s-emanuilov commited on
Commit
65c75be
1 Parent(s): 4e2644a

Update README.md

Browse files

readme.md updates

Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -12,16 +12,16 @@ The CLIP model was developed by researchers at OpenAI to learn about what contri
12
  ## Model type
13
  The model uses a ViT-B/32 Transformer architecture as an image encoder and uses a masked self-attention Transformer as a text encoder. These encoders are trained to maximize the similarity of (image, text) pairs via a contrastive loss.
14
 
15
- ## OpenVINO optimization
16
 
17
  To increase the efficiency of the model during inference, we utilized the OpenVINO™ toolkit for optimization. The table below showcases the inference time improvements achieved with OpenVINO™ compared to the original PyTorch implementation:
18
 
19
- | Metric | PyTorch Inference Time (seconds) | OpenVINO Inference Time (seconds) | Similarity |
20
- |:-------------------|-----------------------------------:|------------------------------------:|-------------:|
21
- | Mean | 0.518564 | 0.461107 | 1 |
22
- | Standard Deviation | 0.109119 | 0.0917191 | 0 |
23
- | Min | 0.390102 | 0.360006 | 1 |
24
- | Max | 0.699677 | 0.620042 | 1 |
25
 
26
  The results indicate that the OpenVINO™ optimization provides a consistent improvement in inference time while maintaining the same level of accuracy (as indicated by the similarity score).
27
 
 
12
  ## Model type
13
  The model uses a ViT-B/32 Transformer architecture as an image encoder and uses a masked self-attention Transformer as a text encoder. These encoders are trained to maximize the similarity of (image, text) pairs via a contrastive loss.
14
 
15
+ ## OpenVINO optimization
16
 
17
  To increase the efficiency of the model during inference, we utilized the OpenVINO™ toolkit for optimization. The table below showcases the inference time improvements achieved with OpenVINO™ compared to the original PyTorch implementation:
18
 
19
+ | Metric | PyTorch Inference Time (sec) | OpenVINO Inference Time (sec) | Similarity |
20
+ |:-------------------|-------------------------------:|---------------------------------:|-------------:|
21
+ | mean | 0.52 | 0.46 | 1 |
22
+ | std | 0.11 | 0.09 | 0 |
23
+ | min | 0.39 | 0.36 | 1 |
24
+ | max | 0.70 | 0.62 | 1 |
25
 
26
  The results indicate that the OpenVINO™ optimization provides a consistent improvement in inference time while maintaining the same level of accuracy (as indicated by the similarity score).
27