR and Federal Data Protection Act thoroughly and can lead compliance efforts within your organisation\\n\\nEnsure all the company\'s employees and staff are acutely aware of their responsibilities under the law\\n\\nConduct regular data protection impact assessments as well as data mapping exercises to ensure maximum efficiency in your compliance efforts\\n\\nNotify the relevant authorities of a data breach as soon as possible\\n\\n## How can Securiti Help\\n\\nData privacy and compliance have become incredibly vital in earning users\' trust globally. Most users now expect most businesses to take all the relevant measures to ensure the data they collect is properly stored, protected, and maintained. Data protection laws have made such efforts legally mandatory"]',
    "What is required for an official complaint to be filed under Germany's Federal Data Protection Act?",
    'Why is tracking data lineage important for data management and security?',
]
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

Information Retrieval

Metric Value
cosine_accuracy@1 0.07
cosine_accuracy@3 0.26
cosine_accuracy@5 0.44
cosine_accuracy@10 0.63
cosine_precision@1 0.07
cosine_precision@3 0.0867
cosine_precision@5 0.088
cosine_precision@10 0.063
cosine_recall@1 0.07
cosine_recall@3 0.26
cosine_recall@5 0.44
cosine_recall@10 0.63
cosine_ndcg@10 0.3151
cosine_mrr@10 0.218
cosine_map@100 0.2318

Information Retrieval

Metric Value
cosine_accuracy@1 0.06
cosine_accuracy@3 0.24
cosine_accuracy@5 0.44
cosine_accuracy@10 0.6
cosine_precision@1 0.06
cosine_precision@3 0.08
cosine_precision@5 0.088
cosine_precision@10 0.06
cosine_recall@1 0.06
cosine_recall@3 0.24
cosine_recall@5 0.44
cosine_recall@10 0.6
cosine_ndcg@10 0.2944
cosine_mrr@10 0.2
cosine_map@100 0.2149

Information Retrieval

Metric Value
cosine_accuracy@1 0.07
cosine_accuracy@3 0.21
cosine_accuracy@5 0.4
cosine_accuracy@10 0.6
cosine_precision@1 0.07
cosine_precision@3 0.07
cosine_precision@5 0.08
cosine_precision@10 0.06
cosine_recall@1 0.07
cosine_recall@3 0.21
cosine_recall@5 0.4
cosine_recall@10 0.6
cosine_ndcg@10 0.2902
cosine_mrr@10 0.1963
cosine_map@100 0.2117

Information Retrieval

Metric Value
cosine_accuracy@1 0.07
cosine_accuracy@3 0.17
cosine_accuracy@5 0.32
cosine_accuracy@10 0.53
cosine_precision@1 0.07
cosine_precision@3 0.0567
cosine_precision@5 0.064
cosine_precision@10 0.053
cosine_recall@1 0.07
cosine_recall@3 0.17
cosine_recall@5 0.32
cosine_recall@10 0.53
cosine_ndcg@10 0.2594
cosine_mrr@10 0.1776
cosine_map@100 0.1946

Training Details

Training Dataset

Unnamed Dataset

  • Size: 900 training samples
  • Columns: positive and anchor
  • Approximate statistics based on the first 1000 samples:
    positive anchor
    type string string
    details
    • min: 512 tokens
    • mean: 512.0 tokens
    • max: 512 tokens
    • min: 7 tokens
    • mean: 22.05 tokens
    • max: 82 tokens
  • Samples:
    positive anchor
    ["orra\n\nThe Andorra personal data protection act came into force on May 17, 2022, by the Andorra Data Protection Authority (ADPA). Learn more about Andorra PDPA\n\n### United Kingdom\n\nThe UK Data Protection Act (DPA) 2018 is the amended version of the Data Protection Act that was passed in 1998. The DPA 2018 implements the GDPR with several additions and restrictions. Learn more about UK DPA\n\n### Botswana\n\nThe Botswana Data Protection came into effect on October 15, 2021 after the issuance of the Data Protection Act (Commencement Date) Order 2021 by the Minister of Presidential Affairs, Governance and Public Administration. Learn more about Botswana DPA\n\n### Zambia\n\nOn March 31, 2021, the Zambian parliament formally passed the Data Protection Act No. 3 of 2021 and the Electronic Communications and Transactions Act No. 4 of 2021. Learn more about Zambia DPA\n\n### Jamaica\n\nOn November 30, 2020, the First Schedule of the Data Protection Act No. 7 of 2020 came into effect following the publication of Supplement No. 160 of Volume CXLIV in the Jamaica Gazette Supplement. Learn more about Jamaica DPA\n\n### Belarus\n\nThe Law on Personal Data Protection of May 7, 2021, No. 99-Z, entered into effect within Belarus on November 15, 2021. Learn more about Belarus DPA\n\n### Russian Federation\n\nThe primary Russian law on data protection, Federal Law No. 152-FZ has been in effect since July 2006. Learn more\n\n### Eswatini\n\nOn March 4, 2022, the Eswatini Communications Commission published the Data Protection Act No. 5 of 2022, simultaneously announcing its immediate enforcement. Learn more\n\n### Oman\n\nThe Royal Decree 6/2022 promulgating the Personal Data Protection Law (PDPL) was passed on February 9, 2022. Learn more\n\n### Sri Lanka\n\nSri Lanka's parliament formally passed the Personal Data Protection Act (PDPA), No. 9 Of 2022, on March 19, 2022. Learn more\n\n### Kuwait\n\nKuwait's DPPR was formally introduced by the CITRA to ensure the Gulf country's data privacy infrastructure. Learn more\n\n### Brunei Darussalam\n\nThe draft Personal Data Protection Order is Brunei’s primary data protection law which came into effect in 2022. Learn more\n\n### India\n\nIndia’"] What is the name of India's data protection law before May 17, 2022?
    [' the affected data subjects and regulatory authority about the breach and whether any of their information has been compromised as a result.\n\n### Data Protection Impact Assessment\n\nThere is no requirement for conducting data protection impact assessment under the PDPA.\n\n### Record of Processing Activities\n\nA data controller must keep and maintain a record of any privacy notice, data subject request, or any other information relating to personal data processed by him in the form and manner that may be determined by the regulatory authority.\n\n### Cross Border Data Transfer Requirements\n\nThe PDPA provides that personal data can be transferred out of Malaysia only when the recipient country is specified as adequate in the Official Gazette. The personal data of data subjects can not be disclosed without the consent of the data subject. The PDPA provides the following exceptions to the cross border data transfer requirements:\n\nWhere the consent of data subject is obtained for transfer; or\n\nWhere the transfer is necessary for the performance of contract between the parties;\n\nThe transfer is for the purpose of any legal proceedings or for the purpose of obtaining legal advice or for establishing, exercising or defending legal rights;\n\nThe data user has taken all reasonable precautions and exercised all due diligence to ensure that the personal data will not in that place be processed in any manner which, if that place is Malaysia, would be a contravention of this PDPA;\n\nThe transfer is necessary in order to protect the vital interests of the data subject; or\n\nThe transfer is necessary as being in the public interest in circumstances as determined by the Minister.\n\n## Data Subject Rights\n\nThe data subjects or the person whose data is being collected has certain rights under the PDPA. The most prominent rights can be categorized under the following:\n\n## Right to withdraw consent\n\nThe PDPA, like some of the other landmark data protection laws such as CPRA and GDPR gives data subjects the right to revoke their consent at any time by way of written notice from having their data collected processed.\n\n## Right to access and rectification\n\nAs per this right, anyone whose data has been collected has the right to request to review their personal data and have it updated. The onus is on the data handlers to respond to such a request as soon as possible while also making it easier for data subjects on how they can request access to their personal data.\n\n## Right to data portability\n\nData subjects have the right to request that their data be stored in a manner where it'] What is the requirement for conducting a data protection impact assessment under the PDPA?
    [" more\n\nPrivacy\n\nAutomate compliance with global privacy regulations\n\nData Mapping Automation\n\nView\n\nData Subject Request Automation\n\nView\n\nPeople Data Graph\n\nView\n\nAssessment Automation\n\nView\n\nCookie Consent\n\nView\n\nUniversal Consent\n\nView\n\nVendor Risk Assessment\n\nView\n\nBreach Management\n\nView\n\nPrivacy Policy Management\n\nView\n\nPrivacy Center\n\nView\n\nLearn more\n\nSecurity\n\nIdentify data risk and enable protection & control\n\nData Security Posture Management\n\nView\n\nData Access Intelligence & Governance\n\nView\n\nData Risk Management\n\nView\n\nData Breach Analysis\n\nView\n\nLearn more\n\nGovernance\n\nOptimize Data Governance with granular insights into your data\n\nData Catalog\n\nView\n\nData Lineage\n\nView\n\nData Quality\n\nView\n\nData Controls Orchestrator\n\nView\n\nSolutions\n\nTechnologies\n\nCovering you everywhere with 1000+ integrations across data systems.\n\nSnowflake\n\nView\n\nAWS\n\nView\n\nMicrosoft 365\n\nView\n\nSalesforce\n\nView\n\nWorkday\n\nView\n\nGCP\n\nView\n\nAzure\n\nView\n\nOracle\n\nView\n\nLearn more\n\nRegulations\n\nAutomate compliance with global privacy regulations.\n\nUS California CCPA\n\nView\n\nUS California CPRA\n\nView\n\nEuropean Union GDPR\n\nView\n\nThailand’s PDPA\n\nView\n\nChina PIPL\n\nView\n\nCanada PIPEDA\n\nView\n\nBrazil's LGPD\n\nView\n\n\+ More\n\nView\n\nLearn more\n\nRoles\n\nIdentify data risk and enable protection & control.\n\nPrivacy\n\nView\n\nSecurity\n\nView\n\nGovernance\n\nView\n\nMarketing\n\nView\n\nResources\n\nBlog\n\nRead through our articles written by industry experts\n\nCollateral\n\nProduct brochures, white papers, infographics, analyst reports and more.\n\nKnowledge Center\n\nLearn about the data privacy, security and governance landscape.\n\nSecuriti Education\n\nCourses and Certifications for data privacy, security and governance professionals.\n\nCompany\n\nAbout Us\n\nLearn all about"] What is Data Subject Request Automation?
  • Loss: MatryoshkaLoss with these parameters:
    {
    "loss": "MultipleNegativesRankingLoss",
    "matryoshka_dims": [
        512,
        256,
        128,
        64
    ],
    "matryoshka_weights": [
        1,
        1,
        1,
        1
    ],
    "n_dims_per_step": -1
}

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: epoch
  • per_device_train_batch_size: 32
  • per_device_eval_batch_size: 16
  • learning_rate: 2e-05
  • num_train_epochs: 5
  • lr_scheduler_type: cosine
  • warmup_ratio: 0.1
  • bf16: True
  • tf32: True
  • load_best_model_at_end: True
  • optim: adamw_torch_fused
  • batch_sampler: no_duplicates

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: epoch
  • prediction_loss_only: True
  • per_device_train_batch_size: 32
  • 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: 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: cosine
  • 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: True
  • 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: True
  • 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_fused
  • 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
  • batch_sampler: no_duplicates
  • multi_dataset_batch_sampler: proportional

Training Logs

Epoch Step Training Loss dim_128_cosine_map@100 dim_256_cosine_map@100 dim_512_cosine_map@100 dim_64_cosine_map@100
0.3448 10 7.9428 - - - -
0.6897 20 6.0138 - - - -
1.0 29 - 0.2011 0.2099 0.2307 0.1829
1.0345 30 5.4431 - - - -
1.3793 40 4.4675 - - - -
1.7241 50 3.7435 - - - -
2.0 58 - 0.2092 0.2161 0.2341 0.1983
2.0690 60 3.6676 - - - -
2.4138 70 3.0414 - - - -
2.7586 80 2.5451 - - - -
3.0 87 - 0.2091 0.2137 0.2426 0.1868
3.1034 90 2.7694 - - - -
3.4483 100 2.3624 - - - -
3.7931 110 2.1016 - - - -
4.0 116 - 0.2139 0.2137 0.2271 0.1964
4.1379 120 2.3842 - - - -
4.4828 130 1.9261 - - - -
4.8276 140 1.9737 - - - -
5.0 145 - 0.2117 0.2149 0.2318 0.1946
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.14
  • Sentence Transformers: 3.0.1
  • Transformers: 4.41.2
  • PyTorch: 2.1.2+cu121
  • Accelerate: 0.31.0
  • 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",
}

MatryoshkaLoss 

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss 

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}