metadata

base_model: BAAI/bge-small-en-v1.5
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:1944
  - loss:BatchAllTripletLoss
widget:
  - source_sentence: 計算費率給定的延期年金的值。
    sentences:
      - "How can I generate a random list of 10 adjectives in Spanish using Ruby? Can you provide me with the Ruby code to generate this list and display it in an HTML page similar to the example given below?\n<!DOCTYPE html>\n<html>\n<head>\n\t<title>Random Adjectives in Spanish</title>\n</head>\n<body>\n\t<h1>Random Adjectives in Spanish</h1>\n\t<p>Here are ten random examples of adjectives in Spanish:</p>\n\t<ul>\n\t\t<li><strong>Bonito</strong> - Beautiful</li>\n\t\t<li><strong>Tranquilo</strong> - Calm</li>\n\t\t<li><strong>Grande</strong> - Big</li>\n\t\t<li><strong>Pequeño</strong> - Small</li>\n\t\t<li><strong>Interesante</strong> - Interesting</li>\n\t\t<li><strong>Dulce</strong> - Sweet</li>\n\t\t<li><strong>Divertido</strong> - Fun</li>\n\t\t<li><strong>Bueno</strong> - Good</li>\n\t\t<li><strong>Malo</strong> - Bad</li>\n\t\t<li><strong>Viejo</strong> - Old</li>\n\t</ul>\n</body>\n</html>\nPlease include the Ruby code to generate the random list of adjectives and display it in the HTML page. "
      - 创建一个尽可能高效的Python排序算法，可以在大数据量下处理包含重复元素的整数数组。你的算法还应处理负数和零。此外，给出相应的复杂度分析。
      - 好的质量要求标注于射线底片上的IR191的国际标准有哪些常见要求？____、____及____。
  - source_sentence: >-
      How can I modify the above Scala code to use XML data instead of CSV data
      for training and test sets? 

      Assuming the XML data is formatted as follows:

      <dataset>
        <instance>
          <age>32</age>
          <income>75000</income>
          <category>Technology</category>
          <gender>Male</gender>
          <label>1</label>
        </instance>
        <instance>
          <age>45</age>
          <income>85000</income>
          <category>Finance</category>
          <gender>Female</gender>
          <label>0</label>
        </instance>
        ...
      </dataset>

      Here is the modified Scala code using Apache Spark's MLlib library and XML
      data:

      // Import necessary libraries

      import org.apache.spark.ml.classification.LogisticRegression

      import org.apache.spark.ml.evaluation.BinaryClassificationEvaluator

      import org.apache.spark.ml.feature.{VectorAssembler, StringIndexer}

      import org.apache.spark.sql.functions.{explode, col}

      // Load the training and test data from XML files

      val trainingData = spark.read
        .format("com.databricks.spark.xml")
        .option("rowTag", "instance")
        .load("training_data.xml")
      val testData = spark.read
        .format("com.databricks.spark.xml")
        .option("rowTag", "instance")
        .load("test_data.xml")
      // Convert categorical features to numerical using StringIndexer

      val categoricalCols = Array("category", "gender")

      val indexers = categoricalCols.map(col => new
      StringIndexer().setInputCol(col).setOutputCol(col + "_index"))

      val pipeline = new Pipeline().setStages(indexers)

      val indexedTrainingData =
      pipeline.fit(trainingData).transform(trainingData)

      val indexedTestData = pipeline.fit(testData).transform(testData)

      // Assemble all features into a single vector column

      val assembler = new VectorAssembler().setInputCols(Array("age", "income",
      "category_index", "gender_index")).setOutputCol("features")

      val trainingDataWithFeatures = assembler.transform(indexedTrainingData)

      val testDataWithFeatures = assembler.transform(indexedTestData)

      // Train the logistic regression model

      val lr = new
      LogisticRegression().setLabelCol("label").setFeaturesCol("features")

      val model = lr.fit(trainingDataWithFeatures)

      // Make predictions on the test data

      val predictions = model.transform(testDataWithFeatures)

      // Evaluate the model using the binary classification evaluator

      val evaluator = new
      BinaryClassificationEvaluator().setLabelCol("label").setRawPredictionCol("rawPrediction")

      val auc = evaluator.evaluate(predictions)  // Area Under the ROC Curve
      (AUC) score

      println("Area Under the ROC Curve (AUC) score: " + auc) 
    sentences:
      - >-
        How can I use Ruby to solve this puzzle and calculate the difference
        between the average population density of metropolitan and
        non-metropolitan areas? And as an added challenge, how can I determine
        the top 3 metropolitan areas with the highest population density and the
        bottom 3 non-metropolitan areas with the lowest population density?

        Here's the data for the puzzle:

        [{"city": "New York", "population": 8398748, "area": 468.9,
        "metropolitan": true},
         {"city": "Los Angeles", "population": 3990456, "area": 502.7, "metropolitan": true},
         {"city": "Chicago", "population": 2705994, "area": 227.3, "metropolitan": true},
         {"city": "Houston", "population": 2325502, "area": 637.5, "metropolitan": true},
         {"city": "Phoenix", "population": 1660272, "area": 517.6, "metropolitan": true},
         {"city": "Philadelphia", "population": 1584138, "area": 347.6, "metropolitan": true},
         {"city": "San Antonio", "population": 1532233, "area": 1196.5, "metropolitan": true},
         {"city": "San Diego", "population": 1425976, "area": 964.5, "metropolitan": true},
         {"city": "Dallas", "population": 1345047, "area": 999.3, "metropolitan": true},
         {"city": "San Jose", "population": 1030119, "area": 467.4, "metropolitan": true},
         {"city": "Austin", "population": 964254, "area": 815.2, "metropolitan": true},
         {"city": "Fort Worth", "population": 895008, "area": 874.4, "metropolitan": true},
         {"city": "Columbus", "population": 892533, "area": 571.8, "metropolitan": true},
         {"city": "San Francisco", "population": 883305, "area": 121.4, "metropolitan": true},
         {"city": "Charlotte", "population": 872498, "area": 771.0, "metropolitan": true},
         {"city": "Indianapolis", "population": 867125, "area": 372.9, "metropolitan": true},
         {"city": "Seattle", "population": 744955, "area": 217.2, "metropolitan": true},
         {"city": "Denver", "population": 716492, "area": 401.2, "metropolitan": true},
         {"city": "Washington", "population": 702455, "area": 177.0, "metropolitan": true},
         {"city": "Boston", "population": 694583, "area": 89.6, "metropolitan": true},
         {"city": "Nashville-Davidson", "population": 669053, "area": 1558.0, "metropolitan": true},
         {"city": "El Paso", "population": 649121, "area": 667.0, "metropolitan": true},
         {"city": "Portland", "population": 632309, "area": 376.5, "metropolitan": true},
         {"city": "Oklahoma City", "population": 631346, "area": 1599.6, "metropolitan": true},
         {"city": "Las Vegas", "population": 644644, "area": 352.0, "metropolitan": true},
         {"city": "Detroit", "population": 673104, "area": 359.4, "metropolitan": true},
         {"city": "Memphis", "population": 652717, "area": 763.4, "metropolitan": true},
         {"city": "Louisville/Jefferson County", "population": 620118, "area": 1595.6, "metropolitan": true},
         {"city": "Baltimore", "population": 602495, "area": 92.1, "metropolitan": true},
         {"city": "Milwaukee", "population": 590157, "area": 248.9, "metropolitan": true},
         {"city": "Albuquerque", "population": 560218, "area": 491.9, "metropolitan": true},
         {"city": "Tucson", "population": 545975, "area": 596.7, "metropolitan": true},
         {"city": "Fresno", "population": 520052, "area": 114.4, "metropolitan": true},
         {"city": "Mesa", "population": 508958, "area": 324.2, "metropolitan": true},
         {"city": "Sacramento", "population": 501901, "area": 97.9, "metropolitan": true},
         {"city": "Atlanta", "population": 498715, "area": 133.2, "metropolitan": true},
         {"city": "Kansas City", "population": 488943, "area": 319.0, "metropolitan": true},
         {"city": "Colorado Springs", "population": 472688, "area": 503.1, "metropolitan": true},
         {"city": "Miami", "population": 463347, "area": 92.9, "metropolitan": true},
         {"city": "Raleigh", "population": 464758, "area": 369.9, "metropolitan": true},
         {"city": "Omaha", "population": 466893, "area": 338.2, "metropolitan": true},
         {"city": "Long Beach", "population": 469450, "area": 170.6, "metropolitan": true},
         {"city": "Virginia Beach", "population": 450980, "area": 497.3, "metropolitan": true},
         {"city": "Oakland", "population": 425195, "area": 78.0, "metropolitan": true},
         {"city": "Minneapolis", "population": 422331, "area": 151.4, "metropolitan": true},
         {"city": "Tulsa", "population": 401800, "area": 484.9, "metropolitan": true},
         {"city": "Wichita", "population": 390591, "area": 355.6, "metropolitan": true},
         {"city": "New Orleans", "population": 390144, "area": 466.9, "metropolitan": true},
         {"city": "Arlington", "population": 388125, "area": 261.8, "metropolitan": true},
         {"city": "Tampa", "population": 352957, "area": 113.4, "metropolitan": true},
         {"city": "Santa Ana", "population": 332725, "area": 71.0, "metropolitan": true},
         {"city": "Anaheim", "population": 336265, "area": 131.7, "metropolitan": true},
         {"city": "St. Louis", "population": 319294, "area": 66.2, "metropolitan": true},
         {"city": "Riverside", "population": 316619, "area": 204.0, "metropolitan": true},
         {"city": "Corpus Christi", "population": 316381, "area": 1606.8, "metropolitan": true},
         {"city": "Lexington-Fayette", "population": 314488, "area": 739.5, "metropolitan": true},
         {"city": "Pittsburgh", "population": 302407, "area": 151.1, "metropolitan": true},
         {"city": "Stockton", "population": 291707, "area": 151.2, "metropolitan": true},
         {"city": "Cincinnati", "population": 301301, "area": 200.3, "metropolitan": true},
         {"city": "St. Paul", "population": 285068, "area": 52.8, "metropolitan": true},
         {"city": "Toledo", "population": 276491, "area": 217.2, "metropolitan": true},
         {"city": "Greensboro", "population": 279639, "area": 283.1, "metropolitan": true},
         {"city": "Newark", "population": 282090, "area": 26.1, "metropolitan": true},
         {"city": "Plano", "population": 269776, "area": 185.1, "metropolitan": true},
         {"city": "Henderson", "population": 270811, "area": 267.4, "metropolitan": true},
         {"city": "Lincoln", "population": 258379, "area": 196.9, "metropolitan": true},
         {"city": "Buffalo", "population": 256902, "area": 136.0, "metropolitan": true},
         {"city": "Jersey City", "population": 247597, "area": 21.1, "metropolitan": true},
         {"city": "Chula Vista", "population": 243916, "area": 52.1, "metropolitan": true},
         {"city": "Fort Wayne", "population": 253691,  
      - >-
        How would you use Python to count the occurrences of a specific word in
        a given text? "the quick brown fox jumped over the lazy dog", "the
      - 可以介绍一下监狱的运作吗？
  - source_sentence: >-
      Can you provide a Scala code that can arrange a list of words into a
      grammatically correct sentence? Specifically, can you arrange the
      following list of words: "dog cat run quickly"? 
    sentences:
      - >-
        How can I use C# to generate a set of 150 words for a crossword puzzle?
        Can you provide a sample C# code that will randomly select 150 words
        from a table called 'crossword_words' in a SQL database and output them
        as a grid for the puzzle?

        Here is a sample C# code that uses the SQL query to select 150 random
        words from the 'crossword_words' table and output them as a grid for the
        crossword puzzle:

        ```csharp

        using System;

        using System.Data.SqlClient;

        namespace CrosswordPuzzleGenerator

        {
            class Program
            {
                static void Main(string[] args)
                {
                    string connectionString = "Data Source=YourServerName;Initial Catalog=YourDatabaseName;Integrated Security=True";
                    SqlConnection connection = new SqlConnection(connectionString);
                    connection.Open();
                    SqlCommand command = new SqlCommand("SELECT TOP 150 * FROM crossword_words ORDER BY NEWID()", connection);
                    SqlDataReader reader = command.ExecuteReader();
                    string[] words = new string[150];
                    int index = 0;
                    while (reader.Read())
                    {
                        words[index] = reader.GetString(0);
                        index++;
                    }
                    reader.Close();
                    connection.Close();
                    // Create grid for crossword puzzle
                    char[,] grid = new char[15, 15];
                    for (int i = 0; i < 15; i++)
                    {
                        for (int j = 0; j < 15; j++)
                        {
                            grid[i, j] = '_';
                        }
                    }
                    // Insert words horizontally
                    foreach (string word in words)
                    {
                        int row = new Random().Next(0, 15);
                        int col = new Random().Next(0, 16 - word.Length);
                        for (int i = 0; i < word.Length; i++)
                        {
                            grid[row, col + i] = word[i];
                        }
                    }
                    // Insert words vertically
                    foreach (string word in words)
                    {
                        int row = new Random().Next(0, 16 - word.Length);
                        int col = new Random().Next(0, 15);
                        for (int i = 0; i < word.Length; i++)
                        {
                            grid[row + i, col] = word[i];
                        }
                    }
                    // Print grid for crossword puzzle
                    for (int i = 0; i < 15; i++)
                    {
                        for (int j = 0; j < 15; j++)
                        {
                            Console.Write(grid[i, j] + " ");
                        }
                        Console.WriteLine();
                    }
                }
            }
        }

        ```

        To use this code, simply replace 'crossword_words' with the name of your
        table in the SQL query and run the C# code. This code will generate a
        crossword puzzle grid with 150 random words from the table. 
      - >-
        对于由 -1，0 和 1 组成的一个二维格子，其宽度和高度均为 n x
        n，请设计一个算法，需要找到并返回最大的连接在一起的相同数字矩阵，矩阵也可以是单格。这里所谓的“连接”，指的是彼此相邻的单元格（从四个方向：左右上下），共享相同数值（要么-1，要么0，要么1）。
      - 请求创建一个Python函数，以确定给定正整数的阶乘是否在特定范围内产生尾随零。
  - source_sentence: >-
      How can I generate a randomized short story of at least 1000 words using
      the given characters Mark, Sarah, and Alice and their actions as input
      data in a MATLAB code?

      I have the following table in MATLAB:

      StoryEvents = [1,2,3,4,5,6,7,8,9,10]';

      Characters =
      {'Mark';'Sarah';'Alice';'Mark';'Sarah';'Alice';'Mark';'Sarah';'Alice';'Mark'};

      Actions = {'Meets Sarah at the park';'Tells Mark about her love for
      art';'Joins Mark and Sarah for lunch';'Discovers Alice is a
      musician';'Helps Alice prepare for a concert';'Invites Mark and Sarah to
      her concert';'Realizes he has feelings for Sarah';'Confesses her love for
      Mark';'Performs at her concert';'Kisses Sarah after the concert'};

      How can I use this table to generate a unique and engaging story each time
      the code is run? 
    sentences:
      - javascript中定义的函数是如何执行的
      - >-
        What R code can be used to create a personalized meal plan for a week
        based on dietary restrictions and preferences?

        Here is an example R code for a personalized meal plan:

        ```{r}

        # Define dietary restrictions and preferences

        vegetarian <- TRUE

        gluten_free <- TRUE

        prefer_spicy <- TRUE

        # Define meal options

        breakfast_options <- c("oatmeal", "smoothie", "avocado toast")

        lunch_options <- c("salad", "vegetarian wrap", "quinoa bowl")

        dinner_options <- c("spicy tofu stir-fry", "zucchini noodles with
        marinara", "stuffed bell peppers")

        # Create meal plan for the week

        days <- c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday",
        "Saturday", "Sunday")

        meal_plan <- data.frame(Day = days)

        for (i in 1:length(days)) {
          if (vegetarian) {
            options <- lunch_options[lunch_options != "vegetarian wrap"]
          } else {
            options <- lunch_options
          }
          if (gluten_free) {
            options <- options[options != "quinoa bowl"]
          }
          if (prefer_spicy) {
            dinner_options <- c(dinner_options, "spicy lentil soup")
          }
          breakfast <- sample(breakfast_options, 1)
          lunch <- sample(options, 1)
          dinner <- sample(dinner_options, 1)
          meal_plan[i, 2:4] <- c(breakfast, lunch, dinner)
        }

        # Print meal plan

        cat("Here is your personalized meal plan for the week:\n")

        print(meal_plan)

        ```

        Feel free to adjust the dietary restrictions and preferences, as well as
        the meal options to create a customized meal plan for yourself or
        others. 
      - 你能创建月球、火星或其他行星的拟人化角色吗？
  - source_sentence: ios如何获取url中的参数
    sentences:
      - 下面是《自尊之外：互动心理与互动机制》中的一段对话，请根据对话写出节选自的原文。
      - >-
        How can the given Ruby code be enhanced to also extract the
        prepositional phrase that contains the adverb modifying the verb in the
        sentence? The updated code must output the complete prepositional phrase
        along with the noun, verb, and adjectives in the sentence. 
      - 小明家2月份计划支出每天150元，通过节约最后每天只用105元，请问这为家庭节省了多少钱？

SentenceTransformer based on BAAI/bge-small-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. 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: BAAI/bge-small-en-v1.5
Maximum Sequence Length: 512 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': 512, 'do_lower_case': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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()
)

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("Snivellus789/router-embedding-tuned-human")
# Run inference
sentences = [
    'ios如何获取url中的参数',
    'How can the given Ruby code be enhanced to also extract the prepositional phrase that contains the adverb modifying the verb in the sentence? The updated code must output the complete prepositional phrase along with the noun, verb, and adjectives in the sentence. ',
    '小明家2月份计划支出每天150元，通过节约最后每天只用105元，请问这为家庭节省了多少钱？',
]
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: 1,944 training samples
Columns: sentence and label
Approximate statistics based on the first 1000 samples:
sentence label
type string int
details
min: 8 tokens
mean: 95.61 tokens
max: 512 tokens

0: ~50.00%
1: ~50.00%

	sentence	label
type	string	int
details	min: 8 tokens mean: 95.61 tokens max: 512 tokens	0: ~50.00% 1: ~50.00%

Samples:

sentence	label
`请输出所有跟政企市场相关的关键词列表`	`0`
`开发一个定制的JavaScript解决方案，用于有效地平衡和排序一个二叉树。你可以假设输入是一个平衡因子擯至2的大O()为Log(N)的AVL树。专注于实现自我调整二叉搜索树的变换，当面对不平衡操作时，如插入或删除节点。确保你的解决方案为潜在的边缘案例做好准备，并具有健壮的错误处理策略。你的代码应该清晰地记录和优化效率。`	`0`
`在一个尚未被公开的领域中，描述五个最具创新性的产品概念。`	`0`

Loss: BatchAllTripletLoss

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16
learning_rate: 2e-05
num_train_epochs: 2
warmup_ratio: 0.1
bf16: True
batch_sampler: no_duplicates

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
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: 2e-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: 2
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.1
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: True
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
eval_on_start: False
eval_use_gather_object: False
batch_sampler: no_duplicates
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss
0.8197	100	0.1444
1.6393	200	0.0905

Framework Versions

Python: 3.10.12
Sentence Transformers: 3.1.1
Transformers: 4.44.2
PyTorch: 2.4.1+cu121
Accelerate: 0.35.0.dev0
Datasets: 3.0.1
Tokenizers: 0.19.1

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",
}

BatchAllTripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}