metadata

base_model: Bofandra/fine-tuning-use-cmlm-multilingual-quran
datasets: []
language: []
library_name: sentence-transformers
pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:6225
  - loss:MegaBatchMarginLoss
widget:
  - source_sentence: >-
      يا أيها الذين آمنوا لا تتخذوا الكافرين أولياء من دون المؤمنين أتريدون أن
      تجعلوا لله عليكم سلطانا مبينا
    sentences:
      - >-
        And when he attained his full strength and was [mentally] mature, We
        bestowed upon him judgement and knowledge. And thus do We reward the
        doers of good.
      - Then Moses threw his staff, and at once it devoured what they falsified.
      - >-
        O you who have believed, do not take the disbelievers as allies instead
        of the believers. Do you wish to give Allah against yourselves a clear
        case?
  - source_sentence: قال لم أكن لأسجد لبشر خلقته من صلصال من حمإ مسنون
    sentences:
      - And We left it as a sign, so is there any who will remember?
      - Gardens of perpetual residence, whose doors will be opened to them.
      - >-
        He said, "Never would I prostrate to a human whom You created out of
        clay from an altered black mud."
  - source_sentence: وسخر لكم الشمس والقمر دائبين وسخر لكم الليل والنهار
    sentences:
      - >-
        And He subjected for you the sun and the moon, continuous [in orbit],
        and subjected for you the night and the day.
      - >-
        And We called him from the side of the mount at [his] right and brought
        him near, confiding [to him].
      - >-
        And We send not the messengers except as bringers of good tidings and
        warners. And those who disbelieve dispute by [using] falsehood to
        [attempt to] invalidate thereby the truth and have taken My verses, and
        that of which they are warned, in ridicule.
  - source_sentence: إذ دخلوا عليه فقالوا سلاما قال إنا منكم وجلون
    sentences:
      - >-
        Indeed, your Lord is most knowing of who strays from His way, and He is
        most knowing of the [rightly] guided.
      - >-
        Then they turned away from him and said, "[He was] taught [and is] a
        madman."
      - >-
        When they entered upon him and said, "Peace." [Abraham] said, "Indeed,
        we are fearful of you."
  - source_sentence: فأما من أوتي كتابه بيمينه فيقول هاؤم اقرءوا كتابيه
    sentences:
      - >-
        So as for he who is given his record in his right hand, he will say,
        "Here, read my record!
      - >-
        And whoever is patient and forgives - indeed, that is of the matters
        [requiring] determination.
      - Indeed, he had [once] been among his people in happiness;

SentenceTransformer based on Bofandra/fine-tuning-use-cmlm-multilingual-quran

This is a sentence-transformers model finetuned from Bofandra/fine-tuning-use-cmlm-multilingual-quran. It maps sentences & paragraphs to a 768-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: Bofandra/fine-tuning-use-cmlm-multilingual-quran
Maximum Sequence Length: 256 tokens
Output Dimensionality: 768 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': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, '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()
)

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("Bofandra/fine-tuning-use-cmlm-multilingual-quran-translation")
# Run inference
sentences = [
    'فأما من أوتي كتابه بيمينه فيقول هاؤم اقرءوا كتابيه',
    'So as for he who is given his record in his right hand, he will say, "Here, read my record!',
    'Indeed, he had [once] been among his people in happiness;',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

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

Training Details

Training Dataset

Unnamed Dataset

Size: 6,225 training samples
Columns: sentence_0 and sentence_1
Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1
type string string
details
min: 3 tokens
mean: 23.11 tokens
max: 163 tokens

min: 6 tokens
mean: 34.63 tokens
max: 180 tokens

	sentence_0	sentence_1
type	string	string
details	min: 3 tokens mean: 23.11 tokens max: 163 tokens	min: 6 tokens mean: 34.63 tokens max: 180 tokens

Samples:

sentence_0	sentence_1
`ومن آياته أنك ترى الأرض خاشعة فإذا أنزلنا عليها الماء اهتزت وربت إن الذي أحياها لمحيي الموتى إنه على كل شيء قدير`	`And of His signs is that you see the earth stilled, but when We send down upon it rain, it quivers and grows. Indeed, He who has given it life is the Giver of Life to the dead. Indeed, He is over all things competent.`
`من دون الله قالوا ضلوا عنا بل لم نكن ندعو من قبل شيئا كذلك يضل الله الكافرين`	`Other than Allah?" They will say, "They have departed from us; rather, we did not used to invoke previously anything." Thus does Allah put astray the disbelievers.`
`أرأيت الذي ينهى`	`Have you seen the one who forbids`

Loss: MegaBatchMarginLoss

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 4
per_device_eval_batch_size: 4
num_train_epochs: 1
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 4
per_device_eval_batch_size: 4
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
num_train_epochs: 1
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: round_robin

Training Logs

Epoch	Step	Training Loss
0.3211	500	0.2393
0.6423	1000	0.1212
0.9634	1500	0.0715

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

@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",
}

MegaBatchMarginLoss

@inproceedings{wieting-gimpel-2018-paranmt,
    title = "{P}ara{NMT}-50{M}: Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations",
    author = "Wieting, John and Gimpel, Kevin",
    editor = "Gurevych, Iryna and Miyao, Yusuke",
    booktitle = "Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2018",
    address = "Melbourne, Australia",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/P18-1042",
    doi = "10.18653/v1/P18-1042",
    pages = "451--462",
}