You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This package is in early development. Expect [breaking changes on minor version updates](https://semver.org/#spec-item-4) until version `1.0.0`.
17
17
18
18
## Example
19
+
Let's use LRP to explain why an image of a cat gets classified as a cat:
19
20
```julia
21
+
using ExplainabilityMethods
20
22
using Flux
21
23
using Metalhead
22
-
using ExplainabilityMethods
23
24
24
25
# Load model
25
26
vgg =VGG19()
26
27
model =strip_softmax(vgg.layers)
27
28
28
29
# Run XAI method
29
30
analyzer =LRPEpsilon(model)
30
-
expl, out =analyze(img, analyzer)
31
+
expl, out =analyze(img, analyzer)
31
32
32
33
# Show heatmap
33
34
heatmap(expl)
34
35
```
36
+
![][heatmap]
35
37
38
+
39
+
## Methods
36
40
Currently, the following analyzers are implemented:
37
41
38
42
```
@@ -45,9 +49,10 @@ Currently, the following analyzers are implemented:
45
49
```
46
50
47
51
One of the design goals of ExplainabilityMethods.jl is extensibility.
48
-
Individual LRP rules like `ZeroRule`, `EpsilonRule`, `GammaRule` and `ZBoxRule` can be composed and are easily extended by custom rules.
52
+
Individual LRP rules like `ZeroRule`, `EpsilonRule`, `GammaRule` and `ZBoxRule`[can be composed][docs-composites] and are easily extended by [custom rules][docs-custom-rules].
0 commit comments