Mickele-hub/cv-ai-embedding-model-test

SentenceTransformer based on sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2. It maps inputs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, classification, clustering, and more.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

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': 384, 'pooling_mode': 'mean', 'include_prompt': True})
)

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("sentence_transformers_model_id")
# Run inference
sentences = [
    "SummaryHighly motivated Sales Associate with extensive customer service and sales experience. Outgoing sales professional with track record of driving increased sales, improving buying experience and elevating company profile with target market.\nHighlightsLabVIEW, Virtuoso (Cadence) Design Suite, P LE CS, Cadence Schematic editor, AWR Microwave Office,\n   MATLAB/SIMULINK, Code Composer Studio, National Instruments PXI, Texas Instruments VLCT for IC Testing-\n   ATE\n   Programming Languages: C, C++, Python, Assembly and VHDL\nExperience06/2014toPresentElectrical Design Engineer|,,Working on developing next generation intelligent down-hole oil drilling tools.Working on automating Calibration process for Accelerometers and Magnetometers Working on developing automated test systems using LabVIEW and Python.toResearch AssistantRockwell Automation, Inc.|,,Worked on developing high accuracy GPS System and also on finding outliers in large collection of Cotton fiber cross-section scan data.toProduct Engineering InternDartmouth College|,,www.physitechelectronics.com) Developed Hardware and Software for various Major projects on embedded systems with variety of sensors.Developed Firmware for a Programmer to program AVR Microcontrollers through USB Port Served as the President of The Electronics Club, SNIST, Hyderabad for two years (2010-2012) Electronix Club is a student's organization founded in 2009 in Sreenidhi Institute of Science and Technology, Hyderabad, India by a group of electronics freaks to learn and share the concepts of electronics and prosper in the process Taught classes on basic Analog Circuits, Microcontroller Programming and Sensor interfaces during this period.\nEducationExpected inMay, 2014totoMaster of Science|Electrical EngineeringTexas Tech University,Lubbock,TXGPA:GPA: 3.9/4.0Electrical Engineering GPA: 3.9/4.0Expected inMay, 2012totoBachelor of Technology|Electronics and Communications EngineeringJawaharlal Nehru Technological University,,GPA:GPA: 3.7/4.0Electronics and Communications Engineering GPA: 3.7/4.0\nSkillsArchitect, basic, C, C++, CAD, Cadence, Calibration, com, Hardware, editor, Electronics, embedded systems, GPS, LabVIEW, MATLAB, Office, Microwave, Composer, Networks, next, oil, Programmer, Programming, prosper, Python, USB, VHDL",
    'At Monument, were building a vertical software platform for the self-storage industry. This is a rare opportunity to reinvent a $30 billion market from the ground up.As a Lead Engineer, you will work closely with engineering and product leadership to design and build the first version of Monuments software platform from scratch, using the most modern technologies available - no tech debt, no legacy systems, just a blank whiteboard. Your ResponsibilitiesDesign highly-scalable solutions for an enterprise application.Lead a team of engineers to deliver quality software, on time and on budget.Empower your teammates to tackle projects by assisting with technical planning, reviewing code, and promoting strong engineering practices. Participate in interviewing and hiring engineers.Troubleshoot and fix technical issues quickly. Who You Are5+ years of experience building web apps; 3+ years building B2B SaaS apps.Experience leading a technical implementation.Experience with JavaScript, React, Node.js, and Nest.js (or similar framework).Experience with one or more object-oriented programming languages.Experience with MySQL or other relational database management systems.Experience optimizing web apps for both performance and scalability. You have a deep understanding of and experience writing effective unit tests.You are passionate about building something from scratch.You thrive in a team setting and are excited to mentor other engineers - you see their successes as your own. You care as much about building great products as you do writing great code. Comp & BenefitsCompetitive salary and equity grantMedical, Dental, and VisionUnlimited PTO Remote Work PolicyMonument supports remote work for this position. We also have an office in Dallas, TX where team members are welcome to work.To ensure alignment and cohesion, Monument holds quarterly in-person all-hands gatherings in the Dallas, Texas area. All team members are expected to attend.',
    'Data EngineerLength of Assignment: 6+ monthsWork Location: Remote US (NJ preferred)\nSUMMARY OF ESSENTIAL JOB FUNCTIONS:Design and develop analytical models and be the face to the data consumers.Perform data curation to meet the business requirements.Build batch and streaming data pipelines.Develop processes for automating, testing, and deploying your work.Identify risks and opportunities of potential logic and data issues within the data environment.Collaborate effectively with the global team and ensure day to day deliverables are met.\nMINIMUN REUIREMENTS:MUST HAVE experience with GCPGoogle Cloud Platform andor Spark.Bachelors degree and 5+ years of industry experience in software development, data engineering, business intelligence, data science, or related field with a track record of manipulating, processing, and extracting value from large datasets.3+ years of experience as Data Engineer or in a similar roleProven experience with AWS andor GCP, Hadoop, Vertica, Talend, Tableau, and other modern technology platforms is required.Cloud to Cloud migration experience preferred.Have knowledge of data management fundamentals and data storage principlesHave knowledge of systems as it pertains to data storage and computing.Strong source to target mapping experience and ETL principlesknowledge.Excellent verbal and written communication skills.Strong quantitative and analytical skills with accuracy and attention to detailAbility to work well independently with minimal supervision and can manage multiple priorities.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000, -0.0941, -0.0238],
#         [-0.0941,  1.0000,  0.1630],
#         [-0.0238,  0.1630,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 100 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 100 samples:

  	sentence1	sentence2	score
  type	string	string	float
  modality	text	text
  details	min: 128 tokens mean: 128.0 tokens max: 128 tokens	min: 60 tokens mean: 126.54 tokens max: 128 tokens	min: 0.0 mean: 0.0 max: 0.0

• Samples:

  sentence1	sentence2	score
  SummaryHighly motivated Sales Associate with extensive customer service and sales experience. Outgoing sales professional with track record of driving increased sales, improving buying experience and elevating company profile with target market. Highlights-Soft Skills: Public Speaking, Public Relations, Team Building, Project Management, Procedure writing, Staff Supervision and Management, Ability to interface with professionals on all levels. Accomplishments, Honors, and Activities -Board of Directors Member for the Food Bank of Corpus Christi from November 2010 to April 2013. -Held Life Insurance License -Basketball Official (Referee) High School Varsity Level. ExperienceAccountant,08/2014-05/2015Aspirus–Owen,WI,Perform daily and routine accounting functions for two main companies and five small royalty companies. Responsibilities include but are not limited to the following: Accounts Payable, Accounts Receivable, Manage and reconcile funds for multiple banks accounts, Payroll, Perfo...	Net2Source Inc. is an award-winning total workforce solutions company recognized by Staffing Industry Analysts for our accelerated growth of 300% in the last 3 years with over 5500+ employees globally, with over 30+ locations in the US and global operations in 32 countries. We believe in providing staffing solutions to address the current talent gap Right Talent Right Time Right Place Right Price and acting as a Career Coach to our consultants. Role: Basel Business AnalystLocation: Washington, D.C.Work Mode: HybridHire Type: 6+ Month Contract (extendable) JD: Role Specific Experience: 6+ years of relevant technical and business work experience The Candidates who have worked on Basel related projects in Credit risks or at least are aware of credit risk. Banking & Financial domain experience, along with knowledge of risk management, familiarity with concepts of finance and accounting Proficiency in MS Excel Experience in defining and implementing technology solutions for a major...	0.0
  Professional SummaryCurrently working with Caterpillar as an contract employee Active in several NPI and CPI projects Provide technical for CAT facilities worldwide 24-hours a day Willing to travel to meet job requirements Familiar with Power Distribution Systems, NETA, NFPA 70E, and OHSA electrical standards Graduating in December 2014 with a Master's Degree in Industrial Management Excellent communication and interpersonal skills SkillsGuest servicesInventory control proceduresMerchandising expertiseLoss preventionCash register operationsProduct promotions Work History08/2012toCurrentFacilities Software Coordinator|City,STATE,Responsible for managing software across all products in Decatur IL, Joliet IL, and North Little Rock AR On-call 24 hours a day to provide support to field reps, assembly workers, and line supervisors around the world including India, Russia, China, and Brazil Traveled to various locations within the company including Russia, Mossville IL, East Peoria IL, and No...	At Salas OBrien we tell our clients that were engineered for impact. This passion for making a difference applies just as much to our team as it does to our projects. Thats why were committed to living our values every day: inspiring, achieving, and connecting as shared owners of our success with a focus on a sustainable future. Building for the long-term means that all of our team members can expect to work on amazing projects with a people-first approach to problem solving. It also means that each member of our team has truly limitless potential to build a unique, meaningful, and high-impact careerand theyll receive great total rewards along the way. About Us Founded in 1975, Salas OBrien is an employee-owned engineering and professional services firm focused on achieving impact for our clients, our team, and the world. We know that tomorrows requirements are todays opportunities, and we are here to design lasting solutions for pressing challenges. We work across a variety of industr...	0.0
  SummaryI started my construction career in June of 2017 in Jacksonville, Florida as an Electrical Apprentice working full time and attending the Jacksonville Electrical Training Alliance. this experience provided invaluable insight into electrical systems work and the related theory. Knowledgeable leader with record of accomplishment on diverse construction projects. Translates directives of architects, constructions managers and clients into actionable on-site directives to consistently meet project demands. Expert in tackling problems head-on and maintaining strict standards. Hardworking and reliable electrician apprentice with strong ability in electrical knowledge and mechanical. Offering leadership, reliability and accuracy. Highly organized, proactive and punctual with team-oriented mentality. SkillsDiagnostics and troubleshootingOSHA requirementsElectrical codes familiaritySafety awarenessPower and hand tool operationSchematics and blueprintsLift up to 20,000 poundsElectrical di...	Schweitzer Engineering Laboratories (SEL) Infrastructure Defense Division seeks a talented individual for a cybersecurity focused (Lead) Software Engineer position. The work we perform is some of the most fascinating, challenging, and rewarding that can be found in the power engineering community. Lead a team, conducting white and black box testing of devices and software for security flaws and vulnerabilities. Identify areas of concern and recommendations for an improved cybersecurity stance. Will require staying up to date and training others about current security vulnerabilities and mitigation processes. Note: United States Citizenship is required. As a Lead Software Engineer, a typical day might include the following: Lead the specification, design, implementation, and test of firmwaresoftware systems.Understand hardware and software interactions, and collaborate with Hardware Engineers.Lead the development of designs and test documentation.Mentor Engineers, Associate Engineers, a...	0.0

• Loss: CosineSimilarityLoss with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss",
      "cos_score_transformation": "torch.nn.modules.linear.Identity"
  }
  

Training Hyperparameters

All Hyperparameters

Click to expand

• per_device_train_batch_size: 8
• num_train_epochs: 3.0
• max_steps: -1
• learning_rate: 5e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0
• optim: adamw_torch_fused
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1.0
• label_smoothing_factor: 0.0
• bf16: False
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: False
• project: huggingface
• trackio_space_id: None
• trackio_bucket_id: None
• trackio_static_space_id: None
• per_device_eval_batch_size: 8
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: False
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: False
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• dataloader_multiprocessing_context: None
• dataloader_in_order: True
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_static_graph: None
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: None
• fsdp_config: None
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}
• warmup_ratio: None

Training Time

• Training: 1.7 minutes

Framework Versions

• Python: 3.13.14
• Sentence Transformers: 6.0.1
• Transformers: 5.17.0
• PyTorch: 2.14.0+cpu
• Accelerate: 1.15.0
• Datasets: 5.0.1
• Tokenizers: 0.23.2

Additional Resources

• Training and Finetuning Embedding Models with Sentence Transformers: the end-to-end guide for training or finetuning Sentence Transformer models.
• Introduction to Matryoshka Embedding Models: variable-size embeddings that can be truncated with minimal quality loss.
• Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval: post-training compression of embedding vectors.
• Multimodal Embedding & Reranker Models with Sentence Transformers: use text, image, audio, and video models through the same API.
• Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers: train multimodal embedding models, with a Visual Document Retrieval walkthrough.

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",
}