Proposed SMILES-Mamba, a two-stage model that leverages both unlabeled and labeled data through a combination of self-supervised pretraining and fine-tuning strategies. SMILES-Mamba exhibits competitive performance across 22 ADMET datasets, highlighting the potential of self-supervised learning in improving molecular property prediction.

High Price ModelEvaluated and ensured the competitiveness of products’ pricing on the PDD platform with other online platforms,including Alibaba.Increased prediction accuracy of the high-price probability by vectorizing text and image data. Improved the overall AUC by about 0.5%, compared with existing model without text and image data.Fair Price ModelMarkedtheproductsontheplatformwithafairpriceasareferenceforpricingandadvertising.Improved the accuracy of similar-item pairs… Show more High Price ModelEvaluated and ensured the competitiveness of products’ pricing on the PDD platform with other online platforms,including Alibaba.Increased prediction accuracy of the high-price probability by vectorizing text and image data. Improved the overall AUC by about 0.5%, compared with existing model without text and image data.Fair Price ModelMarkedtheproductsontheplatformwithafairpriceasareferenceforpricingandadvertising.Improved the accuracy of similar-item pairs by deleting and supplementing the rough pairs and calculated the fair prices by the mean price of selected similar items. Show less

Interrogated drug resistance at the single-cell level for target through source data.Source data had huge differences from the target data in absolute numerical size and measurement dimensions, but the two were intrinsically related. [Pytorch, NumPy, Scikit-learn, Pandas]Applied and finetuned Dynamic Adversarial Adaptation Network and achieved overall F1 score 0.9682, accuracy 0.9680, and average precision 0.9704.