--- tags: - sentence-transformers - sentence-similarity - feature-extraction - dense - generated_from_trainer - dataset_size:2184 - loss:BatchAllTripletLoss base_model: kiarashmo/chembberta-77m-mlm-safetensors widget: - source_sentence: CC(C)CNCc1ccc(-c2ccccc2S(=O)(=O)N2CCCC2)cc1 sentences: - C=C1CC2CCC34CC5OC6C(OC7CCC(CC(=O)CC8C(CC9OC(CCC1O2)CC(C)C9=C)OC(CC(O)CN)C8OC)OC7C6O3)C5O4 - C[NH+](C)CCC=C1c2ccccc2CCc2ccccc21 - CC(C)Cn1cnc2c(N)nc3ccccc3c21 - source_sentence: COC(=O)NC(C(=O)NC(Cc1ccccc1)C(O)CN(Cc1ccc(-c2ccccn2)cc1)NC(=O)C(NC(=O)OC)C(C)(C)C)C(C)(C)C sentences: - C=C1CC2CCC34CC5OC6C(OC7CCC(CC(=O)CC8C(CC9OC(CCC1O2)CC(C)C9=C)OC(CC(O)CN)C8OC)OC7C6O3)C5O4 - C[NH+]1CCCC(CC2c3ccccc3Sc3ccccc32)C1 - C[NH2+]C1(c2ccccc2Cl)CCCCC1=O - source_sentence: C[NH+]1CC(C(=O)NC2(C)OC3(O)C4CCCN4C(=O)C(Cc4ccccc4)N3C2=O)CC2c3cccc4[nH]cc(c34)CC21 sentences: - C[NH+](C)CCC=C1c2ccccc2COc2ccc(CC(=O)[O-])cc21 - C[NH+]1CCC(=C2c3ccccc3CCn3c(C=O)c[nH+]c32)CC1 - COC(=O)NC(C(=O)NC(Cc1ccccc1)C(O)CN(Cc1ccc(-c2ccccn2)cc1)NC(=O)C(NC(=O)OC)C(C)(C)C)C(C)(C)C - source_sentence: C[NH2+]CCCC12CCC(c3ccccc31)c1ccccc12 sentences: - C[N+]1(C)CCC(=C(c2ccccc2)c2ccccc2)CC1 - CC(CN1CC(=O)NC(=O)C1)[NH+]1CC(=O)NC(=O)C1 - C[NH+](C)CCc1c[nH]c2ccc(CC3COC(=O)N3)cc12 - source_sentence: CC(C)CNCc1ccc(-c2ccccc2S(=O)(=O)N2CCCC2)cc1 sentences: - COC(=O)NC(C(=O)NC(Cc1ccccc1)C(O)CN(Cc1ccc(-c2ccccn2)cc1)NC(=O)C(NC(=O)OC)C(C)(C)C)C(C)(C)C - COc1ccc(C(=O)CC(=O)c2ccc(C(C)(C)C)cc2)cc1 - COC1CC(OC2C(C)C(=O)OC(C)C(C)C(OC(C)=O)C(C)C(=O)C3(CO3)CC(C)C(OC3OC(C)CC([NH+](C)C)C3OC(C)=O)C2C)OC(C)C1OC(C)=O pipeline_tag: sentence-similarity library_name: sentence-transformers metrics: - cosine_accuracy - cosine_accuracy_threshold - cosine_f1 - cosine_f1_threshold - cosine_precision - cosine_recall - cosine_ap - cosine_mcc model-index: - name: SentenceTransformer based on kiarashmo/chembberta-77m-mlm-safetensors results: - task: type: binary-classification name: Binary Classification dataset: name: val sim type: val-sim metrics: - type: cosine_accuracy value: 0.611 name: Cosine Accuracy - type: cosine_accuracy_threshold value: 0.8879227638244629 name: Cosine Accuracy Threshold - type: cosine_f1 value: 0.6980609418282548 name: Cosine F1 - type: cosine_f1_threshold value: -0.5465683937072754 name: Cosine F1 Threshold - type: cosine_precision value: 0.5436893203883495 name: Cosine Precision - type: cosine_recall value: 0.9748549323017408 name: Cosine Recall - type: cosine_ap value: 0.6971622829878537 name: Cosine Ap - type: cosine_mcc value: 0.19032555952847827 name: Cosine Mcc --- # SentenceTransformer based on kiarashmo/chembberta-77m-mlm-safetensors This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [kiarashmo/chembberta-77m-mlm-safetensors](https://huggingface.co/kiarashmo/chembberta-77m-mlm-safetensors). 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:** [kiarashmo/chembberta-77m-mlm-safetensors](https://huggingface.co/kiarashmo/chembberta-77m-mlm-safetensors) - **Maximum Sequence Length:** 512 tokens - **Output Dimensionality:** 384 dimensions - **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': 512, 'do_lower_case': False, 'architecture': 'RobertaModel'}) (1): Pooling({'word_embedding_dimension': 384, '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 = [ 'CC(C)CNCc1ccc(-c2ccccc2S(=O)(=O)N2CCCC2)cc1', 'COC(=O)NC(C(=O)NC(Cc1ccccc1)C(O)CN(Cc1ccc(-c2ccccn2)cc1)NC(=O)C(NC(=O)OC)C(C)(C)C)C(C)(C)C', 'COc1ccc(C(=O)CC(=O)c2ccc(C(C)(C)C)cc2)cc1', ] embeddings = model.encode(sentences) print(embeddings.shape) # [3, 384] # Get the similarity scores for the embeddings similarities = model.similarity(embeddings, embeddings) print(similarities) # tensor([[ 1.0000, 0.8293, -0.3326], # [ 0.8293, 1.0000, -0.0993], # [-0.3326, -0.0993, 1.0000]]) ``` ## Evaluation ### Metrics #### Binary Classification * Dataset: `val-sim` * Evaluated with [BinaryClassificationEvaluator](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.BinaryClassificationEvaluator) | Metric | Value | |:--------------------------|:-----------| | cosine_accuracy | 0.611 | | cosine_accuracy_threshold | 0.8879 | | cosine_f1 | 0.6981 | | cosine_f1_threshold | -0.5466 | | cosine_precision | 0.5437 | | cosine_recall | 0.9749 | | **cosine_ap** | **0.6972** | | cosine_mcc | 0.1903 | ## Training Details ### Training Dataset #### Unnamed Dataset * Size: 2,184 training samples * Columns: text and label * Approximate statistics based on the first 1000 samples: | | text | label | |:--------|:-----------------------------------------------------------------------------------|:-----------------------------------------------| | type | string | int | | details | | | * Samples: | text | label | |:------------------------------------------------------------------------|:---------------| | CC(C)CC(NC(=O)CNC(=O)c1cc(Cl)ccc1Cl)B(O)O | 1 | | O=C(NCC(O)CO)c1c(I)c(C(=O)NCC(O)CO)c(I)c(N(CCO)C(=O)CO)c1I | 0 | | Clc1cc(Cl)c(OCC#CI)cc1Cl | 0 | * Loss: [BatchAllTripletLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#batchalltripletloss) ### Evaluation Dataset #### Unnamed Dataset * Size: 282 evaluation samples * Columns: text and label * Approximate statistics based on the first 282 samples: | | text | label | |:--------|:------------------------------------------------------------------------------------|:------------------------------------------------| | type | string | int | | details | | | * Samples: | text | label | |:-------------------------------------------------------------------------------|:---------------| | CC(C)CNCc1ccc(-c2ccccc2S(=O)(=O)N2CCCC2)cc1 | 1 | | CC(C)Cn1cnc2c(N)nc3ccccc3c21 | 0 | | CC(C)CN(CC(O)C(Cc1ccccc1)NC(=O)OC1COC2OCCC12)S(=O)(=O)c1ccc(N)cc1 | 0 | * Loss: [BatchAllTripletLoss](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#batchalltripletloss) ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 32 - `per_device_eval_batch_size`: 32 - `num_train_epochs`: 100 - `warmup_steps`: 100 - `load_best_model_at_end`: True #### All Hyperparameters
Click to expand - `overwrite_output_dir`: False - `do_predict`: False - `eval_strategy`: steps - `prediction_loss_only`: True - `per_device_train_batch_size`: 32 - `per_device_eval_batch_size`: 32 - `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`: 100 - `max_steps`: -1 - `lr_scheduler_type`: linear - `lr_scheduler_kwargs`: {} - `warmup_ratio`: 0.0 - `warmup_steps`: 100 - `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`: True - `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`: None - `hub_always_push`: False - `hub_revision`: None - `gradient_checkpointing`: False - `gradient_checkpointing_kwargs`: None - `include_inputs_for_metrics`: False - `include_for_metrics`: [] - `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 - `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 - `use_liger_kernel`: False - `liger_kernel_config`: None - `eval_use_gather_object`: False - `average_tokens_across_devices`: False - `prompts`: None - `batch_sampler`: batch_sampler - `multi_dataset_batch_sampler`: proportional - `router_mapping`: {} - `learning_rate_mapping`: {}
### Training Logs | Epoch | Step | Training Loss | Validation Loss | val-sim_cosine_ap | |:------:|:----:|:-------------:|:---------------:|:-----------------:| | 7.2464 | 500 | 4.0383 | 5.2239 | 0.6972 | ### Framework Versions - Python: 3.9.23 - Sentence Transformers: 5.0.0 - Transformers: 4.53.3 - PyTorch: 2.5.0+cu118 - Accelerate: 1.9.0 - Datasets: 4.0.0 - Tokenizers: 0.21.2 ## Citation ### BibTeX #### Sentence Transformers ```bibtex @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 ```bibtex @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} } ```