Hugging Face's logo

Marqo
/
marqo-fashionSigLIP

Zero-Shot Image Classification
OpenCLIP
ONNX
Safetensors
Transformers.js
English
siglip
clip
e-commerce
fashion
multimodal retrieval
custom_code
Model card Files Community
1
DavidJung commited on
Commit
c141ae8
•
1 Parent(s): 35a2c37

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -63,8 +63,8 @@ Average evaluation results on 6 public multimodal fashion datasets ([Atlas](http
63
  **Text-To-Image (Averaged across 6 datasets)**
64
  | Model | AvgRecall | Recall@1 | Recall@10 | MRR |
65
  |----------------------------|-------------|------------|-------------|-----------|
66
- | FashionCLIP2.0 | 0.163 | 0.077 | 0.249 | 0.165 |
67
  | Marqo-FashionSigLIP | **0.231** | **0.121** | **0.340** | **0.239** |
 
68
  | OpenFashionCLIP | 0.132 | 0.060 | 0.204 | 0.135 |
69
  | ViT-B-16-laion2b_s34b_b88k | 0.174 | 0.088 | 0.261 | 0.180 |
70
  | ViT-B-16-SigLIP-webli | 0.212 | 0.111 | 0.314 | 0.214 |
@@ -72,8 +72,8 @@ Average evaluation results on 6 public multimodal fashion datasets ([Atlas](http
72
  **Category-To-Product (Averaged across 5 datasets)**
73
  | Model | AvgP | P@1 | P@10 | MRR |
74
  |----------------------------|-----------|-----------|-----------|-----------|
75
- | FashionCLIP2.0 | 0.684 | 0.681 | 0.686 | 0.741 |
76
  | Marqo-FashionSigLIP | **0.737** | **0.758** | **0.716** | **0.812** |
 
77
  | OpenFashionCLIP | 0.646 | 0.653 | 0.639 | 0.720 |
78
  | ViT-B-16-laion2b_s34b_b88k | 0.662 | 0.673 | 0.652 | 0.743 |
79
  | ViT-B-16-SigLIP-webli | 0.688 | 0.690 | 0.685 | 0.751 |
@@ -81,8 +81,8 @@ Average evaluation results on 6 public multimodal fashion datasets ([Atlas](http
81
  **Sub-Category-To-Product (Averaged across 4 datasets)**
82
  | Model | AvgP | P@1 | P@10 | MRR |
83
  |----------------------------|-----------|-----------|-----------|-----------|
84
- | FashionCLIP2.0 | 0.657 | 0.676 | 0.638 | 0.733 |
85
  | Marqo-FashionSigLIP | **0.725** | **0.767** | **0.683** | **0.811** |
 
86
  | OpenFashionCLIP | 0.598 | 0.619 | 0.578 | 0.689 |
87
  | ViT-B-16-laion2b_s34b_b88k | 0.638 | 0.651 | 0.624 | 0.712 |
88
  | ViT-B-16-SigLIP-webli | 0.643 | 0.643 | 0.643 | 0.726 |
 
63
  **Text-To-Image (Averaged across 6 datasets)**
64
  | Model | AvgRecall | Recall@1 | Recall@10 | MRR |
65
  |----------------------------|-------------|------------|-------------|-----------|
 
66
  | Marqo-FashionSigLIP | **0.231** | **0.121** | **0.340** | **0.239** |
67
+ | FashionCLIP2.0 | 0.163 | 0.077 | 0.249 | 0.165 |
68
  | OpenFashionCLIP | 0.132 | 0.060 | 0.204 | 0.135 |
69
  | ViT-B-16-laion2b_s34b_b88k | 0.174 | 0.088 | 0.261 | 0.180 |
70
  | ViT-B-16-SigLIP-webli | 0.212 | 0.111 | 0.314 | 0.214 |
 
72
  **Category-To-Product (Averaged across 5 datasets)**
73
  | Model | AvgP | P@1 | P@10 | MRR |
74
  |----------------------------|-----------|-----------|-----------|-----------|
 
75
  | Marqo-FashionSigLIP | **0.737** | **0.758** | **0.716** | **0.812** |
76
+ | FashionCLIP2.0 | 0.684 | 0.681 | 0.686 | 0.741 |
77
  | OpenFashionCLIP | 0.646 | 0.653 | 0.639 | 0.720 |
78
  | ViT-B-16-laion2b_s34b_b88k | 0.662 | 0.673 | 0.652 | 0.743 |
79
  | ViT-B-16-SigLIP-webli | 0.688 | 0.690 | 0.685 | 0.751 |
 
81
  **Sub-Category-To-Product (Averaged across 4 datasets)**
82
  | Model | AvgP | P@1 | P@10 | MRR |
83
  |----------------------------|-----------|-----------|-----------|-----------|
 
84
  | Marqo-FashionSigLIP | **0.725** | **0.767** | **0.683** | **0.811** |
85
+ | FashionCLIP2.0 | 0.657 | 0.676 | 0.638 | 0.733 |
86
  | OpenFashionCLIP | 0.598 | 0.619 | 0.578 | 0.689 |
87
  | ViT-B-16-laion2b_s34b_b88k | 0.638 | 0.651 | 0.624 | 0.712 |
88
  | ViT-B-16-SigLIP-webli | 0.643 | 0.643 | 0.643 | 0.726 |