tags:

• sentence-transformers
• sentence-similarity
• feature-extraction
• dense
• generated_from_trainer
• dataset_size:362
• loss:CosineSimilarityLoss base_model: sentence-transformers/all-MiniLM-L6-v2 widget:
• source_sentence: What's her experience like? sentences:
  • What's her experience like?
  • Alyza and her teammates delivered a 94% improvement in time efficiency.
  • Alyza developed dashboards for Sell-In and Sell-Out analysis, analyzed and prepared sales data for meetings, collaborated on analyzing promotion-sales correlations, predicted future sales, analyzed stock on hand and offtake data, designed monthly sales plans, updated performance data, managed master data, created promotional cooperation letters, and addressed claim issues.
• source_sentence: Good afternoon sentences:
  • Good afternoon! I'm here to help you learn about Alyza Rahima Pramudya. What would you like to know about her?
  • It's a predictive model to identify Telco customers likely to churn, helping reduce customer loss.
  • Can you tell me about the Urban Visual Pollutants Detection project?
• source_sentence: What responsibilities did she have at Auto2000? sentences:
  • Can you name some of her technical projects and applications?
  • As a Digital Project Consultant, Alyza identified, assessed, developed, tested, and implemented Robotic Process Automation (RPA) using UiPath, designed and developed Power BI dashboards, and developed automation scripts for report generation.
  • Hello! I'm here to help you learn about Alyza Rahima Pramudya. What would you like to know about her education, work experience, projects, or achievements?
• source_sentence: Can you tell me about the news classification project? sentences:
  • Can you tell me about the news classification project?
  • Can you describe her duties as a Digital Project Consultant?
  • Alyza placement at Auto2000 was part of the Astra1st program.
• source_sentence: What prestigious programs has Alyza been selected for? sentences:
  • 'Alyza''s projects include DearCSV, Ask Me Girl!, Prompt & Prejudice, Dog Breed Classifierz, IKN Sentiment App, Frezz : Fruit Freshness Detector, Covid-19 in US: Weather & Socioeconomic Factors, Urban Visual Pollutants Detection, WHO: Life Expectancy Analysis, News Category Classification, Jakarta Air Quality Classification, Diabetes Classification & Regression, and Telco Customer Churn Prediction.'
  • Alyza is an Astra1st Batch XII Awardee (chosen from over 6,900 applicants, 0.62% acceptance rate) and her team was honored as the Best Team in Astra1st Batch XII. She is also a Mastering AI Batch IV Awardee, receiving a full scholarship for the bootcamp by Skill Academy Pro x Ruangguru Engineering Academy.
  • Alyza worked as a Digital Project Consultant from June 2024 to November 2024. pipeline_tag: sentence-similarity library_name: sentence-transformers

基于 sentence-transformers/all-MiniLM-L6-v2 的 SentenceTransformer

这是一个基于 sentence-transformers/all-MiniLM-L6-v2 微调的 sentence-transformers 模型。它将句子和段落映射到 384 维的稠密向量空间，可用于语义文本相似度、语义搜索、改写挖掘、文本分类、聚类等任务。

模型详情

模型描述

• 模型类型: Sentence Transformer
• 基础模型: sentence-transformers/all-MiniLM-L6-v2 <!-- at revision c9745ed1d9f207416be6d2e6f8de32d1f16199bf -->
• 最大序列长度: 256 个 token
• 输出维度: 384 维
• 相似度函数: 余弦相似度 <!-- - 训练数据集: 未知 --> <!-- - 语言: 未知 --> <!-- - 许可证: 未知 -->

模型来源

• 文档: Sentence Transformers 文档
• 代码库: Sentence Transformers GitHub
• Hugging Face: Sentence Transformers on Hugging Face

完整模型架构

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False, 'architecture': '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})
  (2): Normalize()
)

使用方法

直接使用 (Sentence Transformers)

首先安装 Sentence Transformers 库:

pip install -U sentence-transformers

然后你可以加载此模型并运行推理。

from sentence_transformers import SentenceTransformer

# 从 🤗 Hub 下载
model = SentenceTransformer("sentence_transformers_model_id")
# 运行推理
sentences = [
    'What prestigious programs has Alyza been selected for?',
    'Alyza is an Astra1st Batch XII Awardee (chosen from over 6,900 applicants, 0.62% acceptance rate) and her team was honored as the Best Team in Astra1st Batch XII. She is also a Mastering AI Batch IV Awardee, receiving a full scholarship for the bootcamp by Skill Academy Pro x Ruangguru Engineering Academy.',
    "Alyza's projects include DearCSV, Ask Me Girl!, Prompt & Prejudice, Dog Breed Classifierz, IKN Sentiment App, Frezz : Fruit Freshness Detector, Covid-19 in US: Weather & Socioeconomic Factors, Urban Visual Pollutants Detection, WHO: Life Expectancy Analysis, News Category Classification, Jakarta Air Quality Classification, Diabetes Classification & Regression, and Telco Customer Churn Prediction.",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# 获取嵌入的相似度分数
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9866, 0.9782],
#         [0.9866, 1.0000, 0.9715],
#         [0.9782, 0.9715, 1.0000]])

<!--

直接使用 (Transformers)

<details><summary>点击查看 Transformers 中的直接使用方法</summary>

</details> -->

<!--

下游使用 (Sentence Transformers)

你可以在自己的数据集上微调此模型。

<details><summary>点击展开</summary>

</details> -->

<!--

超出范围的使用

列出模型可能被预见性地滥用的方式，并说明用户不应对模型做什么。 -->

<!--

偏见、风险和局限性

这个模型有哪些已知或可预见的问题？你也可以在这里标记已知的失败案例或模型的弱点。 -->

<!--

建议

针对可预见的问题有哪些建议？例如，过滤显式内容。 -->

训练详情

训练数据集

未命名数据集

• 大小: 362 个训练样本
• 列: <code>sentence_0</code>, <code>sentence_1</code>, 和 <code>label</code>
• 基于前 362 个样本的近似统计:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	<ul><li>min: 3 tokens</li><li>mean: 12.76 tokens</li><li>max: 23 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 28.02 tokens</li><li>max: 137 tokens</li></ul>	<ul><li>min: 1.0</li><li>mean: 1.0</li><li>max: 1.0</li></ul>

• 样本:

  sentence_0	sentence_1	label
  <code>What is alyza's full name?</code>	<code>Alyza's full name is Alyza Rahima Pramudya.</code>	<code>1.0</code>
  <code>Can you tell me about Prompt & Prejudice?</code>	<code>Prompt & Prejudice creates dreamy romance ideas based on user inputs or random generation.</code>	<code>1.0</code>
  <code>How does the News Category Classification project work?</code>	<code>How does the News Category Classification project work?</code>	<code>1.0</code>

• 损失函数: <code>CosineSimilarityLoss</code> 参数如下:

  {
      \"loss_fct\": \"torch.nn.modules.loss.MSELoss\"
  }
  

训练超参数

非默认超参数

• num_train_epochs: 10
• multi_dataset_batch_sampler: round_robin

所有超参数

<details><summary>点击展开</summary>

• 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
• torch_empty_cache_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
• num_train_epochs: 10
• 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: None
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• include_for_metrics: []
• 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
• include_tokens_per_second: False
• include_num_input_tokens_seen: False
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

</details>

框架版本

• Python: 3.13.3
• Sentence Transformers: 5.0.0
• Transformers: 4.52.4
• PyTorch: 2.7.1+cpu
• Accelerate: 1.8.1
• Datasets: 3.6.0
• Tokenizers: 0.21.2

引用

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\",
}

<!--

术语表

明确定义术语以便各受众理解。 -->

<!--

模型卡作者

列出创建模型卡的人员，为模型卡构建中的详细工作提供认可和责任。 -->

<!--

模型卡联系方式

为有模型卡更新、建议或问题的人提供联系模型卡作者的方式。 -->