Update README.md
Browse files
README.md
CHANGED
|
@@ -59,30 +59,29 @@ ornithology and machine learning.
|
|
| 59 |
Mean performance of AudioProtoPNet models with one, five, ten, and twenty prototypes per class for the
|
| 60 |
validation dataset POW and the seven test datasets, averaged over five different random seeds. The 'Score'
|
| 61 |
column represents the average of the respective metric across all test datasets. Best values for each metric are
|
| 62 |
-
**bolded
|
| 63 |
similarly, the model with only one prototype per class showed slightly lower performance.
|
| 64 |
|
| 65 |
-
|
|
| 66 |
-
|
| 67 |
-
| AudioProtoPNet-1
|
| 68 |
-
|
|
| 69 |
-
|
|
| 70 |
-
| AudioProtoPNet-5
|
| 71 |
-
|
|
| 72 |
-
|
|
| 73 |
-
| AudioProtoPNet-10
|
| 74 |
-
|
|
| 75 |
-
|
|
| 76 |
-
| AudioProtoPNet-20
|
| 77 |
-
|
|
| 78 |
-
|
|
| 79 |
|
| 80 |
**Table 2: Comparative Performance of AudioProtoPNet, ConvNeXt, and Perch**
|
| 81 |
|
| 82 |
Mean performance of AudioProtoPNet-5, ConvNeXt, and Perch for the validation dataset POW and the seven
|
| 83 |
test datasets, averaged over five different random seeds. The 'Score' column represents the average of the
|
| 84 |
-
respective metric across all test datasets. Best values for each metric are **bolded
|
| 85 |
-
*underlined*. AudioProtoPNet-5 notably outperformed both Perch and ConvNeXt in terms of cmAP, AUROC,
|
| 86 |
and top-1 accuracy scores.
|
| 87 |
|
| 88 |
| Model | Metric | POW | PER | NES | UHH | HSN | NBP | SSW | SNE | Score |
|
|
@@ -97,6 +96,8 @@ and top-1 accuracy scores.
|
|
| 97 |
| | AUROC | 0.84 | 0.70 | 0.90 | 0.76 | 0.86 | 0.91 | 0.91 | 0.83 | 0.84 |
|
| 98 |
| | T1-Acc | 0.85 | 0.48 | **0.66** | **0.57** | 0.58 | 0.69 | 0.62 | 0.69 | 0.61 |
|
| 99 |
|
|
|
|
|
|
|
| 100 |
## Example
|
| 101 |
|
| 102 |
This model can be easily loaded and used for inference with the `transformers` library.
|
|
|
|
| 59 |
Mean performance of AudioProtoPNet models with one, five, ten, and twenty prototypes per class for the
|
| 60 |
validation dataset POW and the seven test datasets, averaged over five different random seeds. The 'Score'
|
| 61 |
column represents the average of the respective metric across all test datasets. Best values for each metric are
|
| 62 |
+
**bolded**. While models with five, ten, and twenty prototypes performed
|
| 63 |
similarly, the model with only one prototype per class showed slightly lower performance.
|
| 64 |
|
| 65 |
+
| | Metric | POW | PER | NES | UHH | HSN | NBP | SSW | SNE | Score |
|
| 66 |
+
|----------------------|---------|-------|-------|-------|-------|-------|-------|-------|-------|-------|
|
| 67 |
+
| AudioProtoPNet-1 | cmAP | 0.49 | **0.30** | 0.36 | 0.28 | 0.50 | 0.66 | 0.40 | 0.32 | 0.40 |
|
| 68 |
+
| | AUROC | 0.88 | 0.79 | 0.92 | 0.85 | 0.91 | 0.92 | 0.96 | 0.84 | 0.88 |
|
| 69 |
+
| | T1-Acc | **0.87** | 0.59 | 0.49 | 0.42 | 0.64 | 0.71 | 0.64 | 0.70 | 0.60 |
|
| 70 |
+
| AudioProtoPNet-5 | cmAP | **0.50** | **0.30** | **0.38** | **0.31** | **0.54** | **0.68** | 0.42 | 0.33 | **0.42** |
|
| 71 |
+
| | AUROC | 0.88 | 0.79 | 0.93 | **0.87** | **0.92** | **0.93** | **0.97** | **0.88** | **0.90** |
|
| 72 |
+
| | T1-Acc | 0.84 | 0.59 | **0.52** | **0.49** | **0.65** | 0.71 | 0.66 | 0.74 | **0.62** |
|
| 73 |
+
| AudioProtoPNet-10 | cmAP | **0.50** | **0.30** | **0.38** | 0.30 | **0.54** | **0.68** | 0.42 | **0.34** | **0.42** |
|
| 74 |
+
| | AUROC | 0.88 | **0.80** | **0.94** | 0.86 | **0.92** | **0.93** | **0.97** | 0.86 | **0.90** |
|
| 75 |
+
| | T1-Acc | 0.85 | 0.59 | **0.52** | 0.47 | 0.64 | **0.72** | 0.67 | 0.74 | **0.62** |
|
| 76 |
+
| AudioProtoPNet-20 | cmAP | **0.50** | **0.30** | **0.38** | **0.31** | **0.54** | **0.68** | **0.43** | 0.33 | **0.42** |
|
| 77 |
+
| | AUROC | **0.89** | **0.80** | **0.94** | 0.86 | **0.92** | **0.93** | **0.97** | 0.87 | **0.90** |
|
| 78 |
+
| | T1-Acc | **0.87** | **0.60** | **0.52** | 0.42 | **0.65** | **0.72** | **0.68** | **0.75** | **0.62** |
|
| 79 |
|
| 80 |
**Table 2: Comparative Performance of AudioProtoPNet, ConvNeXt, and Perch**
|
| 81 |
|
| 82 |
Mean performance of AudioProtoPNet-5, ConvNeXt, and Perch for the validation dataset POW and the seven
|
| 83 |
test datasets, averaged over five different random seeds. The 'Score' column represents the average of the
|
| 84 |
+
respective metric across all test datasets. Best values for each metric are **bolded**. AudioProtoPNet-5 notably outperformed both Perch and ConvNeXt in terms of cmAP, AUROC,
|
|
|
|
| 85 |
and top-1 accuracy scores.
|
| 86 |
|
| 87 |
| Model | Metric | POW | PER | NES | UHH | HSN | NBP | SSW | SNE | Score |
|
|
|
|
| 96 |
| | AUROC | 0.84 | 0.70 | 0.90 | 0.76 | 0.86 | 0.91 | 0.91 | 0.83 | 0.84 |
|
| 97 |
| | T1-Acc | 0.85 | 0.48 | **0.66** | **0.57** | 0.58 | 0.69 | 0.62 | 0.69 | 0.61 |
|
| 98 |
|
| 99 |
+
|
| 100 |
+
|
| 101 |
## Example
|
| 102 |
|
| 103 |
This model can be easily loaded and used for inference with the `transformers` library.
|