Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Hardware
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

Mohamed-Gamil
/
multilingual-e5-small-JapaneseTeacher

Sentence Similarity
sentence-transformers
Safetensors
English
bert
feature-extraction
Generated from Trainer
dataset_size:4961
loss:MultipleNegativesRankingLoss
Eval Results (legacy)
text-embeddings-inference
Model card Files Files and versions
xet
Community

Instructions to use Mohamed-Gamil/multilingual-e5-small-JapaneseTeacher with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

  • Libraries
  • sentence-transformers

    How to use Mohamed-Gamil/multilingual-e5-small-JapaneseTeacher with sentence-transformers:

    from sentence_transformers import SentenceTransformer
    
    model = SentenceTransformer("Mohamed-Gamil/multilingual-e5-small-JapaneseTeacher")
    
    sentences = [
        "page_content='| 私わたしが困こまっていたとき、                              | 兄あにがいつも助たすけてくれ(た / ました)                 |\n|-------------------------------------------------------------|-------------------------------------------------------------|\n| Time Clause                                                 | Main Clause                                                 |\n| When I was in trouble, my elder brother always helped [me]. | When I was in trouble, my elder brother always helped [me]. |  \nとき is grammatically a noun \"時: time\" and written in Hiragana or Kanji. By using relative clauses with とき, you can express time clauses like \"when\" in English. Thus, you can directly connect sentences to とき without using conjugations. When you connect nouns and na-adjectives, you can also use の instead of the state-of-being style, e.g. 学生のとき VS. 学生だったとき. There is a certain difference in nuance between とき and ときに. When you use とき, main clauses should indicate habitual actions, ongoing states, and constant states.  \n| 学生がくせいのとき、祭まつりが好すき(だった / でした)。  When [I] was a student, [I] liked festivals.   |\n|---|\n| 子こ供どもだったとき、よくポケモンをして(いた / いました)。  When [I] was a child, [I] often played Pokemon.   |  \nWhen it comes to ときに, the に comes from the particle に which indicates specific time. Therefore, when you use ときに, main clauses should indicate non-habitual actions or one-time events.  \n| 学生がくせいのときに、自じ転車てんしゃで旅行りょこう(した / しました)。  When [I] was a student, [I] traveled by bicycle.   |\n|---|\n| 地じ震しんが来きたときに、泣ないて(しまった / しまいました)。  [I] (unintentionally) cried when the earthquake came.   |  \nSince it is a relative clause with a noun, you can substitute other words which have similar meanings to とき. For example, you can use \" 頃 ころ : (approximate) time\" which indicates a wider range of time than とき.  \n| 学生がくせいの頃ころ、祭まつりが好すき(だった / でした)。                 |\n|-----------------------------------------------------------------------------|\n| 学生がくせいの頃ころに、自じ転車てんしゃで旅行りょこう(した / しました)。 |  \nConsidering both of the characteristics, if time clauses indicate actions which take place within a short time, とき is more suitable than 頃 and vice versa. Take a look at the following comparison.  \n| 地じ震しんが起おきたときは、火ひを使つかってはいけ(ない / ません)。  => Natural (As for when earthquakes happen, you must not use fire.)   |\n|---|\n| 地震が起きた頃は、火を使ってはいけ(ない / ません)。  => Wrong   |\n| 学生がくせいのとき、祭まつりが好すき(だった / でした)。  => Natural   |\n| 学生がくせいの頃ころ、祭まつりが好すき(だった / でした)。  => More natural   |' metadata={'h3': 'とき and ときに: When'}",
        "Can you give an example of how それ is used in a conversation about shoes?",
        "Is there a difference between 'I'm singing' and 'I'm singing (and still singing)'?",
        "What is the correct usage of とき versus 頃ころ in time expressions?"
    ]
    embeddings = model.encode(sentences)
    
    similarities = model.similarity(embeddings, embeddings)
    print(similarities.shape)
    # [4, 4]
  • Notebooks
  • Google Colab
  • Kaggle
multilingual-e5-small-JapaneseTeacher
493 MB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 3 commits
Mohamed-Gamil's picture
Mohamed-Gamil
Finetune v2: more epochs, different chunking method (smaller chunks)
3d20cda verified 11 months ago
  • 1_Pooling
    Add new SentenceTransformer model 11 months ago
  • .gitattributes
    1.57 kB
    Add new SentenceTransformer model 11 months ago
  • README.md
    30.7 kB
    Finetune v2: more epochs, different chunking method (smaller chunks) 11 months ago
  • config.json
    628 Bytes
    Add new SentenceTransformer model 11 months ago
  • config_sentence_transformers.json
    205 Bytes
    Add new SentenceTransformer model 11 months ago
  • model.safetensors
    471 MB
    xet
    Finetune v2: more epochs, different chunking method (smaller chunks) 11 months ago
  • modules.json
    349 Bytes
    Add new SentenceTransformer model 11 months ago
  • sentence_bert_config.json
    53 Bytes
    Add new SentenceTransformer model 11 months ago
  • sentencepiece.bpe.model
    5.07 MB
    xet
    Add new SentenceTransformer model 11 months ago
  • special_tokens_map.json
    965 Bytes
    Add new SentenceTransformer model 11 months ago
  • tokenizer.json
    17.1 MB
    xet
    Add new SentenceTransformer model 11 months ago
  • tokenizer_config.json
    1.2 kB
    Add new SentenceTransformer model 11 months ago