dima806
/

hand_gestures_image_detection

Image Classification

Model card Files Files and versions

dima806 commited on May 6, 2024

Commit

a894d50

·

verified ·

1 Parent(s): 0a88e50

Update README.md

Files changed (1) hide show

README.md +22 -22

README.md CHANGED Viewed

@@ -8,33 +8,33 @@ Returns hand gesture based on image.
 See https://www.kaggle.com/code/dima806/hand-gestures-image-detection-vit for more details.
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/6449300e3adf50d864095b90/gkYmH05qTQvdPG-KYbLht.png)
 ```
 Classification report:
                  precision    recall  f1-score   support
-           call     0.9090    0.9455    0.9269     11825
-        dislike     0.9706    0.9759    0.9732     11825
-           fist     0.9760    0.9871    0.9815     11826
-           four     0.8818    0.8880    0.8849     11826
-           like     0.9479    0.8905    0.9183     11825
-           mute     0.9751    0.9966    0.9857     11826
-             ok     0.9356    0.9509    0.9432     11825
-            one     0.9582    0.9335    0.9457     11826
-           palm     0.9659    0.9494    0.9576     11825
-          peace     0.8914    0.9116    0.9014     11826
- peace_inverted     0.9725    0.9646    0.9685     11825
-           rock     0.9265    0.8987    0.9124     11826
-           stop     0.9338    0.9690    0.9510     11826
-  stop_inverted     0.9790    0.9400    0.9591     11826
-          three     0.8701    0.8742    0.8721     11826
-         three2     0.9631    0.9556    0.9593     11826
-         two_up     0.9400    0.9718    0.9556     11826
-two_up_inverted     0.9709    0.9590    0.9649     11825
-       accuracy                         0.9423    212861
-      macro avg     0.9426    0.9423    0.9423    212861
-   weighted avg     0.9426    0.9423    0.9423    212861
 ```

 See https://www.kaggle.com/code/dima806/hand-gestures-image-detection-vit for more details.
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/6449300e3adf50d864095b90/3qhMpgH6NA_GI27z6reIl.png)
 ```
 Classification report:
                  precision    recall  f1-score   support
+           call     0.9063    0.9521    0.9286     11825
+        dislike     0.9695    0.9784    0.9739     11826
+           fist     0.9779    0.9862    0.9820     11826
+           four     0.8909    0.8941    0.8925     11826
+           like     0.9563    0.8864    0.9200     11825
+           mute     0.9757    0.9954    0.9855     11825
+             ok     0.9367    0.9534    0.9450     11826
+            one     0.9610    0.9371    0.9489     11825
+           palm     0.9686    0.9509    0.9596     11826
+          peace     0.8972    0.9120    0.9045     11825
+ peace_inverted     0.9781    0.9610    0.9695     11826
+           rock     0.9238    0.9156    0.9197     11826
+           stop     0.9382    0.9701    0.9539     11825
+  stop_inverted     0.9820    0.9414    0.9613     11825
+          three     0.8766    0.8823    0.8794     11826
+         three2     0.9665    0.9573    0.9619     11826
+         two_up     0.9426    0.9714    0.9568     11826
+two_up_inverted     0.9687    0.9646    0.9667     11826
+       accuracy                         0.9450    212861
+      macro avg     0.9454    0.9450    0.9450    212861
+   weighted avg     0.9454    0.9450    0.9450    212861
 ```