Update README.md
Browse files
README.md
CHANGED
@@ -9,7 +9,7 @@ tags:
|
|
9 |
|
10 |
# Baseline model for audio classification of orthopera and hemiptera
|
11 |
|
12 |
-
|
13 |
|
14 |
## Installation
|
15 |
|
@@ -31,8 +31,8 @@ The model has been originally trained on AudioSet by Kong et. al.
|
|
31 |
|
32 |
### Dataset
|
33 |
|
34 |
-
The model has been further trained (finetuned) on audio segments of the ECOSoundSet dataset, which is soon to be submitted,
|
35 |
-
and will referenced here as soon as it is publicly available.
|
36 |
|
37 |
|
38 |
### Features
|
@@ -46,7 +46,7 @@ The model has been trained for 30 epochs. At the end of each epoch, the model wa
|
|
46 |
|
47 |
## Evaluation
|
48 |
|
49 |
-
The performance on the test set reached a
|
50 |
|
51 |
|
52 |
## Acknowledgments
|
|
|
9 |
|
10 |
# Baseline model for audio classification of orthopera and hemiptera
|
11 |
|
12 |
+
This baseline model, utilized in the ECOSoundSet paper, was trained to tag audio files with one or more of 86 species from the Orthoptera and Hemiptera insect orders.
|
13 |
|
14 |
## Installation
|
15 |
|
|
|
31 |
|
32 |
### Dataset
|
33 |
|
34 |
+
The model has been further trained (finetuned) on 4s long audio segments of the ECOSoundSet dataset, which is soon to be submitted,
|
35 |
+
and will be referenced here as soon as it is publicly available.
|
36 |
|
37 |
|
38 |
### Features
|
|
|
46 |
|
47 |
## Evaluation
|
48 |
|
49 |
+
The performance on the test set reached a (macro) f1-score of 0.56.
|
50 |
|
51 |
|
52 |
## Acknowledgments
|