tx · 4kvCRNuM7vDGJzNvHyybgikCvkVLpNaQg79JDe7vdHvY

3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY:  -0.01000000 Waves

2024.03.20 11:02 [3026084] smart account 3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY > SELF 0.00000000 Waves

{
    "type": 13,
    "id": "4kvCRNuM7vDGJzNvHyybgikCvkVLpNaQg79JDe7vdHvY",
    "fee": 1000000,
    "feeAssetId": null,
    "timestamp": 1710921742139,
    "version": 2,
    "chainId": 84,
    "sender": "3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY",
    "senderPublicKey": "2AWdnJuBMzufXSjTvzVcawBQQhnhF1iXR6QNVgwn33oc",
    "proofs": [
        "5EKGViadaf8ULJMDEcH5mNa8j4iVhAZmTebZ5fG3JhpJQzMjMJwz6FKaBA9Pry4XvHz98PLXqbBy1KjKcEaee3kG"
    ],
    "script": "base64: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",
    "height": 3026084,
    "applicationStatus": "succeeded",
    "spentComplexity": 0
}

View: original | compacted
Prev: 93J7ZQZ4A2gErEDM76v3ufnNkgYrkUokkebJzjty2jH1
Next: BSfEC4cMC8bJB6QuEh7zoz7DZQEW1cv5iLZgKP7bH7Av

Diff: 
Old
New
Differences
1
1
 
{-# STDLIB_VERSION 5 #-}
2
2
 
{-# SCRIPT_TYPE ACCOUNT #-}
3
3
 
{-# CONTENT_TYPE DAPP #-}
4
-
let layer1Weights = [[4051769, 4062273], [-5948515, -6010085]]
4
+
let layer1Weights = [[4721113, -5002107], [6226846, -6353789]]
5
5
 
6
-
let layer1Biases = [-6307843, 2229872]
6
+
let layer1Biases = [-2521378, 3389498]
7
7
 
8
-
let layer2Weights = [[-8372358, -8139317]]
8
+
let layer2Weights = [[8109936, -7559760]]
9
9
 
10
-
let layer2Biases = [4083679]
10
+
let layer2Biases = [3490942]
11
11
 
12
12
 
func sigmoid (z) = {
13
13
 
    let e = 2718281
14
14
 
    let base = 1000000
15
15
 
    let negativeZ = (-1 * z)
16
16
 
    let expPart = fraction(e, negativeZ, base)
17
-
    fraction(base, 1000000, (base + expPart))
17
+
    fraction(base, base, (base + expPart))
18
18
 
    }
19
19
 
20
20
 
44
44
 
45
45
 
46
46
 
@Callable(i)
47
-
func predict (inputData) = {
48
-
    let input1 = if (if ((inputData == 0))
49
-
        then true
50
-
        else (inputData == 1))
51
-
        then 0
52
-
        else 1000000
53
-
    let input2 = if (if ((inputData == 0))
54
-
        then true
55
-
        else (inputData == 2))
56
-
        then 0
57
-
        else 1000000
58
-
    let result = xorNeuralNetwork(input1, input2)
47
+
func predict (input1,input2) = {
48
+
    let scaledInput1 = if ((input1 == 1))
49
+
        then 1000000
50
+
        else 0
51
+
    let scaledInput2 = if ((input2 == 2))
52
+
        then 1000000
53
+
        else 0
54
+
    let result = xorNeuralNetwork(scaledInput1, scaledInput2)
59
55
 
[IntegerEntry("result", result)]
60
56
 
    }
61
57
 

Full: 
Old
New
Differences
1
1
 
{-# STDLIB_VERSION 5 #-}
2
2
 
{-# SCRIPT_TYPE ACCOUNT #-}
3
3
 
{-# CONTENT_TYPE DAPP #-}
4
-
let layer1Weights = [[4051769, 4062273], [-5948515, -6010085]]
4
+
let layer1Weights = [[4721113, -5002107], [6226846, -6353789]]
5
5
 
6
-
let layer1Biases = [-6307843, 2229872]
6
+
let layer1Biases = [-2521378, 3389498]
7
7
 
8
-
let layer2Weights = [[-8372358, -8139317]]
8
+
let layer2Weights = [[8109936, -7559760]]
9
9
 
10
-
let layer2Biases = [4083679]
10
+
let layer2Biases = [3490942]
11
11
 
12
12
 
func sigmoid (z) = {
13
13
 
    let e = 2718281
14
14
 
    let base = 1000000
15
15
 
    let negativeZ = (-1 * z)
16
16
 
    let expPart = fraction(e, negativeZ, base)
17
-
    fraction(base, 1000000, (base + expPart))
17
+
    fraction(base, base, (base + expPart))
18
18
 
    }
19
19
 
20
20
 
21
21
 
func dotProduct (a,b) = {
22
22
 
    let product0 = fraction(a[0], b[0], 1000000)
23
23
 
    let product1 = fraction(a[1], b[1], 1000000)
24
24
 
    (product0 + product1)
25
25
 
    }
26
26
 
27
27
 
28
28
 
func forwardPass (input,weights,biases) = {
29
29
 
    let sum0 = (dotProduct(input, weights[0]) + biases[0])
30
30
 
    let sum1 = (dotProduct(input, weights[1]) + biases[1])
31
31
 
    let sig0 = sigmoid(sum0)
32
32
 
    let sig1 = sigmoid(sum1)
33
33
 
[sig0, sig1]
34
34
 
    }
35
35
 
36
36
 
37
37
 
func xorNeuralNetwork (input1,input2) = {
38
38
 
    let input = [input1, input2]
39
39
 
    let hiddenLayerOutput = forwardPass(input, layer1Weights, layer1Biases)
40
40
 
    let outputLayerSum = (dotProduct(hiddenLayerOutput, layer2Weights[0]) + layer2Biases[0])
41
41
 
    let output = sigmoid(outputLayerSum)
42
42
 
    output
43
43
 
    }
44
44
 
45
45
 
46
46
 
@Callable(i)
47
-
func predict (inputData) = {
48
-
    let input1 = if (if ((inputData == 0))
49
-
        then true
50
-
        else (inputData == 1))
51
-
        then 0
52
-
        else 1000000
53
-
    let input2 = if (if ((inputData == 0))
54
-
        then true
55
-
        else (inputData == 2))
56
-
        then 0
57
-
        else 1000000
58
-
    let result = xorNeuralNetwork(input1, input2)
47
+
func predict (input1,input2) = {
48
+
    let scaledInput1 = if ((input1 == 1))
49
+
        then 1000000
50
+
        else 0
51
+
    let scaledInput2 = if ((input2 == 2))
52
+
        then 1000000
53
+
        else 0
54
+
    let result = xorNeuralNetwork(scaledInput1, scaledInput2)
59
55
 
[IntegerEntry("result", result)]
60
56
 
    }
61
57
 
62
58
 

github/deemru/w8io/6500d08 
26.93 ms ◑