---
base_model: sentence-transformers/stsb-xlm-r-multilingual
datasets: []
language: []
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:193860
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: ഓ, അതെ, ഞാൻ അവരെ കഴുത്ത് ഞെരിച്ച് കൊല്ലുമായിരുന്നു എന്ന ചിന്തയെക്കുറിച്ച്
വായിച്ചത് ഞാൻ ഓർക്കുന്നു.
sentences:
- A major privacy related disaster might be an exception.
- আমি এটি সম্পর্কে পড়েছি এবং ভেবেছিলাম যে আমাকে তাদের শ্বাসরোধ করতে হবে।
- How do you like it out there?
- source_sentence: A male tennis player hits a tennis ball at a tennis match.
sentences:
- You can shower outside in nature with privacy.
- രണ്ട് കാറുകൾ ഓടുന്നു.
- একজন লোক টেনিস খেলছে।
- source_sentence: ఒక గోధుమ మరియు తెలుపు కుక్క ఒక ఇంటి ప్రాంగణంలో ఆడుతోంది.
sentences:
- आँगन में एक कुत्ता खेल रहा है।
- तालप्रिय ग्वाडेलोपियन पर्यटकांच्या संख्येपेक्षा जास्त असतील.
- সবুজ পোশাকে কিছু মানুষ
- source_sentence: A baby wearing a pink outfit with flowers on it has its mouth open.
sentences:
- बैठकीच्या खोलीच्या भिंती पांढऱ्या आहेत.
- a baby is in a pink outfit
- এটা ভালো হবে, কিন্তু আমি স্বাধীনতা উপভোগ করি।
- source_sentence: A baby wearing a watch.
sentences:
- એક બાળક ઘડિયાળ પહેરી રહ્યું છે.
- एक बेसबॉल खिलाड़ी गेंद पर झूलता है।
- The mans legs are touching.
model-index:
- name: SentenceTransformer based on sentence-transformers/stsb-xlm-r-multilingual
results:
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts dev
type: sts-dev
metrics:
- type: pearson_cosine
value: 0.8658165968626415
name: Pearson Cosine
- type: spearman_cosine
value: 0.8714077275778997
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8695576458225691
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8700925845327402
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8694747813672388
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8703633875862249
name: Spearman Euclidean
- type: pearson_dot
value: 0.7735824081876905
name: Pearson Dot
- type: spearman_dot
value: 0.7728637057026586
name: Spearman Dot
- type: pearson_max
value: 0.8695576458225691
name: Pearson Max
- type: spearman_max
value: 0.8714077275778997
name: Spearman Max
- task:
type: semantic-similarity
name: Semantic Similarity
dataset:
name: sts test
type: sts-test
metrics:
- type: pearson_cosine
value: 0.8413454674808685
name: Pearson Cosine
- type: spearman_cosine
value: 0.8516557437790466
name: Spearman Cosine
- type: pearson_manhattan
value: 0.8406890199541754
name: Pearson Manhattan
- type: spearman_manhattan
value: 0.8401478064056196
name: Spearman Manhattan
- type: pearson_euclidean
value: 0.8405040750844844
name: Pearson Euclidean
- type: spearman_euclidean
value: 0.8402979769379469
name: Spearman Euclidean
- type: pearson_dot
value: 0.7261415217517116
name: Pearson Dot
- type: spearman_dot
value: 0.7095416925344771
name: Spearman Dot
- type: pearson_max
value: 0.8413454674808685
name: Pearson Max
- type: spearman_max
value: 0.8516557437790466
name: Spearman Max
---
# SentenceTransformer based on sentence-transformers/stsb-xlm-r-multilingual
This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/stsb-xlm-r-multilingual](https://huggingface.co/sentence-transformers/stsb-xlm-r-multilingual). 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:** [sentence-transformers/stsb-xlm-r-multilingual](https://huggingface.co/sentence-transformers/stsb-xlm-r-multilingual)
- **Maximum Sequence Length:** 128 tokens
- **Output Dimensionality:** 768 tokens
- **Similarity Function:** Cosine Similarity
### Model Sources
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers)
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
### Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: XLMRobertaModel
(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})
)
```
## Usage
### Direct Usage (Sentence Transformers)
First install the Sentence Transformers library:
```bash
pip install -U sentence-transformers
```
Then you can load this model and run inference.
```python
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
'A baby wearing a watch.',
'એક બાળક ઘડિયાળ પહેરી રહ્યું છે.',
'The mans legs are touching.',
]
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]
```
## Evaluation
### Metrics
#### Semantic Similarity
* Dataset: `sts-dev`
* Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8658 |
| **spearman_cosine** | **0.8714** |
| pearson_manhattan | 0.8696 |
| spearman_manhattan | 0.8701 |
| pearson_euclidean | 0.8695 |
| spearman_euclidean | 0.8704 |
| pearson_dot | 0.7736 |
| spearman_dot | 0.7729 |
| pearson_max | 0.8696 |
| spearman_max | 0.8714 |
#### Semantic Similarity
* Dataset: `sts-test`
* Evaluated with [EmbeddingSimilarityEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)
| Metric | Value |
|:--------------------|:-----------|
| pearson_cosine | 0.8413 |
| **spearman_cosine** | **0.8517** |
| pearson_manhattan | 0.8407 |
| spearman_manhattan | 0.8401 |
| pearson_euclidean | 0.8405 |
| spearman_euclidean | 0.8403 |
| pearson_dot | 0.7261 |
| spearman_dot | 0.7095 |
| pearson_max | 0.8413 |
| spearman_max | 0.8517 |
## Training Details
### Training Dataset
#### Unnamed Dataset
* Size: 193,860 training samples
* Columns: query, positive, and negative
* Approximate statistics based on the first 1000 samples:
| | query | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string | string |
| details | - min: 4 tokens
- mean: 26.96 tokens
- max: 128 tokens
| - min: 5 tokens
- mean: 14.19 tokens
- max: 128 tokens
| - min: 5 tokens
- mean: 14.86 tokens
- max: 60 tokens
|
* Samples:
| query | positive | negative |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------|
| Delos was not only an important religious center, but also a major meeting point for trade between East and West during the Hellenistic and Roman eras. | The East and West met at Delos to trade. | All of the buildings are open to visitors. |
| काळ्या रंगाचा शर्ट घातलेली एक स्त्री तिच्या उजवीकडे पाहते, तर तिच्या डाव्या बाजूला निळ्या रंगाची बनियान घातलेला एक माणूस काचेतून पाणी पितो. | કાળા શર્ટમાં મહિલા તેના લખાણ તરફ જોઈ રહી હતી જ્યારે તેની બાજુના સજ્જન તેની તરસ છીપાવી રહ્યા હતા. | Armies of Cathar heretics and Roman church battled near Albi la Rouge. |
| કોંક્રિટ પગથિયા પર બેઠેલા ધાબળામાં વીંટાળેલા નાના બાળક સાથેનું દંપતી | సంబంధంలో ఉన్న ఇద్దరు వ్యక్తులు ఒక బిడ్డతో కూర్చున్నారు. | যারা আইনি সহায়তা চাইছেন তাদের জন্য এনজেপি ইন্টারনেট ভিত্তিক সহায়তা এবং সহায়তা প্রদান করে। |
* Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Evaluation Dataset
#### Unnamed Dataset
* Size: 48,465 evaluation samples
* Columns: query, positive, and negative
* Approximate statistics based on the first 1000 samples:
| | query | positive | negative |
|:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|
| type | string | string | string |
| details | - min: 5 tokens
- mean: 27.21 tokens
- max: 128 tokens
| - min: 4 tokens
- mean: 14.41 tokens
- max: 128 tokens
| - min: 5 tokens
- mean: 14.52 tokens
- max: 61 tokens
|
* Samples:
| query | positive | negative |
|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------|:-------------------------------------------------------------|
| काउबॉय टोपी और जींस पहने आदमी लकड़ी की इमारत के सामने खड़ा है। | लाकडी इमारतीसमोर उभा असलेला एक माणूस. | ஒரு சூட் அணிந்த ஒரு மனிதர் தெருவைக் கடக்கிறார். |
| 7 The Malcolm Baldridge National Quality Award and the President's Quality Award are given to organizations for their overall achievements in quality and performance. | ఒక సంస్థ బాగా పనిచేస్తే, వారికి రెండు అవార్డులు లభిస్తాయి. | விட்டிங்டன் ஒரு வண்டியில் சவாரி செய்தார். |
| ఈ ఫుట్బాల్ జట్టు ఎరుపు చొక్కాలు మరియు ఎరుపు శిరస్త్రాణాలు ధరిస్తుంది. | তারা ফুটবল খেলছে। | एक आदमी मेट्रो में है। |
* Loss: [MultipleNegativesRankingLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
```json
{
"scale": 20.0,
"similarity_fct": "cos_sim"
}
```
### Training Hyperparameters
#### Non-Default Hyperparameters
- `eval_strategy`: steps
- `per_device_train_batch_size`: 128
- `per_device_eval_batch_size`: 128
- `num_train_epochs`: 1
- `warmup_ratio`: 0.1
- `bf16`: True
- `batch_sampler`: no_duplicates
#### All Hyperparameters
Click to expand
- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 128
- `per_device_eval_batch_size`: 128
- `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`: 1.0
- `num_train_epochs`: 1
- `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