SentenceTransformer based on distilbert/distilroberta-base
This is a sentence-transformers model finetuned from distilbert/distilroberta-base. 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: distilbert/distilroberta-base
- Maximum Sequence Length: 512 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': 512, 'do_lower_case': False}) with Transformer model: RobertaModel
(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:
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("iqrakiran/my-sentence-transformer-model")
# Run inference
sentences = [
"I remember generating alternative predictions. We said that I might get nervous but if I stay focused on what I want to do, I can probably do it. It's just hard to believe when I'm in this situation.",
"It's normal to feel nervous in difficult situations. As a counsellor, I suggest continuing to test these alternative predictions directly in real-life experiences. As you encounter situations where you manage to handle things even while nervous, your belief in your abilities and the predictions will strengthen. Remember to stay focused and trust yourself.",
'If you are a counsellor, please answer the questions based on the description of the patient.',
]
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: 4,093 training samples
- Columns:
sentence_0
,sentence_1
, andsentence_2
- Approximate statistics based on the first 1000 samples:
sentence_0 sentence_1 sentence_2 type string string string details - min: 11 tokens
- mean: 57.82 tokens
- max: 197 tokens
- min: 30 tokens
- mean: 76.95 tokens
- max: 149 tokens
- min: 22 tokens
- mean: 22.0 tokens
- max: 22 tokens
- Samples:
sentence_0 sentence_1 sentence_2 Lately, I've been feeling really stressed out and I keep having this thought that my partner might be cheating on me. It started when I noticed changes in their behavior a couple of months ago. At times, I feel like my partner acts distant, but when I confront them about it, they assure me that everything is fine between us. I feel like maybe it's not a strong enough thought to be classified as a delusion, but it's really bothering me and affecting our relationship.
Thank you for sharing your concerns and feelings with me. It's important to know that it's normal for people to experience doubts and insecurities in relationships sometimes. I understand that feeling stressed and having negative thoughts is difficult to deal with. If you're open to it, perhaps we could explore these thoughts and feelings further to better understand their origin and find ways to cope with them. Additionally, involving your partner in a constructive conversation about your feelings and concerns might help to improve your relationship and provide reassurance.
If you are a counsellor, please answer the questions based on the description of the patient.
I think counting would mean that somebody cares about me and I matter to them. In my life, for example, I think Ms. Jones is upset that I am not as involved or doing the things that she saw me do last year. So I guess I matter to her.
It's important to recognize that feeling valued and cared for by others can contribute to our overall well-being. It's great that you have someone like Ms. Jones in your life who notices your changes and seems to care about you. Can you tell me more about your relationship with Ms. Jones or the changes that have occurred in the past year?
If you are a counsellor, please answer the questions based on the description of the patient.
Lately, I've been feeling quite down, and it's been affecting my daily life. I've also been having a hard time concentrating and making decisions, and often find myself lost in thought. On top of that, I've started hearing voices that aren't there, but it only happens occasionally. It all started when I moved to a new city a few months ago. I feel so isolated, and the stress of adjusting to a new environment might be causing these issues.
Thank you for sharing your feelings and experiences with me. It sounds like you are struggling with some challenging emotions and experiences since your move to a new city. Adjusting to a new environment can be a significant source of stress. It's essential to acknowledge what you're going through and understand that seeking help is healthy. Can you provide further details about the voices you've been hearing? How often do they occur, and how exactly do they affect your daily life? Also, can you share your previous mental health history, if any? Your additional information will help me get a better understanding of your situation and provide the best support possible.
If you are a counsellor, please answer the questions based on the description of the patient.
- Loss:
TripletLoss
with these parameters:{ "distance_metric": "TripletDistanceMetric.EUCLIDEAN", "triplet_margin": 5 }
Training Hyperparameters
Non-Default Hyperparameters
per_device_train_batch_size
: 16per_device_eval_batch_size
: 16num_train_epochs
: 10multi_dataset_batch_sampler
: round_robin
All Hyperparameters
Click to expand
overwrite_output_dir
: Falsedo_predict
: Falseeval_strategy
: noprediction_loss_only
: Trueper_device_train_batch_size
: 16per_device_eval_batch_size
: 16per_gpu_train_batch_size
: Noneper_gpu_eval_batch_size
: Nonegradient_accumulation_steps
: 1eval_accumulation_steps
: Nonetorch_empty_cache_steps
: Nonelearning_rate
: 5e-05weight_decay
: 0.0adam_beta1
: 0.9adam_beta2
: 0.999adam_epsilon
: 1e-08max_grad_norm
: 1num_train_epochs
: 10max_steps
: -1lr_scheduler_type
: linearlr_scheduler_kwargs
: {}warmup_ratio
: 0.0warmup_steps
: 0log_level
: passivelog_level_replica
: warninglog_on_each_node
: Truelogging_nan_inf_filter
: Truesave_safetensors
: Truesave_on_each_node
: Falsesave_only_model
: Falserestore_callback_states_from_checkpoint
: Falseno_cuda
: Falseuse_cpu
: Falseuse_mps_device
: Falseseed
: 42data_seed
: Nonejit_mode_eval
: Falseuse_ipex
: Falsebf16
: Falsefp16
: Falsefp16_opt_level
: O1half_precision_backend
: autobf16_full_eval
: Falsefp16_full_eval
: Falsetf32
: Nonelocal_rank
: 0ddp_backend
: Nonetpu_num_cores
: Nonetpu_metrics_debug
: Falsedebug
: []dataloader_drop_last
: Falsedataloader_num_workers
: 0dataloader_prefetch_factor
: Nonepast_index
: -1disable_tqdm
: Falseremove_unused_columns
: Truelabel_names
: Noneload_best_model_at_end
: Falseignore_data_skip
: Falsefsdp
: []fsdp_min_num_params
: 0fsdp_config
: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}fsdp_transformer_layer_cls_to_wrap
: Noneaccelerator_config
: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}deepspeed
: Nonelabel_smoothing_factor
: 0.0optim
: adamw_torchoptim_args
: Noneadafactor
: Falsegroup_by_length
: Falselength_column_name
: lengthddp_find_unused_parameters
: Noneddp_bucket_cap_mb
: Noneddp_broadcast_buffers
: Falsedataloader_pin_memory
: Truedataloader_persistent_workers
: Falseskip_memory_metrics
: Trueuse_legacy_prediction_loop
: Falsepush_to_hub
: Falseresume_from_checkpoint
: Nonehub_model_id
: Nonehub_strategy
: every_savehub_private_repo
: Falsehub_always_push
: Falsegradient_checkpointing
: Falsegradient_checkpointing_kwargs
: Noneinclude_inputs_for_metrics
: Falseeval_do_concat_batches
: Truefp16_backend
: autopush_to_hub_model_id
: Nonepush_to_hub_organization
: Nonemp_parameters
:auto_find_batch_size
: Falsefull_determinism
: Falsetorchdynamo
: Noneray_scope
: lastddp_timeout
: 1800torch_compile
: Falsetorch_compile_backend
: Nonetorch_compile_mode
: Nonedispatch_batches
: Nonesplit_batches
: Noneinclude_tokens_per_second
: Falseinclude_num_input_tokens_seen
: Falseneftune_noise_alpha
: Noneoptim_target_modules
: Nonebatch_eval_metrics
: Falseeval_on_start
: Falseeval_use_gather_object
: Falsebatch_sampler
: batch_samplermulti_dataset_batch_sampler
: round_robin
Training Logs
Epoch | Step | Training Loss |
---|---|---|
1.9531 | 500 | 0.2321 |
3.9062 | 1000 | 0.0 |
5.8594 | 1500 | 0.0001 |
7.8125 | 2000 | 0.0 |
9.7656 | 2500 | 0.0 |
Framework Versions
- Python: 3.10.12
- Sentence Transformers: 3.0.1
- Transformers: 4.44.2
- PyTorch: 2.4.0+cu121
- Accelerate: 0.33.0
- Datasets: 2.21.0
- 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",
}
TripletLoss
@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}
}
- Downloads last month
- 4
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for iqrakiran/my-sentence-transformer-model
Base model
distilbert/distilroberta-base