tx · HQYVbfLQm6q7Et1MgYX3SjNGapVKGrrmzK5duBWV2fLi

3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY:  -0.01000000 Waves

2024.05.04 11:47 [3091161] smart account 3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY > SELF 0.00000000 Waves

{
    "type": 13,
    "id": "HQYVbfLQm6q7Et1MgYX3SjNGapVKGrrmzK5duBWV2fLi",
    "fee": 1000000,
    "feeAssetId": null,
    "timestamp": 1714812492950,
    "version": 2,
    "chainId": 84,
    "sender": "3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY",
    "senderPublicKey": "2AWdnJuBMzufXSjTvzVcawBQQhnhF1iXR6QNVgwn33oc",
    "proofs": [
        "2ZVGoA64vK9mEiXosJoWafczHnpCQRhkpGKQbYExvXnJTjB2JqrLXCuh9DVn7rXXmYJS7YSe5kbUr2WdfxkrcRm5"
    ],
    "script": "base64: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",
    "height": 3091161,
    "applicationStatus": "succeeded",
    "spentComplexity": 0
}

View: original | compacted
Prev: CrFqEGY8DWvwsYq3yExJteW66i3hdNFg277CA3AXDGFm
Next: BgJxwNe2CWQhu6e8YQTPKDw8KySzHCR28WNTUgBG6V1v

Diff: 
Old
New
Differences
31
31
 
32
32
 
33
33
 
func feedforward (inputs) = {
34
-
    let layer1Result1 = sigmoid_approx((dotProduct(inputs, layer1Weights[0]) + layer1Biases[0]))
35
-
    let layer1Result2 = sigmoid_approx((dotProduct(inputs, layer1Weights[1]) + layer1Biases[1]))
34
+
    let dp1 = dotProduct(inputs, layer1Weights[0])
35
+
    let dp2 = dotProduct(inputs, layer1Weights[1])
36
+
    let layer1Result1 = sigmoid_approx((dp1 + layer1Biases[0]))
37
+
    let layer1Result2 = sigmoid_approx((dp2 + layer1Biases[1]))
36
38
 
    let layer2Inputs = [layer1Result1, layer1Result2]
37
-
    sigmoid_approx((dotProduct(layer2Inputs, layer2Weights[0]) + layer2Biases[0]))
39
+
    let dp3 = dotProduct(layer2Inputs, layer2Weights[0])
40
+
    let output = sigmoid_approx((dp3 + layer2Biases[0]))
41
+
    $Tuple6(output, dp1, dp2, layer1Result1, layer1Result2, dp3)
38
42
 
    }
39
43
 
40
44
 
41
45
 
@Callable(i)
42
46
 
func predict (input1,input2) = {
43
47
 
    let inputs = [input1, input2]
44
-
    let prediction = feedforward(inputs)
45
-
[IntegerEntry("prediction", prediction)]
48
+
    let $t014341517 = feedforward(inputs)
49
+
    let prediction = $t014341517._1
50
+
    let dp1 = $t014341517._2
51
+
    let dp2 = $t014341517._3
52
+
    let layer1Result1 = $t014341517._4
53
+
    let layer1Result2 = $t014341517._5
54
+
    let dp3 = $t014341517._6
55
+
[IntegerEntry("prediction", prediction), IntegerEntry("dotProduct1", dp1), IntegerEntry("dotProduct2", dp2), IntegerEntry("layer1Result1", layer1Result1), IntegerEntry("layer1Result2", layer1Result2), IntegerEntry("dotProduct3", dp3)]
46
56
 
    }
47
57
 
48
58
 

Full: 
Old
New
Differences
1
1
 
{-# STDLIB_VERSION 5 #-}
2
2
 
{-# SCRIPT_TYPE ACCOUNT #-}
3
3
 
{-# CONTENT_TYPE DAPP #-}
4
4
 
let layer1Weights = [[600496, 600733], [414197, 414253]]
5
5
 
6
6
 
let layer1Biases = [-259050, -635637]
7
7
 
8
8
 
let layer2Weights = [[832966, -897141]]
9
9
 
10
10
 
let layer2Biases = [-381179]
11
11
 
12
12
 
func relu (x) = if ((x > 0))
13
13
 
    then x
14
14
 
    else 0
15
15
 
16
16
 
17
17
 
func sigmoid_approx (x) = if ((-5000 > x))
18
18
 
    then 0
19
19
 
    else if ((0 > x))
20
20
 
        then 5000
21
21
 
        else if ((5000 > x))
22
22
 
            then 9500
23
23
 
            else 10000
24
24
 
25
25
 
26
26
 
func dotProduct (v1,v2) = {
27
27
 
    let sum1 = ((v1[0] * v2[0]) / 10000)
28
28
 
    let sum2 = ((v1[1] * v2[1]) / 10000)
29
29
 
    (sum1 + sum2)
30
30
 
    }
31
31
 
32
32
 
33
33
 
func feedforward (inputs) = {
34
-
    let layer1Result1 = sigmoid_approx((dotProduct(inputs, layer1Weights[0]) + layer1Biases[0]))
35
-
    let layer1Result2 = sigmoid_approx((dotProduct(inputs, layer1Weights[1]) + layer1Biases[1]))
34
+
    let dp1 = dotProduct(inputs, layer1Weights[0])
35
+
    let dp2 = dotProduct(inputs, layer1Weights[1])
36
+
    let layer1Result1 = sigmoid_approx((dp1 + layer1Biases[0]))
37
+
    let layer1Result2 = sigmoid_approx((dp2 + layer1Biases[1]))
36
38
 
    let layer2Inputs = [layer1Result1, layer1Result2]
37
-
    sigmoid_approx((dotProduct(layer2Inputs, layer2Weights[0]) + layer2Biases[0]))
39
+
    let dp3 = dotProduct(layer2Inputs, layer2Weights[0])
40
+
    let output = sigmoid_approx((dp3 + layer2Biases[0]))
41
+
    $Tuple6(output, dp1, dp2, layer1Result1, layer1Result2, dp3)
38
42
 
    }
39
43
 
40
44
 
41
45
 
@Callable(i)
42
46
 
func predict (input1,input2) = {
43
47
 
    let inputs = [input1, input2]
44
-
    let prediction = feedforward(inputs)
45
-
[IntegerEntry("prediction", prediction)]
48
+
    let $t014341517 = feedforward(inputs)
49
+
    let prediction = $t014341517._1
50
+
    let dp1 = $t014341517._2
51
+
    let dp2 = $t014341517._3
52
+
    let layer1Result1 = $t014341517._4
53
+
    let layer1Result2 = $t014341517._5
54
+
    let dp3 = $t014341517._6
55
+
[IntegerEntry("prediction", prediction), IntegerEntry("dotProduct1", dp1), IntegerEntry("dotProduct2", dp2), IntegerEntry("layer1Result1", layer1Result1), IntegerEntry("layer1Result2", layer1Result2), IntegerEntry("dotProduct3", dp3)]
46
56
 
    }
47
57
 
48
58
 

github/deemru/w8io/6500d08 
31.71 ms ◑