Category Classification on Zappos50k in-domain (test)
97.17Top-1 AccMSCon
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MSConLoss=Multi-Similarity Contrastive, Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 97.17 | 0.27 | |
| XEnt MTLoss=Cross-Entropy (Multi-task), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 96.98 | 0.29 | |
| SupCon CatLoss=SupCon (Category), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 96.95 | 0.29 | |
| XEnt CatLoss=Cross-Entropy (Category), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 96.64 | 0.34 | |
| SimCLRLoss=SimCLR, Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 90.05 | 0.43 | |
| XEnt CloLoss=Cross-Entropy (Closure), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 88.99 | 0.33 | |
| SupCon CloLoss=SupCon (Closure), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 83.62 | 0.3 | |
| CSNLoss=Conditional Similarity Network, Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 83.33 | 0.32 | |
| XEnt GendLoss=Cross-Entropy (Gender), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 81.96 | 0.32 | |
| SupCon GenLoss=SupCon (Gender), Evaluation Protocol=Fine-tune classification layer on frozen embedding2023.07 | 76.4 | 0.28 |