--- tags: - ColBERT - PyLate - sentence-transformers - sentence-similarity - feature-extraction - generated_from_trainer - dataset_size:10000000 - loss:Contrastive base_model: answerdotai/ModernBERT-base datasets: - bclavie/msmarco-10m-triplets pipeline_tag: sentence-similarity library_name: PyLate metrics: - MaxSim_accuracy@1 - MaxSim_accuracy@3 - MaxSim_accuracy@5 - MaxSim_accuracy@10 - MaxSim_precision@1 - MaxSim_precision@3 - MaxSim_precision@5 - MaxSim_precision@10 - MaxSim_recall@1 - MaxSim_recall@3 - MaxSim_recall@5 - MaxSim_recall@10 - MaxSim_ndcg@10 - MaxSim_mrr@10 - MaxSim_map@100 model-index: - name: PyLate model based on answerdotai/ModernBERT-base results: - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoClimateFEVER type: NanoClimateFEVER metrics: - type: MaxSim_accuracy@1 value: 0.24 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.5 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.56 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.72 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.24 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.18 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.136 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.09 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.12 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.23166666666666663 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.26733333333333337 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.3456666666666666 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.28659074634439036 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.38899206349206344 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.22237332831909778 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoDBPedia type: NanoDBPedia metrics: - type: MaxSim_accuracy@1 value: 0.72 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.88 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.92 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.92 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.72 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.6333333333333333 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.608 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.514 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.08698943062762281 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.15785394286086044 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.25560775585402196 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.35140662490283503 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.6310474103595507 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.8023333333333333 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.4766459270446468 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoFEVER type: NanoFEVER metrics: - type: MaxSim_accuracy@1 value: 0.84 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.98 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 1.0 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 1.0 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.84 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.33333333333333326 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.20799999999999996 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.10599999999999998 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.7866666666666667 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.9266666666666667 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.9566666666666667 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.9633333333333333 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.8973997659405684 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.905 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.8618368368368369 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoFiQA2018 type: NanoFiQA2018 metrics: - type: MaxSim_accuracy@1 value: 0.46 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.66 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.74 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.78 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.46 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.32666666666666666 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.244 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.14 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.2685793650793651 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.47251587301587306 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.5722857142857143 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.6081190476190477 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.5290995192750131 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.579 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.4613817582717077 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoHotpotQA type: NanoHotpotQA metrics: - type: MaxSim_accuracy@1 value: 0.88 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.96 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.96 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.98 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.88 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.49999999999999983 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.32799999999999996 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.17199999999999996 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.44 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.75 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.82 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.86 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.8190215640428958 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.9195238095238095 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.7542864635927516 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoMSMARCO type: NanoMSMARCO metrics: - type: MaxSim_accuracy@1 value: 0.46 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.74 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.76 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.86 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.46 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.24666666666666665 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.15200000000000002 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.08599999999999998 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.46 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.74 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.76 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.86 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.6683749348060735 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.6068333333333333 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.6153476502857308 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoNFCorpus type: NanoNFCorpus metrics: - type: MaxSim_accuracy@1 value: 0.46 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.52 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.54 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.56 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.46 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.3533333333333333 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.29600000000000004 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.23600000000000002 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.043124364328590487 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.07617110540472594 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.0888523411805491 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.10700228981811344 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.30567453687850876 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.494 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.13042780558280634 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoNQ type: NanoNQ metrics: - type: MaxSim_accuracy@1 value: 0.38 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.66 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.78 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.9 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.38 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.22 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.15600000000000003 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.092 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.37 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.62 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.73 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.84 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.6117132890493197 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.5528888888888889 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.5368699134199134 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoQuoraRetrieval type: NanoQuoraRetrieval metrics: - type: MaxSim_accuracy@1 value: 0.9 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.98 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 1.0 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 1.0 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.9 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.3933333333333333 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.24799999999999997 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.13599999999999998 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.7973333333333333 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.9386666666666668 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.9593333333333334 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.9933333333333334 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.9500644303763519 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.945 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.9287460317460318 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoSCIDOCS type: NanoSCIDOCS metrics: - type: MaxSim_accuracy@1 value: 0.42 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.62 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.7 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.82 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.42 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.30666666666666664 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.24 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.172 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.08866666666666667 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.19066666666666662 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.24666666666666665 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.3526666666666666 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.34178587115776565 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.5458571428571429 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.254228684034207 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoArguAna type: NanoArguAna metrics: - type: MaxSim_accuracy@1 value: 0.14 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.48 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.66 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.82 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.14 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.15999999999999998 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.132 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.08199999999999999 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.14 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.48 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.66 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.82 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.46517990763149614 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.3529603174603174 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.3567092878593359 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoSciFact type: NanoSciFact metrics: - type: MaxSim_accuracy@1 value: 0.66 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.76 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.8 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.84 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.66 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.2733333333333333 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.176 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.09399999999999999 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.625 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.745 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.79 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.83 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.7435557624884191 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.720888888888889 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.7171289121218996 name: Maxsim Map@100 - task: type: py-late-information-retrieval name: Py Late Information Retrieval dataset: name: NanoTouche2020 type: NanoTouche2020 metrics: - type: MaxSim_accuracy@1 value: 0.7346938775510204 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.9591836734693877 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.9591836734693877 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.9591836734693877 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.7346938775510204 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.7006802721088434 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.6285714285714286 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.5122448979591837 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.05035830341896132 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.14458823716252048 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.20896761189253432 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.3315966017926261 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.5870329612095947 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.8401360544217688 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.4348641866623347 name: Maxsim Map@100 - task: type: nano-beir name: Nano BEIR dataset: name: NanoBEIR mean type: NanoBEIR_mean metrics: - type: MaxSim_accuracy@1 value: 0.5611302982731554 name: Maxsim Accuracy@1 - type: MaxSim_accuracy@3 value: 0.7460910518053375 name: Maxsim Accuracy@3 - type: MaxSim_accuracy@5 value: 0.7983987441130298 name: Maxsim Accuracy@5 - type: MaxSim_accuracy@10 value: 0.8583987441130299 name: Maxsim Accuracy@10 - type: MaxSim_precision@1 value: 0.5611302982731554 name: Maxsim Precision@1 - type: MaxSim_precision@3 value: 0.3559497645211931 name: Maxsim Precision@3 - type: MaxSim_precision@5 value: 0.2732747252747253 name: Maxsim Precision@5 - type: MaxSim_precision@10 value: 0.18709576138147568 name: Maxsim Precision@10 - type: MaxSim_recall@1 value: 0.3289783177016312 name: Maxsim Recall@1 - type: MaxSim_recall@3 value: 0.4979842942392806 name: Maxsim Recall@3 - type: MaxSim_recall@5 value: 0.5627471864009862 name: Maxsim Recall@5 - type: MaxSim_recall@10 value: 0.6356249664717402 name: Maxsim Recall@10 - type: MaxSim_ndcg@10 value: 0.6028108230430728 name: Maxsim Ndcg@10 - type: MaxSim_mrr@10 value: 0.6656472178615036 name: Maxsim Mrr@10 - type: MaxSim_map@100 value: 0.519295906598254 name: Maxsim Map@100 --- # PyLate model based on answerdotai/ModernBERT-base This is a [PyLate](https://github.com/lightonai/pylate) model finetuned from [answerdotai/ModernBERT-base](https://huggingface.co/answerdotai/ModernBERT-base) on the [msmarco-10m-triplets](https://huggingface.co/datasets/bclavie/msmarco-10m-triplets) dataset. It maps sentences & paragraphs to sequences of 128-dimensional dense vectors and can be used for semantic textual similarity using the MaxSim operator. ## Model Details ### Model Description - **Model Type:** PyLate model - **Base model:** [answerdotai/ModernBERT-base](https://huggingface.co/answerdotai/ModernBERT-base) - **Document Length:** 512 tokens - **Query Length:** 32 tokens - **Output Dimensionality:** 128 tokens - **Similarity Function:** MaxSim - **Training Dataset:** - [msmarco-10m-triplets](https://huggingface.co/datasets/bclavie/msmarco-10m-triplets) ### Model Sources - **Documentation:** [PyLate Documentation](https://lightonai.github.io/pylate/) - **Repository:** [PyLate on GitHub](https://github.com/lightonai/pylate) - **Hugging Face:** [PyLate models on Hugging Face](https://huggingface.co/models?library=PyLate) ### Full Model Architecture ``` ColBERT( (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'ModernBertModel'}) (1): Dense({'in_features': 768, 'out_features': 128, 'bias': False, 'activation_function': 'torch.nn.modules.linear.Identity', 'use_residual': False}) ) ``` ## Usage First install the PyLate library: ```bash pip install -U pylate ``` ### Retrieval Use this model with PyLate to index and retrieve documents. The index uses [FastPLAID](https://github.com/lightonai/fast-plaid) for efficient similarity search. #### Indexing documents Load the ColBERT model and initialize the PLAID index, then encode and index your documents: ```python from pylate import indexes, models, retrieve # Step 1: Load the ColBERT model model = models.ColBERT( model_name_or_path="pylate_model_id", ) # Step 2: Initialize the PLAID index index = indexes.PLAID( index_folder="pylate-index", index_name="index", override=True, # This overwrites the existing index if any ) # Step 3: Encode the documents documents_ids = ["1", "2", "3"] documents = ["document 1 text", "document 2 text", "document 3 text"] documents_embeddings = model.encode( documents, batch_size=32, is_query=False, # Ensure that it is set to False to indicate that these are documents, not queries show_progress_bar=True, ) # Step 4: Add document embeddings to the index by providing embeddings and corresponding ids index.add_documents( documents_ids=documents_ids, documents_embeddings=documents_embeddings, ) ``` Note that you do not have to recreate the index and encode the documents every time. Once you have created an index and added the documents, you can re-use the index later by loading it: ```python # To load an index, simply instantiate it with the correct folder/name and without overriding it index = indexes.PLAID( index_folder="pylate-index", index_name="index", ) ``` #### Retrieving top-k documents for queries Once the documents are indexed, you can retrieve the top-k most relevant documents for a given set of queries. To do so, initialize the ColBERT retriever with the index you want to search in, encode the queries and then retrieve the top-k documents to get the top matches ids and relevance scores: ```python # Step 1: Initialize the ColBERT retriever retriever = retrieve.ColBERT(index=index) # Step 2: Encode the queries queries_embeddings = model.encode( ["query for document 3", "query for document 1"], batch_size=32, is_query=True, # # Ensure that it is set to False to indicate that these are queries show_progress_bar=True, ) # Step 3: Retrieve top-k documents scores = retriever.retrieve( queries_embeddings=queries_embeddings, k=10, # Retrieve the top 10 matches for each query ) ``` ### Reranking If you only want to use the ColBERT model to perform reranking on top of your first-stage retrieval pipeline without building an index, you can simply use rank function and pass the queries and documents to rerank: ```python from pylate import rank, models queries = [ "query A", "query B", ] documents = [ ["document A", "document B"], ["document 1", "document C", "document B"], ] documents_ids = [ [1, 2], [1, 3, 2], ] model = models.ColBERT( model_name_or_path="pylate_model_id", ) queries_embeddings = model.encode( queries, is_query=True, ) documents_embeddings = model.encode( documents, is_query=False, ) reranked_documents = rank.rerank( documents_ids=documents_ids, queries_embeddings=queries_embeddings, documents_embeddings=documents_embeddings, ) ``` ## Evaluation ### Metrics #### Py Late Information Retrieval * Dataset: `['NanoClimateFEVER', 'NanoDBPedia', 'NanoFEVER', 'NanoFiQA2018', 'NanoHotpotQA', 'NanoMSMARCO', 'NanoNFCorpus', 'NanoNQ', 'NanoQuoraRetrieval', 'NanoSCIDOCS', 'NanoArguAna', 'NanoSciFact', 'NanoTouche2020']` * Evaluated with pylate.evaluation.pylate_information_retrieval_evaluator.PyLateInformationRetrievalEvaluator | Metric | NanoClimateFEVER | NanoDBPedia | NanoFEVER | NanoFiQA2018 | NanoHotpotQA | NanoMSMARCO | NanoNFCorpus | NanoNQ | NanoQuoraRetrieval | NanoSCIDOCS | NanoArguAna | NanoSciFact | NanoTouche2020 | |:--------------------|:-----------------|:------------|:-----------|:-------------|:-------------|:------------|:-------------|:-----------|:-------------------|:------------|:------------|:------------|:---------------| | MaxSim_accuracy@1 | 0.24 | 0.72 | 0.84 | 0.46 | 0.88 | 0.46 | 0.46 | 0.38 | 0.9 | 0.42 | 0.14 | 0.66 | 0.7347 | | MaxSim_accuracy@3 | 0.5 | 0.88 | 0.98 | 0.66 | 0.96 | 0.74 | 0.52 | 0.66 | 0.98 | 0.62 | 0.48 | 0.76 | 0.9592 | | MaxSim_accuracy@5 | 0.56 | 0.92 | 1.0 | 0.74 | 0.96 | 0.76 | 0.54 | 0.78 | 1.0 | 0.7 | 0.66 | 0.8 | 0.9592 | | MaxSim_accuracy@10 | 0.72 | 0.92 | 1.0 | 0.78 | 0.98 | 0.86 | 0.56 | 0.9 | 1.0 | 0.82 | 0.82 | 0.84 | 0.9592 | | MaxSim_precision@1 | 0.24 | 0.72 | 0.84 | 0.46 | 0.88 | 0.46 | 0.46 | 0.38 | 0.9 | 0.42 | 0.14 | 0.66 | 0.7347 | | MaxSim_precision@3 | 0.18 | 0.6333 | 0.3333 | 0.3267 | 0.5 | 0.2467 | 0.3533 | 0.22 | 0.3933 | 0.3067 | 0.16 | 0.2733 | 0.7007 | | MaxSim_precision@5 | 0.136 | 0.608 | 0.208 | 0.244 | 0.328 | 0.152 | 0.296 | 0.156 | 0.248 | 0.24 | 0.132 | 0.176 | 0.6286 | | MaxSim_precision@10 | 0.09 | 0.514 | 0.106 | 0.14 | 0.172 | 0.086 | 0.236 | 0.092 | 0.136 | 0.172 | 0.082 | 0.094 | 0.5122 | | MaxSim_recall@1 | 0.12 | 0.087 | 0.7867 | 0.2686 | 0.44 | 0.46 | 0.0431 | 0.37 | 0.7973 | 0.0887 | 0.14 | 0.625 | 0.0504 | | MaxSim_recall@3 | 0.2317 | 0.1579 | 0.9267 | 0.4725 | 0.75 | 0.74 | 0.0762 | 0.62 | 0.9387 | 0.1907 | 0.48 | 0.745 | 0.1446 | | MaxSim_recall@5 | 0.2673 | 0.2556 | 0.9567 | 0.5723 | 0.82 | 0.76 | 0.0889 | 0.73 | 0.9593 | 0.2467 | 0.66 | 0.79 | 0.209 | | MaxSim_recall@10 | 0.3457 | 0.3514 | 0.9633 | 0.6081 | 0.86 | 0.86 | 0.107 | 0.84 | 0.9933 | 0.3527 | 0.82 | 0.83 | 0.3316 | | **MaxSim_ndcg@10** | **0.2866** | **0.631** | **0.8974** | **0.5291** | **0.819** | **0.6684** | **0.3057** | **0.6117** | **0.9501** | **0.3418** | **0.4652** | **0.7436** | **0.587** | | MaxSim_mrr@10 | 0.389 | 0.8023 | 0.905 | 0.579 | 0.9195 | 0.6068 | 0.494 | 0.5529 | 0.945 | 0.5459 | 0.353 | 0.7209 | 0.8401 | | MaxSim_map@100 | 0.2224 | 0.4766 | 0.8618 | 0.4614 | 0.7543 | 0.6153 | 0.1304 | 0.5369 | 0.9287 | 0.2542 | 0.3567 | 0.7171 | 0.4349 | #### Nano BEIR * Dataset: `NanoBEIR_mean` * Evaluated with pylate.evaluation.nano_beir_evaluator.NanoBEIREvaluator | Metric | Value | |:--------------------|:-----------| | MaxSim_accuracy@1 | 0.5611 | | MaxSim_accuracy@3 | 0.7461 | | MaxSim_accuracy@5 | 0.7984 | | MaxSim_accuracy@10 | 0.8584 | | MaxSim_precision@1 | 0.5611 | | MaxSim_precision@3 | 0.3559 | | MaxSim_precision@5 | 0.2733 | | MaxSim_precision@10 | 0.1871 | | MaxSim_recall@1 | 0.329 | | MaxSim_recall@3 | 0.498 | | MaxSim_recall@5 | 0.5627 | | MaxSim_recall@10 | 0.6356 | | **MaxSim_ndcg@10** | **0.6028** | | MaxSim_mrr@10 | 0.6656 | | MaxSim_map@100 | 0.5193 | ## Training Details ### Training Dataset #### msmarco-10m-triplets * Dataset: [msmarco-10m-triplets](https://huggingface.co/datasets/bclavie/msmarco-10m-triplets) at [8c5139a](https://huggingface.co/datasets/bclavie/msmarco-10m-triplets/tree/8c5139a245a5997992605792faa49ec12a6eb5f2) * Size: 10,000,000 training samples * Columns: query, positive, and negative * Approximate statistics based on the first 1000 samples: | | query | positive | negative | |:--------|:---------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------| | type | string | string | string | | details | | | | * Samples: | query | positive | negative | |:-----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | the most important factor that influences k+ secretion is __________. | The regulation of K+ distribution between the intracellular and extracellular space is referred to as internal K+ balance. The most important factors regulating this movement under normal conditions are insulin and catecholamines (1). | They are both also important for secretion and flow of bile: 1 Cholecystokinin: The name of this hormone describes its effect on the biliary system-cholecysto = gallbladder and kinin = movement. 2 Secretin: This hormone is secreted in response to acid in the duodenum. | | how much did the mackinac bridge cost to build | The cost to design the project was $3,500,000 (Steinman Company). The cost to construct the bridge was $70, 268,500. Two primary contractors were hired to build the bridge: American Bridge for superstructure - $44,532,900; and Merritt-Chapman and Scott of New York for the foundations - $25,735,600. | When your child needs a dental tooth bridge, you need to know the average cost so you can factor the price into your budget. Several factors affect the price of a bridge, which can run between $700 to $1,500 per tooth. If you have insurance or your child is covered by Medicaid, part of the cost may be covered. | | when do concussion symptoms appear | Then you can get advice on what to do next. For milder symptoms, the doctor may recommend rest and ask you to watch your child closely for changes, such as a headache that gets worse. Symptoms of a concussion don't always show up right away, and can develop within 24 to 72 hours after an injury. | Concussion: A traumatic injury to soft tissue, usually the brain, as a result of a violent blow, shaking, or spinning. A brain concussion can cause immediate but temporary impairment of brain functions, such as thinking, vision, equilibrium, and consciousness. After a person has had a concussion, he or she is at increased risk for recurrence. Moreover, after a person has several concussions, less of a blow can cause injury, and the person can require more time to recover. | * Loss: pylate.losses.contrastive.Contrastive ### Training Hyperparameters #### Non-Default Hyperparameters - `eval_strategy`: steps - `per_device_train_batch_size`: 64 - `learning_rate`: 3e-05 - `max_steps`: 50000 - `fp16`: True - `dataloader_drop_last`: True - `dataloader_num_workers`: 8 - `ddp_find_unused_parameters`: False - `torch_compile`: True - `torch_compile_backend`: inductor - `eval_on_start`: 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`: 64 - `per_device_eval_batch_size`: 8 - `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`: 3e-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`: 3.0 - `max_steps`: 50000 - `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`: True - `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`: True - `dataloader_num_workers`: 8 - `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} - `parallelism_config`: 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`: False - `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`: True - `torch_compile_backend`: inductor - `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`: True - `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`: {}