oumaimansir/cv-matcher-finetuned

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 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})
  (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 = [
    'Name: Suresh Patel || Email: suresh.patel94@gmail.com | Phone: +91-8821351429 | LinkedIn: linkedin.com/in/suresh-patel5 || Location: Delhi, India | Open to: Remote / Hybrid || --- || PROFESSIONAL SUMMARY || Data scientist with 3 years applying Hadoop, Kubeflow and Hive to solve complex business problems. Experience across recommendation systems, fraud detection, and demand forecasting. || --- || SKILLS || Python, Hadoop, Hive, SQL, Kubeflow, NLP, dbt, Tableau, XGBoost || --- || WORK EXPERIENCE || ML Engineer | InMobi | Delhi | 2022 – Present || • Designed demand forecasting pipeline using Python, reducing inventory waste by 78% and saving ₹35Cr || • Deployed NLP model using Tableau for 4M+ user queries with sub-14ms inference latency || • Developed fraud detection model using Tableau and NLP with 97.7% precision, preventing ₹21Cr in losses || --- || Junior ML Engineer | Swiggy | Delhi | 2020 – 2021 || • Led A/B testing framework used by 20+ product teams, increasing experiment velocity by 51% || • Mentored team of 44 junior data scientists and established ML best practices || --- || EDUCATION || M.S. in Machine Learning | VIT Mumbai | 2020 | CGPA: 9.3',
    'Job Title: Research Scientist || Company: TCS | Location: Jaipur / Remote | Type: Full-Time || Experience: 3+ years | Salary: ₹26–42 LPA || Team Size: ~24 | Interview Rounds: 6 || --- || ABOUT THE ROLE || We are looking for a passionate Research Scientist to join our Data Science team at TCS. You will work on high-impact problems affecting millions of users and collaborate with some of the best minds in the industry. || --- || RESPONSIBILITIES || • Translate complex model outputs into actionable business insights for non-technical stakeholders || • Work closely with data engineering to ensure data quality and feature availability || • Build end-to-end ml pipelines from data ingestion to model deployment and monitoring || • Develop and maintain data pipelines using statistics and pandas || • Design and run rigorous a/b and multi-armed bandit experiments || --- || REQUIRED SKILLS: Statistics, SQL, Pandas, Spark, Feature Engineering, Transformers || NICE TO HAVE: dbt, BERT, Looker || --- || PERKS: Health & dental insurance, Mental health support, Annual performance bonus, Flexible work hours || Domain: Data Science',
    'Job Title: Performance Marketer || Company: Morgan Stanley | Location: Delhi / Remote | Type: Full-Time || Experience: 1+ years | Salary: ₹20–30 LPA || Team Size: ~18 | Interview Rounds: 6 || --- || ABOUT THE ROLE || We are looking for a passionate Performance Marketer to join our Marketing team at Morgan Stanley. You will work on high-impact problems affecting millions of users and collaborate with some of the best minds in the industry. || --- || RESPONSIBILITIES || • Identify and test new growth channels through rapid experimentation || • Develop and execute email marketing campaigns with segmentation and personalisation || • Work with creative and brand teams to produce compelling campaign assets || • Analyse campaign performance using hubspot and present insights to leadership || • Manage performance marketing budgets and optimise for cac, roas, and ltv || --- || REQUIRED SKILLS: HubSpot, Content Marketing, Video Marketing, LinkedIn Ads, PPC, SEO, Social Media || NICE TO HAVE: Salesforce, Meta Ads, Brand Strategy || --- || PERKS: Remote-first culture, Gym membership, Health & dental insurance, Quarterly offsites || Domain: Marketing',
]
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, 0.6883, 0.4044],
#         [0.6883, 1.0000, 0.3296],
#         [0.4044, 0.3296, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 500 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: 256 tokens mean: 256.0 tokens max: 256 tokens	min: 256 tokens mean: 256.0 tokens max: 256 tokens	min: 0.05 mean: 0.49 max: 0.98

• Samples:

  sentence_0	sentence_1	label
  Name: Rohan Iyer || Email: rohan.iyer44@gmail.com | Phone: +91-7672100521 | LinkedIn: linkedin.com/in/rohan-iyer9 || Location: Kolkata, India | Open to: Remote / Hybrid || --- || PROFESSIONAL SUMMARY || Security engineer with 11 years building and operating SOC capabilities at HDFC Bank. Expert in Workday, Onboarding and threat hunting. Reduced MTTD by 58%. || --- || SKILLS || Onboarding, Workforce Planning, Change Management, Conflict Resolution, Organisational Development, Workday, SuccessFactors || --- || WORK EXPERIENCE || CISO | HDFC Bank | Kolkata | 2018 – Present || • Led incident response for 33+ security events, containing breaches with zero data exfiltration || • Designed and deployed cloud-native SIEM using Change Management, processing 40B+ log events daily || • Implemented Zero Trust architecture across 3K endpoints, reducing lateral movement risk by 56% || --- || Junior CISO | TCS | Kolkata | 2016 – 2018 || • Reduced mean time to detect (MTTD) from 89 hours to 18 hours by...	Job Title: Head of HR || Company: Deloitte | Location: Hyderabad / Remote | Type: Full-Time || Experience: 4+ years | Salary: ₹29–73 LPA || Team Size: ~6 | Interview Rounds: 4 || --- || ABOUT THE ROLE || We are looking for a passionate Head of HR to join our Human Resources team at Deloitte. You will work on high-impact problems affecting millions of users and collaborate with some of the best minds in the industry. || --- || RESPONSIBILITIES || • Partner with business unit heads to identify people needs and build workforce plans || • Resolve employee grievances, conduct investigations, and ensure fair outcomes || • Champion diversity, equity, and inclusion initiatives across the organisation || • Own end-to-end recruitment for technical and business roles across all levels || • Develop and implement l&d programmes aligned with business capability gaps || --- || REQUIRED SKILLS: HRIS, HRMS, LinkedIn Recruiter, Training & Development, Payroll, Change Management, Onboarding, Succession P...	0.33
  Name: Arjun Kulkarni || Email: arjun.kulkarni59@gmail.com | Phone: +91-7205273021 | LinkedIn: linkedin.com/in/arjun-kulkarni9 || Location: Noida, India | Open to: Remote / Hybrid || --- || PROFESSIONAL SUMMARY || Talent acquisition specialist with 12 years. Hired 50+ candidates across tech, product, and business functions. Passionate about Employer Branding and data-driven recruiting. || --- || SKILLS || Onboarding, Employee Relations, Workforce Planning, HR Analytics, Change Management, Exit Management, LinkedIn Recruiter, Performance Management, SAP HR, Employer Branding, Payroll || --- || WORK EXPERIENCE || Recruiter | Mphasis | Noida | 2016 – Present || • Managed full-cycle recruitment for 33+ roles in 12 months to support rapid expansion phase || • Ensured compliance across 4 states with varying labour laws during company acquisition || • Reduced annual attrition from 96% to 8% through stay interviews, compensation benchmarking, and career pathing || --- || Junior Recruiter | Orac...	Job Title: People Ops Manager || Company: Airbnb | Location: Bangalore / Remote | Type: Full-Time || Experience: 4+ years | Salary: ₹13–33 LPA || Team Size: ~12 | Interview Rounds: 3 || --- || ABOUT THE ROLE || We are looking for a passionate People Ops Manager to join our Human Resources team at Airbnb. You will work on high-impact problems affecting millions of users and collaborate with some of the best minds in the industry. || --- || RESPONSIBILITIES || • Develop and implement l&d programmes aligned with business capability gaps || • Use hr analytics to identify trends in attrition, engagement, and hiring efficiency || • Own end-to-end recruitment for technical and business roles across all levels || • Resolve employee grievances, conduct investigations, and ensure fair outcomes || • Administer payroll, benefits, and statutory compliance across locations || --- || REQUIRED SKILLS: Performance Management, Conflict Resolution, LinkedIn Recruiter, Training & Development, Employee Rel...	0.77
  Name: Gaurav Malhotra || Email: gaurav.malhotra26@gmail.com | Phone: +91-8148872110 | LinkedIn: linkedin.com/in/gaurav-malhotra1 || Location: Gurgaon, India | Open to: Remote / Hybrid || --- || PROFESSIONAL SUMMARY || 3+ years as a ML Engineer specialising in Power BI and Go. Built production ML systems at Google serving 38M+ users. Published 5 internal research papers. || --- || SKILLS || Pandas, Hypothesis Testing, XGBoost, Python, Power BI, NumPy, EKS, Go, ArgoCD || --- || WORK EXPERIENCE || ML Engineer | Google | Gurgaon | 2020 – Present || • Built Python-based recommendation engine that increased click-through rate by 19% and revenue by ₹37Cr annually || • Led A/B testing framework used by 21+ product teams, increasing experiment velocity by 46% || • Reduced model training time by 15% by migrating Hypothesis Testing workloads to distributed Python cluster || --- || Junior ML Engineer | ShareChat | Gurgaon | 2017 – 2018 || • Mentored team of 7 junior data scientists and established...	Job Title: Head of DevOps || Company: Adobe | Location: Chennai / Remote | Type: Full-Time || Experience: 7+ years | Salary: ₹24–51 LPA || Team Size: ~39 | Interview Rounds: 5 || --- || ABOUT THE ROLE || We are looking for a passionate Head of DevOps to join our DevOps team at Adobe. You will work on high-impact problems affecting millions of users and collaborate with some of the best minds in the industry. || --- || RESPONSIBILITIES || • Optimise cloud costs through rightsizing, reserved capacity, and architectural improvements || • Participate in on-call rotation, respond to incidents, and drive rca to prevent recurrence || • Collaborate with development teams to embed devops practices and platform tooling || • Implement comprehensive observability including metrics, logs, and distributed tracing || • Define and enforce slos, slis, and error budgets for critical services || --- || REQUIRED SKILLS: Linux, Bash, Consul, Pulumi, Kubernetes, Grafana || NICE TO HAVE: Go, New Relic, FluxC...	0.68

• 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
• 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: 3
• 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: 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: 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 Time

• Training: 19.4 minutes

Framework Versions

• Python: 3.14.5
• Sentence Transformers: 5.6.0
• Transformers: 5.12.1
• PyTorch: 2.12.1+cpu
• Accelerate: 1.14.0
• Datasets: 5.0.0
• 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",
}