Ng)dddlmZddlmZddlmZmZddlmZm Z Gddej Z dS)) annotations)Iterable)Tensornn)SentenceTransformerutilcNeZdZejfd fd Zdd Zedd ZxZ S) MarginMSELossmodelrreturnNonect||_||_t j|_dS)a Compute the MSE loss between the ``|sim(Query, Pos) - sim(Query, Neg)|`` and ``|gold_sim(Query, Pos) - gold_sim(Query, Neg)|``. By default, sim() is the dot-product. The gold_sim is often the similarity score from a teacher model. In contrast to :class:`MultipleNegativesRankingLoss`, the two passages do not have to be strictly positive and negative, both can be relevant or not relevant for a given query. This can be an advantage of MarginMSELoss over MultipleNegativesRankingLoss, but note that the MarginMSELoss is much slower to train. With MultipleNegativesRankingLoss, with a batch size of 64, we compare one query against 128 passages. With MarginMSELoss, we compare a query only against two passages. Args: model: SentenceTransformerModel similarity_fct: Which similarity function to use. References: - For more details, please refer to https://arxiv.org/abs/2010.02666. - `Training Examples > MS MARCO <../../examples/training/ms_marco/README.html>`_ - `Unsupervised Learning > Domain Adaptation <../../examples/domain_adaptation/README.html>`_ Requirements: 1. (query, passage_one, passage_two) triplets 2. Usually used with a finetuned teacher M in a knowledge distillation setup Inputs: +-----------------------------------------------+-----------------------------------------------+ | Texts | Labels | +===============================================+===============================================+ | (query, passage_one, passage_two) triplets | M(query, passage_one) - M(query, passage_two) | +-----------------------------------------------+-----------------------------------------------+ Relations: - :class:`MSELoss` is equivalent to this loss, but without a margin through the negative pair. Example: With gold labels, e.g. if you have hard scores for sentences. Imagine you want a model to embed sentences with similar "quality" close to each other. If the "text1" has quality 5 out of 5, "text2" has quality 1 out of 5, and "text3" has quality 3 out of 5, then the similarity of a pair can be defined as the difference of the quality scores. So, the similarity between "text1" and "text2" is 4, and the similarity between "text1" and "text3" is 2. If we use this as our "Teacher Model", the label becomes similraity("text1", "text2") - similarity("text1", "text3") = 4 - 2 = 2. Positive values denote that the first passage is more similar to the query than the second passage, while negative values denote the opposite. :: from sentence_transformers import SentenceTransformer, SentenceTransformerTrainer, losses from datasets import Dataset model = SentenceTransformer("microsoft/mpnet-base") train_dataset = Dataset.from_dict({ "text1": ["It's nice weather outside today.", "He drove to work."], "text2": ["It's so sunny.", "He took the car to work."], "text3": ["It's very sunny.", "She walked to the store."], "label": [0.1, 0.8], }) loss = losses.MarginMSELoss(model) trainer = SentenceTransformerTrainer( model=model, train_dataset=train_dataset, loss=loss, ) trainer.train() We can also use a teacher model to compute the similarity scores. In this case, we can use the teacher model to compute the similarity scores and use them as the silver labels. This is often used in knowledge distillation. :: from sentence_transformers import SentenceTransformer, SentenceTransformerTrainer, losses from datasets import Dataset student_model = SentenceTransformer("microsoft/mpnet-base") teacher_model = SentenceTransformer("all-mpnet-base-v2") train_dataset = Dataset.from_dict({ "query": ["It's nice weather outside today.", "He drove to work."], "passage1": ["It's so sunny.", "He took the car to work."], "passage2": ["It's very sunny.", "She walked to the store."], }) def compute_labels(batch): emb_queries = teacher_model.encode(batch["query"]) emb_passages1 = teacher_model.encode(batch["passage1"]) emb_passages2 = teacher_model.encode(batch["passage2"]) return { "label": teacher_model.similarity_pairwise(emb_queries, emb_passages1) - teacher_model.similarity_pairwise(emb_queries, emb_passages2) } train_dataset = train_dataset.map(compute_labels, batched=True) # In this example, the labels become -0.036 and 0.68, respectively loss = losses.MarginMSELoss(student_model) trainer = SentenceTransformerTrainer( model=student_model, train_dataset=train_dataset, loss=loss, ) trainer.train() N)super__init__r similarity_fctrMSELossloss_fct)selfr r __class__s f/var/www/html/ai-engine/env/lib/python3.11/site-packages/sentence_transformers/losses/MarginMSELoss.pyrzMarginMSELoss.__init__ s<L  ,  sentence_featuresIterable[dict[str, Tensor]]labelsrcfd|D}|d}|d}|d}||}||}||z } | |S)NcFg|]}|dS)sentence_embedding)r ).0sentence_featurers r z)MarginMSELoss.forward..xs-mmmGW +,,-ABmmmrr)rr) rrrrepsembeddings_queryembeddings_posembeddings_neg scores_pos scores_neg margin_preds ` rforwardzMarginMSELoss.forwardvsmmmm[lmmm7aa(()9>JJ (()9>JJ  :- }}[&111rstrcdS)NuY @misc{hofstätter2021improving, title={Improving Efficient Neural Ranking Models with Cross-Architecture Knowledge Distillation}, author={Sebastian Hofstätter and Sophia Althammer and Michael Schröder and Mete Sertkan and Allan Hanbury}, year={2021}, eprint={2010.02666}, archivePrefix={arXiv}, primaryClass={cs.IR} } )rs rcitationzMarginMSELoss.citations   r)r rr r )rrrrr r)r r+) __name__ __module__ __qualname__rpairwise_dot_scorerr*propertyr. __classcell__)rs@rr r sBFBYi%i%i%i%i%i%i%V 2 2 2 2   X     rr N) __future__rcollections.abcrtorchrrsentence_transformersrrModuler r-rrr:s""""""$$$$$$;;;;;;;;DDDDDBIDDDDDr