Maaz2002/ats-finetuned-model

SentenceTransformer based on sentence-transformers/all-mpnet-base-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-mpnet-base-v2. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/all-mpnet-base-v2
• Maximum Sequence Length: 384 tokens
• Output Dimensionality: 768 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': 'MPNetModel'})
  (1): Pooling({'embedding_dimension': 768, 'pooling_mode': 'mean', '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("sentence_transformers_model_id")
# Run inference
sentences = [
    'Career OverviewHighly capable, versatile, and accomplished IT professional with more than 20 years of experience in planning and implementing all aspects of a Linux UNIX-based network. Demonstrated ability in network design and configuration as well as user-oriented, security conscious history expertise in problem solving with outstanding research skills, able to work efficiently and deliver results regardless of previous experience. Possess well-honed leadership and interpersonal skills, experienced in supervising various IT professionals, engineers, and Ph.D. holders as well as experts in other technical fields. Reliable team player, capable of working independently and prioritizing tasks, utilizing flexible approaches. QualificationsLinux System Administration (RedHat, CentOS, Ubuntu) UNIX (Dec, Sun, HP, and SGI variants) System Administration Cisco SAN Switch Configuration (MDS9000) Network Hardware Configuration (Bay Networks, Cisco, Linksys) iSCSI Legato Networker EMC Navisphere Platform LSF C and FORTRAN Programming Shell, Python and Perl scripting Linux Kernel Modification Security Software including Nmap, TCP Wrappers Email Software including SendMail and Postfix and Mailing List Software Configuration Monitoring Software such as Nagios LDAP NFS NIS Education and TrainingDuke UniversityDurham,NCExpected in Ph.D Elementary Particle Physics-GPA Elementary Particle PhysicsDuke UniversityDurham,NCExpected in A.M Physics-GPA PhysicsUniversity of North CarolinaChapel Hill,NCExpected in B.S Physics-GPA Physics Work ExperienceScientific Atlanta Cisco Systems ST Microelectronics-IT Systems Principle Engineer Senior Software Engineer Senior Staff Development Support SpecialistCity,STATE,01 1999-PresentEvaluate hardware and software needs, research options, recommend best fit and implement upgrades and changes to network and server hardware.Develop plan for growth of network hardware, including servers and other infrastructure.Serve as the sole administrator, providing oversight and direction in all aspects of a network.Recommend and facilitate the move from Solaris to Linux systems, implementing Linux test hosts and eventual move to Linux-based hosts.Automate steps in ASIC design process through creation of python and perl scripts.Manage all aspects of host security for network.Key Accomplishments Planned growth, completed installation, maintenance and configuration of both hardware and software as storage capacity grew from 60GB to 20TB.Investigated, planned, organized and executed upgrades to network infrastructure to move to gigabit Ethernet and transferred network information storage from NIS to LDAP.Achieved approximately 99.9 uptime, researching and implementing changes to ensure network availability, speed, and reliability Wrote Perl scripts to implement network changes as well as configured and maintained LSF workload management system.Handled the monitoring and correction of bottlenecks in NFS access speed of data assisted in the hiring of IT personnel.Provided helpdesk services for users of the Linux network Configured and maintained a Flexlm license server for ASIC tools that was used by engineers company-wide.CSC OAO Dyntel Corporation-EPA Contractor Team MemberCity,STATE,01 1995-01 1999Supervised system administrators accountable for all Unix systems (mix of Solaris, Irix, HP-UX, Ultrix, and Tru64 Unix), Windows servers, as well as the design and implementation of the campus network.Provided guidance to all Windows, UNIX, and Network personnel, directing and monitoring the proper maintenance, optimization, and troubleshooting on Windows domain controllers, file servers, and print servers as well as UNIX servers and workstations.Key Accomplishments Rendered supervision to the group that facilitated smooth e-mail transition, network infrastructure transition and server administration which received commendation from customers and the national office, leading companies with contract to consider Athens as one of their top sites.Successfully administered security measures for the site, becoming the first site to detect an outside attack among all EPA systems.Utilized low cost solutions (including freeware where appropriate), gathered through research to address network needs. SkillsASIC, Automate, Bay Networks, C, Cisco, hardware, Dec, directing, direction, e-mail, Email, Ethernet, file servers, FORTRAN, HP, hiring, HP-UX, Irix, LDAP, Legato, Linux, Mailing, access, office, Windows, NFS, network hardware, Network, NIS, optimization, Perl scripting, Perl, personnel, print servers, design process, Programming, Python, RedHat, researching, research, SAN, SendMail, servers, SGI, Shell, scripts, Solaris, Sun, supervision, Switch, System Administration, troubleshooting, Ultrix, UNIX, UNIX servers, upgrades ',
    ' Calling all innovators find your future at Fiserv. Were Fiserv, a global leader in Fintech and payments, and we move money and information in a way that moves the world. We connect financial institutions, corporations, merchants, and consumers to one another millions of times a day quickly, reliably, and securely. Any time you swipe your credit card, pay through a mobile app, or withdraw money from the bank, were involved. If you want to make an impact on a global scale, come make a difference at Fiserv. What does a successful Software Engineering Manager do at Fiserv?Fiserv is looking for a Software Engineering Manager to join our Core Acquiring Front End engineering group within Fiservs Global Business Solutions (Payment Acceptance) business.You will be joining an organization responsible for mission critical platforms that support our payments acquiring business. We process over 350 million payment transactions per day with a peak throughput of over 8,000 transactions per second all while maintaining millisecond fast response time. All this is done for some of the largest merchants in the US.This fast-paced online transaction processing environment requires a passionate critical problem solver who can balance platform stability and resiliency along with modern engineering practices and solutions. What you will do Manage cross-functional engineering teamsHands-on engineer pickup stories, code reviews, and support production End-to end platform owner and engineer design, build, and support resilient services solutionsPartner with business owners to define platform roadmaps and scope for initiatives featuresDrive prioritization of technical investment on in-scope platformsDesign performant, efficient, effective, and reusable technical solutions (partnering with architecture)Provide thought leadership across acquiring enginesPartner effectively with DevOps, Production Support, and peer platform teams What you will need to have 15 years of experience in software engineering roles using C and Java10 years of experience leading high-performing software engineering teams5 years of experience in financial services or payments processingAbility to clearly communicate with executive leadership What would be great to have Bachelor s degree in computer science, engineering, or equivalent work experienceExperience in AIXExperience building solutions in public cloud (AWS, Azure, GCP) Perks at work Were FiservProud of our commitment to your overall well-being with a growing offering of physical, mental, emotional, and financial benefits from day one.Maintain a healthy work-life balance with paid holidays, generous time off policies, including Unlimited Recharge Refuel for qualifying associates, and free counseling through our EAP.Plan for your future with competitive salaries, the Fiserv 401(k) Savings Plan, and our Employee Stock Purchase Plan.Recognize and be recognized by colleagues with our Living Proof program where you can exchange points for a variety of rewards.Prioritize your health with a variety of medical, dental, vision, life, and disability insurance options and a range of well-being resources through our Fuel Your Life program.Advance your career with training, development, certification, and internal mobility opportunities.Join Employee Resource Groups that promote our diverse and inclusive culture where associates can share perspectives, exchange ideas, and elevate careers. What you should know about us Fiserv is a global fintech leader with 40,000-plus (and growing) associates proudly serving clients in more than 100 countries. As a FORTUNE 500 company, one of Fast Companys Most Innovative Companies, and a top scorer on Bloombergs Gender-Equality Index, we are committed to excellence and purposeful innovation. Our commitment to Diversity and Inclusion Fiserv is an Equal Opportunity Employer, and we welcome and encourage diversity in our workforce that reflects our world. All qualified applicants will receive consideration for employment without regard to race, color, religion, sexual orientation, gender identity, national origin, disability, protected veteran status, or any other category protected by law. We will ensure that individuals with disabilities are provided reasonable accommodation to participate in the job application or interview process, to perform essential job functions, and to receive other benefits and privileges of employment. Please contact us to request accommodation. Warning about fake job posts Please be aware of fraudulent job postings that are not affiliated with Fiserv. Fraudulent job postings may be used by cyber criminals to target your personally identifiable information andor to steal money or financial information. Any communications from a Fiserv representative will come from a legitimate business email address. We will not hire through text message, social media, or email alone, and any interviews will be conducted in person or through a secure video call. We wont ask you for sensitive information nor will we ask you to pay anything during the hiring process. We also wont send you a check to cash on Fiservs behalf. If you see suspicious activity or believe that you have been the victim of a job posting scam, you should report it to your local FBI field office or to the FBIs Internet Crime Complaint Center.',
    ' The Senior Manager, Data Architecture Data Engineering is responsible for translating business strategic goals into future-state data engineering and data science solutions. This includes delivery full life-cycle from execution roadmaps, to implementation, and to production operations. This role drives the creation and maintenance of enterprise data and analytics capabilities, data reference architectures, and integrated data analytics technology roadmaps that will guide our strategic investments and enterprise strategy.Translate business strategic goals into actionable future-state data engineering and data science architectures and solutions.Delivery of full life-cycle Data Solutions from execution roadmaps, to implementation, and to operations. This includes a specific focus on ELT and ETL Data Pipeline solutions from inception to production.Lead end to end solution delivery of enterprise data warehouse, data marts, and data lake with governance.Drive the creation and maintenance of enterprise data and analytics capabilities, data solutions, and integrated data analytics technology roadmaps that will guide our strategic investments and enterprise strategy.Deliver architecture connecting Data Science, Data Engineering, and Data Analytics practices together to deliver impactful data products.Strong partnership with DevOps, DataOps, business partners, engineering teams, and data management.A minimum of ten (10) years of experience in IT with a minimum (6) years in data architecture and data solutions are required.Successful track-record in managing a team of Data Engineers and Data ScientistsSkilled in hybrid cloud data platforms and analytical technologies including Azure andor GCP data infrastructure, Spark, Data Lake, Data Warehouse and Data Marts, BI reporting with PowerBI, Metadata Management, Data Governance, Master Data Management, Data Catalog, and Business Glossary.Ability to document, communicate, implement, and operationalize complex architectures.Experience with ELT and ETL Data Pipeline solutions from inception to production.Proficient in SQL, NoSQL and columnar databases.Experience with SparkPyspark and JSON.Excellent grasp of Computer Science fundamentals - data structures and algorithms.Must have Panda Data Frames knowledge and best practices, execution engines, DAGs (airflow, dagster, luigi, etc), and Kubernetes.Proficient with GitLabFamiliarity with Azure data factory preferred, Google data flow or AWS GlueExperience with Biomedical Informatics, S3Azure blob store, Machine Learning and AI, Unstructured Data, and Data Democratization is preferred.Experience with Agile methodologies, Kanban preferred, and using tools like Jira and Confluence to facilitate the work.Bachelor s degree in computer science or a related discipline required. Master s degree preferred. RECRUITERS NOTES The role will definitely be hands-on and will require excellent communication skills. The hiring manager mentioned the following technologies AImachine learning data warehouses product support Cloud experienceAzure must be open to new technologies. Working with the Product Development Team. Act as the point of contact will create a group that helps the product teams. Create data warehouses for analytics. MUST have excellent communication skills verbal and written. This person will envision strategy and must be able to effectively communicate that strategy to others. The organization has asked me not to advertise the compensation. If you contact me I would be more than willing to share it with you. For over 50 years this non-profit organization has made it their mission to protect patients from unsafe and ineffective medical technologies and practices. More than 5,000 healthcare institutions and systems worldwide including four out of every five U.S. hospitals rely on the services of this organization to guide their operational and strategic decisions. In addition, they serve public and private payers, federal and state agencies, policymakers, ministries of health, associations, and accrediting agencies. Benefit from a healthy work-life balance while staying on the leading edge of technology and thriving in an innovative startup-like culture minus the risk. Sleep well knowing you are helping achieve a world where safe, high-quality healthcare is accessible to everyone. Excellent benefits to include work from home, flex-time and 19 days starting vacation. Its a nice place to work!',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.6981, 0.4169],
#         [0.6981, 1.0000, 0.5805],
#         [0.4169, 0.5805, 1.0000]])

Evaluation

Metrics

Semantic Similarity

• Dataset: ats-val
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8525
spearman_cosine	0.7721

Training Details

Training Dataset

Unnamed Dataset

• Size: 4,461 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 100 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  modality	text	text
  details	min: 312 tokens mean: 382.82 tokens max: 384 tokens	min: 16 tokens mean: 309.28 tokens max: 384 tokens	min: 0.2 mean: 0.49 max: 0.9

• Samples:

  sentence_0	sentence_1	label
  Career OverviewHighly 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. QualificationsDatabases File SystemsOracle , MS AccessNetezzaHadoop Big DataAmazon Redshift , S3Software ProductsOracle D2K, Forms Reports 10gOracle Apps 11gHadoop Impala Hive SparkAmazon Web services, Redshift Programming LanguagesOracle PL SQL SQL, Impala SQLPythonJava(Basics), CVisual Basic, Coding in MS Excel using VBA, Power QueryUnix VMSSoftware ToolsPL SQL Developer, TOAD, DB Visualizer, NPS Admin tool Aginity for Netezza RedshiftForm Report BuilderElastic Search, Logstash and KibanaCloudwatch Exports to excelEclipse, PyCharmMS Office, MS Project. MS Visio, Power Query with ExcelPutty, Attachmate Reflection,HPSD, Remedy, Service PointNSI, VSS, IBM Rational Clear Case Clear Quest, Tortoise SVN, GitHubHP Quality Center , JIRAHadoop...	Job Description I am actively seeking an experienced Data Engineer to join an innovative SaaS company within the AI AdTech space. The ideal candidate is passionate about data, is highly self-motivated, enjoys collaborative work environments, and has a seasoned background in software development. Requirements At least 8 years of professional engineering experience At least 5 years of hands-on data engineering experience with ScalaSpark At least 3 years of hands-on software development experience with Java Spring for REST API development Highly proficient in designing, building, deploying, etc. new ETL data pipelines, databases, features and systems Expert level Spark knowledge Experience with Relational databases Cloud technology experience (AWS is highly preferred) B.S. or higher in Computer Science, or related field Bonus Points Background working in a SaaS product based environment Experience andor exposure to the full software development life cycle (SDLC) of a product Lead andor m...	0.871
  Professional ProfileExpert in Functional Testing .6-Year Record of Proven ResultsSenior software QA testerwith full system development life cycle experience, including designing, developing and implementing test plans, high level calendar test cases and test processes with attention to detail resulting in high quality Change Request and production change implementationsHands-on technology professionalaccustomed to working in complex, project-based environments. Multifaceted experience in QA software testing, system testing, user-acceptance testing and production certification testing.Backed by strong credentialsincluding a Bachelor s degree, certificate in business management and advanced command of various testing tools, agile methodologies and cross-platform skills in Windows, and Unix. Summary of SkillsTest Plans, Cases ProcessesFunctional RequirementsDocumentationScrum Agile MethodologiesUI Compatibility TestingTest strategiesDefect Bug TrackingUAT and System TestingTesting Automat...	Lead Software Developer New York City - Hybrid 150,000 - 190,000 One of our key clients is building a platform for small businesses to connect, transact, and grow together. A simple, easy-to-use interface that will serve as a contemporary solution for businesses payments needs. They are looking for a strong engineering lead to lead their full stack team. Someone who can develop, test, and maintain software applications and system to meet security, scalability, and performance requirements, and to ensure that their services run reliably and efficiently at scale with their background across the stack. What they re looking for Degree in Computer Science or equivalent8 years of experience in software engineering with strong Fintech background, Payments experience is a plusExperience working cross-functionality across multiple teams to deliver high-quality products.Experience with Java, AWS, SQL. Benefits Competitive salaryBonus potential Equity stock optionHealthcarevisiondental 100 cover...	0.586
  Career OverviewHighly 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. QualificationsDevelopment Using Apache Tomcat Server, Web Services, MVC pattern and OOP. Programming Languages C , Java, JavaScript, CSS3 and HTML5 Work ExperienceSoftware Development Intern,11 2016-PresentKion Group Indianapolis,IN,Design, develop and test Android and iOS applications.Involving with all phases of mobile app development from initial concept, through design, development, testing and deployment.Work with product managers to clarify function flows and provide technical insight on optimization.Work with the entire project team to balance creative objectives with business and technical requirements.Data Center Operations Associate,01 2016-PresentBoulder County, Co Boulder,CO,Develop internal application to process tracking d...	Job Requisition ID 23WD71706 Position Overview As a global leader in 3D design, engineering, and entertainment software, Autodesk helps people imagine, design, and create a better world. Autodesk accelerates better design through an unparalleled depth of experience and a broad portfolio of software to give customers the power to solve their design, business, and environmental challenges. In addition to designers, architects, engineers, and media and entertainment professionals, Autodesk helps students, educators, and casual creators unlock their creative ideas through user-friendly applications. As a DevOps Engineer on our Platform Engineering Team, you will work with team to build new capabilities to the Platform framework. You will have the opportunity to collaborate with frontend and backend engineers, QAQE, Architects, product owners, web producers and other key stakeholders to shape conceptual ideas into the elegantly engineered framework. This is your chance to work with a team ...	0.5

• Loss: CosineSimilarityLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• num_train_epochs: 10
• fp16: True
• per_device_eval_batch_size: 16
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• per_device_train_batch_size: 16
• num_train_epochs: 10
• 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
• label_smoothing_factor: 0.0
• bf16: False
• fp16: True
• 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: 16
• 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
• 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
• warmup_ratio: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	ats-val_spearman_cosine
1.0	279	-	0.5856
1.7921	500	0.0476	-
2.0	558	-	0.6475
3.0	837	-	0.6801
3.5842	1000	0.0266	-
4.0	1116	-	0.7240
5.0	1395	-	0.7350
5.3763	1500	0.0166	-
6.0	1674	-	0.7540
7.0	1953	-	0.7693
7.1685	2000	0.0116	-
8.0	2232	-	0.7702
8.9606	2500	0.0084	-
9.0	2511	-	0.7721

Training Time

• Training: 47.8 minutes

Framework Versions

• Python: 3.12.13
• Sentence Transformers: 5.6.0
• Transformers: 5.12.1
• PyTorch: 2.11.0+cu128
• Accelerate: 1.14.0
• Datasets: 4.0.0
• Tokenizers: 0.22.2

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",
}