metadata

base_model: saraleivam/GURU2-paraphrase-multilingual-MiniLM-L12-v2
datasets: []
language: []
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:503
  - loss:SoftmaxLoss
widget:
  - source_sentence: >-
      #La posición de Ingeniero QA Manual deberá:Priorización de la ejecución de
      pruebas.Experiencia en la ejecución de pruebas manuales y en la
      documentación de resultados.Proponer la estrategia de automatización de
      pruebas y las mejoras a los procesos de automatización.Manejo de
      plataformas como Atlassian Jira, Atlassian Confluence y
      GitLab.Conocimiento en motores de bases de datos y lenguajes de
      programación como .NET, Java, PHP y Python.
    sentences:
      - >-
        Streamlined Project Management with Trello: AI
        Integration.Business.Business Essentials.Be able to Initialize and
        Structure a Go-To-Market Plan Using Trello and AI Tools. Be able to
        Generate and Organize GTM Strategies Using ChatGPT and Trello. Be able
        to Enhance GTM Plan Content and Workflow Efficiency with Trello’s AI
        Tools
      - >-
        Microsoft Power BI Data Analyst.Information Technology.Security.Learn to
        use Power BI to connect to data sources and transform them into
        meaningful insights.  . Prepare Excel data for analysis in Power BI
        using the most common formulas and functions in a worksheet.     . Learn
        to use the visualization and report capabilities of Power BI to create
        compelling reports and dashboards.  . Demonstrate your new skills with a
        capstone project and prepare for the industry-recognized Microsoft
        PL-300 Certification exam.   
      - >-
        Getting Started in GIMP.Computer Science.Design and Product.Become
        Familiar with Using GIMP and GIMPs User Interface. Use Crop and Text
        Tools. Basic Color Correction Techniques
  - source_sentence: >-
      #La posición de Gerente Jurídico deberá:Apoyar al negocio en la
      elaboración y revisión y/o negociación de contratos, así como brindar la
      asesoría y soporte legal en las etapas precontractuales y
      postcontractuales.Asesorar a la Gerencia General en diversos temas,
      incluyendo integraciones empresariales.Participar activamente en la toma
      de decisiones estratégicas para la operación local y en diversos temas en
      los países a cargo.Asesorar asuntos legales de propiedad intelectual tales
      como, registros de marcas, patentes y permisos para uso de sustancias y
      equipos.Fungir como Representante Legal, Apoderado General y Oficial de
      Cumplimiento de todas las compañías en los países a cargo.Representación
      de los accionistas extranjeros en juntas y asambleas y/o llevar la
      secretaría corporativa de la sociedad.Revisar estrategia de respuesta y
      defensa jurídica en litigios, derechos de petición, tutelas y otras
      demandas, y asignación de firmas jurídicas externas.Atender y soportar
      legalmente en las adquisiciones de bienes y servicios de las
      compañías.Garantizar soporte legal de requerimientos de clientes internos
      y externos.
    sentences:
      - >-
        Data Science.Data Science.Data Analysis.Use R to clean, analyze, and
        visualize data.. Navigate the entire data science pipeline from data
        acquisition to publication. . Use GitHub to manage data science
        projects.. Perform regression analysis, least squares and inference
        using regression models.
      - >-
        Fundraising and Development Foundations.Business.Business
        Strategy.Communication, Leadership and Management, Organizational
        Development, Planning, Strategy, Strategy and Operations, Decision
        Making, People Development, Prospecting and Qualification, Business
        Development, Writing
      - >-
        Interfacing with the Arduino.Physical Science and Engineering.Electrical
        Engineering.Internet Of Things, Computer Programming
  - source_sentence: >-
      #Reportando a Sales Manager ColombiaLograr un crecimiento sostenible de
      los ingresos mediante la negociación, cierre, implementación y
      cumplimiento de acuerdos con cuentas clave (usuario final y
      revendedor).Negociar importantes oportunidades rentables tanto con cuentas
      clave nuevas como existentes. Experiencia manejando varios países a nivel
      Latam.Gestionar una cartera de cuentas clave con ingresos totales de 2 a 3
      millones de dólares.
    sentences:
      - >-
        The Blues: Understanding and Performing an American Art Form.Arts and
        Humanities.Music and Art.Students will be able to describe the blues as
        an important musical form. . Students will be able to explain
        differences in jazz and other variations of the blues. 
      - >-
        Managing Employee Performance.Business.Business Essentials.Employee
        Relations, Human Resources, Leadership and Management, Organizational
        Development, People Development, People Management, Performance
        Management, Professional Development, Strategy and Operations, Training,
        Culture
      - >-
        Marketing on TikTok.Business.Marketing.Social Media, Marketing, Media
        Strategy & Planning, Strategy, Digital Marketing
  - source_sentence: >-
      #Reportando a la Dirección Comercial. Esta persona sera la responsable de
      velar por la gestión de las cuentas ya activas y consecución de nuevos
      clientes en el sector financiero. Su rol principal sera tomar la base
      instalada de clientes e incrementar el consumo de soluciones que
      desarrolla la compañía, contribuyendo de esta manera al incremento de
      ventas para la organización
    sentences:
      - >-
        IBM Full Stack Software Developer.Information Technology.Cloud
        Computing.Master the most up-to-date practical skills and tools that
        full stack developers use in their daily roles. Learn how to deploy and
        scale applications using Cloud Native methodologies and tools such as
        Containers, Kubernetes, Microservices, and Serverless. Develop software
        with front-end development languages and tools such as HTML, CSS,
        JavaScript, React, and Bootstrap. Build your GitHub portfolio by
        applying your skills to multiple labs and hands-on projects, including a
        capstone
      - >-
        Strategising: Management for Global Competitive
        Advantage.Business.Business Strategy.Analyse how technology and
        innovation can disrupt and reshape your organisation. Evaluate the
        different ways supply chains can effectively meet your customer's
        demands  . Understand the different  strategies that your organisation
        can implement in order to remain competitive. Develop your understanding
        of how organisations can make positive contributions to society while
        effectively maintaining their bottom line
      - >-
        Scripting with Python and SQL for Data Engineering.Data Science.Data
        Analysis.Extract data from different sources and map it to Python data
        structures.. Design Scripts to connect and query a SQL database from
        within Python.. Apply scraping techniques to read and extract data from
        a website.
  - source_sentence: >-
      la posición de Ejecutivo Comercial Ingeniero Agrónomo Zootecnista deberá:*
      Hacer la apertura de mercado en la zona de Caldas.* Hacer las visitas
      comerciales a los diferentes clientes.
    sentences:
      - >-
        IBM Full-Stack JavaScript Developer.Computer Science.Software
        Development.Master the full-stack development languages, frameworks,
        tools, and technologies to develop job-ready skills valued by
        employers.. Write, deploy, and scale cloud-native back-end applications
        using Node, NoSQL databases, containers, microservices, and serverless..
        Develop websites and front-end software using HTML, CSS, JavaScript, and
        React.. Employ DevOps practices and Agile methodologies to continuously
        build and deploy software using CI/CD tools.
      - >-
        Data Science with Databricks for Data Analysts.Data Science.Data
        Analysis.Discover how Databricks and Apache Spark simplify big data
        processing and optimize data analysis. . Frame business problems for
        data science and machine learning to make the most out of big data
        analytic workflows.. Solve real-world business problems quickly using
        Databricks to power the most popular data science techniques. 
      - >-
        Power System: Generation, Transmission and Protection.Physical Science
        and Engineering.Electrical Engineering.Overview of generators and
        auxiliary system, electrical aspects in a thermal power plant (balance
        of plants) and related power plant control system. Indian grid scenario,
        transmission line parameters with real case study, modelling of
        transmission line parameters using MATLAB. Modern Trends in Electrical
        Design of EHV, modelling of transmission lines, mechanical design of AC
        transmission line. Protection system from generation, transmission to
        distribution including switchgear practical aspects and Gas Insulated
        Substations

SentenceTransformer based on saraleivam/GURU2-paraphrase-multilingual-MiniLM-L12-v2

This is a sentence-transformers model finetuned from saraleivam/GURU2-paraphrase-multilingual-MiniLM-L12-v2. It maps sentences & paragraphs to a 384-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: saraleivam/GURU2-paraphrase-multilingual-MiniLM-L12-v2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 384 tokens
Similarity Function: Cosine Similarity

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, '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("saraleivam/GURU3-paraphrase-multilingual-MiniLM-L12-v2")
# Run inference
sentences = [
    'la posición de Ejecutivo Comercial Ingeniero Agrónomo Zootecnista deberá:* Hacer la apertura de mercado en la zona de Caldas.* Hacer las visitas comerciales a los diferentes clientes.',
    'IBM Full-Stack JavaScript Developer.Computer Science.Software Development.Master the full-stack development languages, frameworks, tools, and technologies to develop job-ready skills valued by employers.. Write, deploy, and scale cloud-native back-end applications using Node, NoSQL databases, containers, microservices, and serverless.. Develop websites and front-end software using HTML, CSS, JavaScript, and React.. Employ DevOps practices and Agile methodologies to continuously build and deploy software using CI/CD tools.',
    'Data Science with Databricks for Data Analysts.Data Science.Data Analysis.Discover how Databricks and Apache Spark simplify big data processing and optimize data analysis. . Frame business problems for data science and machine learning to make the most out of big data analytic workflows.. Solve real-world business problems quickly using Databricks to power the most popular data science techniques. ',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Training Details

Training Dataset

Unnamed Dataset

Size: 503 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 label
type string string int
details
min: 5 tokens
mean: 87.62 tokens
max: 128 tokens

min: 16 tokens
mean: 64.98 tokens
max: 128 tokens

0: ~73.76%
1: ~7.75%
2: ~18.49%

	sentence1	sentence2	label
type	string	string	int
details	min: 5 tokens mean: 87.62 tokens max: 128 tokens	min: 16 tokens mean: 64.98 tokens max: 128 tokens	0: ~73.76% 1: ~7.75% 2: ~18.49%

Samples:

sentence1	sentence2	label
`Ingenierio electrónico especializado en la implementación de modelos físicos. Experiencia en C++.`	`C++ Programming for Unreal Game Development.Computer Science.Software Development.Computer Programming, C Programming Language Family, Computer Programming Tools, Programming Principles`	`0`
`Analista de datos con años de experiencia. Gran interés hacia el Big Data.`	`Data Literacy: Exploring and Visualizing Data.Data Science.Data Analysis.Data Analysis, Data Management, Data Visualization, Data Visualization Software, Interactive Data Visualization, SAS (Software), Statistical Visualization, Business Analysis, Data Analysis Software, Exploratory Data Analysis, Statistical Analysis, Statistical Programming`	`0`
`Buscamos profesional en profesional Economía, Administración de empresas, Finanzas con MBA. Mínimo 8 años de experiencia en finanzas corporativas, liderando procesos de levantamiento de deuda, liderando equipos multidisciplinarios. Nivel avanzado de inglés`	`Advanced Data Modeling.Information Technology.Data Management.Deploy basic data modeling skills and navigate modern storage options for a data warehouse.. Demonstrate data modeling skills within a real-world project environment.`	`2`

Loss: SoftmaxLoss

Training Hyperparameters

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 8
per_device_eval_batch_size: 8
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
learning_rate: 5e-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: 3.0
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
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: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
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: False
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
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: batch_sampler
multi_dataset_batch_sampler: proportional

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.0.1
Transformers: 4.41.2
PyTorch: 2.3.0+cu121
Accelerate: 0.31.0
Datasets: 2.20.0
Tokenizers: 0.19.1

Citation

BibTeX

Sentence Transformers and SoftmaxLoss

@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",
}