tx · FW5PA4wXaJvFdUbFe4iU7VZQgY6XboWoXMhe3W5iUXmx

3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY:  -0.01000000 Waves

2024.03.24 18:25 [3032312] smart account 3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY > SELF 0.00000000 Waves

{
    "type": 13,
    "id": "FW5PA4wXaJvFdUbFe4iU7VZQgY6XboWoXMhe3W5iUXmx",
    "fee": 1000000,
    "feeAssetId": null,
    "timestamp": 1711293939384,
    "version": 2,
    "chainId": 84,
    "sender": "3N3n75UqB8G1GKmXFr4zPhKCjGcqJPRSuJY",
    "senderPublicKey": "2AWdnJuBMzufXSjTvzVcawBQQhnhF1iXR6QNVgwn33oc",
    "proofs": [
        "3AoZn1s57vdLnLQZZuYpottVB4r43i5PV1sGTSEXTEhDWbWVVRaUMPbKG3BaRBGvBs9tPcosP5rGqNTE2WyYWWvu"
    ],
    "script": "base64:AAIFAAAAAAAAAAgIAhIECgIBAQAAAAgAAAAADWxheWVyMVdlaWdodHMJAARMAAAAAgkABEwAAAACAAAAAAAASAnZCQAETAAAAAIA//////+zrIUFAAAAA25pbAkABEwAAAACCQAETAAAAAIAAAAAAABfA54JAARMAAAAAgD//////58MgwUAAAADbmlsBQAAAANuaWwAAAAADGxheWVyMUJpYXNlcwkABEwAAAACAP//////2YbeCQAETAAAAAIAAAAAAAAzuDoFAAAAA25pbAAAAAANbGF5ZXIyV2VpZ2h0cwkABEwAAAACCQAETAAAAAIAAAAAAAB7v3AJAARMAAAAAgD//////4ylsAUAAAADbmlsBQAAAANuaWwAAAAADGxheWVyMkJpYXNlcwkABEwAAAACAAAAAAAANUR+BQAAAANuaWwBAAAAB3NpZ21vaWQAAAACAAAAAXoAAAALZGVidWdQcmVmaXgEAAAAAWUAAAAAAAApekkEAAAABGJhc2UAAAAAAAAPQkAEAAAACXBvc2l0aXZlWgMJAABmAAAAAgAAAAAAAAAAAAUAAAABegkBAAAAAS0AAAABBQAAAAF6BQAAAAF6BAAAAAdleHBQYXJ0CQAAawAAAAMFAAAAAWUFAAAABGJhc2UFAAAACXBvc2l0aXZlWgQAAAAIc2lnVmFsdWUJAABrAAAAAwUAAAAEYmFzZQUAAAAEYmFzZQkAAGQAAAACBQAAAARiYXNlBQAAAAdleHBQYXJ0CQAFFAAAAAIJAARMAAAAAgkBAAAADEludGVnZXJFbnRyeQAAAAIJAAEsAAAAAgUAAAALZGVidWdQcmVmaXgCAAAACXBvc2l0aXZlWgUAAAAJcG9zaXRpdmVaCQAETAAAAAIJAQAAAAxJbnRlZ2VyRW50cnkAAAACCQABLAAAAAIFAAAAC2RlYnVnUHJlZml4AgAAAAdleHBQYXJ0BQAAAAdleHBQYXJ0CQAETAAAAAIJAQAAAAxJbnRlZ2VyRW50cnkAAAACCQABLAAAAAIFAAAAC2RlYnVnUHJlZml4AgAAAAhzaWdWYWx1ZQUAAAAIc2lnVmFsdWUFAAAAA25pbAUAAAAIc2lnVmFsdWUBAAAACmRvdFByb2R1Y3QAAAACAAAAAWEAAAABYgQAAAAIcHJvZHVjdDAJAABrAAAAAwkAAZEAAAACBQAAAAFhAAAAAAAAAAAACQABkQAAAAIFAAAAAWIAAAAAAAAAAAAAAAAAAAAPQkAEAAAACHByb2R1Y3QxCQAAawAAAAMJAAGRAAAAAgUAAAABYQAAAAAAAAAAAQkAAZEAAAACBQAAAAFiAAAAAAAAAAABAAAAAAAAD0JACQAAZAAAAAIFAAAACHByb2R1Y3QwBQAAAAhwcm9kdWN0MQEAAAALZm9yd2FyZFBhc3MAAAAEAAAABWlucHV0AAAAB3dlaWdodHMAAAAGYmlhc2VzAAAABWxheWVyBAAAAARzdW0wCQAAZAAAAAIJAQAAAApkb3RQcm9kdWN0AAAAAgUAAAAFaW5wdXQJAAGRAAAAAgUAAAAHd2VpZ2h0cwAAAAAAAAAAAAkAAZEAAAACBQAAAAZiaWFzZXMAAAAAAAAAAAAEAAAABHN1bTEJAABkAAAAAgkBAAAACmRvdFByb2R1Y3QAAAACBQAAAAVpbnB1dAkAAZEAAAACBQAAAAd3ZWlnaHRzAAAAAAAAAAABCQABkQAAAAIFAAAABmJpYXNlcwAAAAAAAAAAAQQAAAALJHQwMTMzMTEzODgJAQAAAAdzaWdtb2lkAAAAAgUAAAAEc3VtMAkAASwAAAACBQAAAAVsYXllcgIAAAAETDFOMQQAAAANc2lnbW9pZERlYnVnMAgFAAAACyR0MDEzMzExMzg4AAAAAl8xBAAAAARzaWcwCAUAAAALJHQwMTMzMTEzODgAAAACXzIEAAAACyR0MDEzOTMxNDUwCQEAAAAHc2lnbW9pZAAAAAIFAAAABHN1bTEJAAEsAAAAAgUAAAAFbGF5ZXICAAAABEwxTjIEAAAADXNpZ21vaWREZWJ1ZzEIBQAAAAskdDAxMzkzMTQ1MAAAAAJfMQQAAAAEc2lnMQgFAAAACyR0MDEzOTMxNDUwAAAAAl8yCQAFFAAAAAIJAARMAAAAAgUAAAAEc2lnMAkABEwAAAACBQAAAARzaWcxCQAETAAAAAIFAAAABHN1bTAJAARMAAAAAgUAAAAEc3VtMQUAAAADbmlsCQAETgAAAAIFAAAADXNpZ21vaWREZWJ1ZzAFAAAADXNpZ21vaWREZWJ1ZzEBAAAAEHhvck5ldXJhbE5ldHdvcmsAAAACAAAABmlucHV0MQAAAAZpbnB1dDIEAAAABWlucHV0CQAETAAAAAIFAAAABmlucHV0MQkABEwAAAACBQAAAAZpbnB1dDIFAAAAA25pbAQAAAALJHQwMTYyODE3MjAJAQAAAAtmb3J3YXJkUGFzcwAAAAQFAAAABWlucHV0BQAAAA1sYXllcjFXZWlnaHRzBQAAAAxsYXllcjFCaWFzZXMCAAAAAkhMBAAAABFoaWRkZW5MYXllck91dHB1dAgFAAAACyR0MDE2MjgxNzIwAAAAAl8xBAAAAAtoaWRkZW5EZWJ1ZwgFAAAACyR0MDE2MjgxNzIwAAAAAl8yBAAAAAskdDAxNzI1MTg2MAkBAAAAB3NpZ21vaWQAAAACCQAAZAAAAAIJAQAAAApkb3RQcm9kdWN0AAAAAgkABEwAAAACCQABkQAAAAIFAAAAEWhpZGRlbkxheWVyT3V0cHV0AAAAAAAAAAAACQAETAAAAAIJAAGRAAAAAgUAAAARaGlkZGVuTGF5ZXJPdXRwdXQAAAAAAAAAAAEFAAAAA25pbAkAAZEAAAACBQAAAA1sYXllcjJXZWlnaHRzAAAAAAAAAAAACQABkQAAAAIFAAAADGxheWVyMkJpYXNlcwAAAAAAAAAAAAIAAAACT0wEAAAAC291dHB1dERlYnVnCAUAAAALJHQwMTcyNTE4NjAAAAACXzEEAAAABm91dHB1dAgFAAAACyR0MDE3MjUxODYwAAAAAl8yCQAFFAAAAAIJAARMAAAAAgUAAAAGb3V0cHV0CQAETAAAAAIJAABkAAAAAgkBAAAACmRvdFByb2R1Y3QAAAACCQAETAAAAAIJAAGRAAAAAgUAAAARaGlkZGVuTGF5ZXJPdXRwdXQAAAAAAAAAAAAJAARMAAAAAgkAAZEAAAACBQAAABFoaWRkZW5MYXllck91dHB1dAAAAAAAAAAAAQUAAAADbmlsCQABkQAAAAIFAAAADWxheWVyMldlaWdodHMAAAAAAAAAAAAJAAGRAAAAAgUAAAAMbGF5ZXIyQmlhc2VzAAAAAAAAAAAACQAETAAAAAIJAAGRAAAAAgUAAAARaGlkZGVuTGF5ZXJPdXRwdXQAAAAAAAAAAAIJAARMAAAAAgkAAZEAAAACBQAAABFoaWRkZW5MYXllck91dHB1dAAAAAAAAAAAAwUAAAADbmlsCQAETgAAAAIFAAAAC2hpZGRlbkRlYnVnBQAAAAtvdXRwdXREZWJ1ZwAAAAEAAAABaQEAAAAQcHJlZGljdF9vcmlnaW5hbAAAAAIAAAAGaW5wdXQxAAAABmlucHV0MgQAAAAMc2NhbGVkSW5wdXQxAwkAAAAAAAACBQAAAAZpbnB1dDEAAAAAAAAAAAEAAAAAAAAPQkAAAAAAAAAAAAAEAAAADHNjYWxlZElucHV0MgMJAAAAAAAAAgUAAAAGaW5wdXQyAAAAAAAAAAABAAAAAAAAD0JAAAAAAAAAAAAABAAAAAskdDAyMjU0MjMzNQkBAAAAEHhvck5ldXJhbE5ldHdvcmsAAAACBQAAAAxzY2FsZWRJbnB1dDEFAAAADHNjYWxlZElucHV0MgQAAAAObmV0d29ya091dHB1dHMIBQAAAAskdDAyMjU0MjMzNQAAAAJfMQQAAAAMZGVidWdFbnRyaWVzCAUAAAALJHQwMjI1NDIzMzUAAAACXzIJAAROAAAAAgkABEwAAAACCQEAAAAMSW50ZWdlckVudHJ5AAAAAgIAAAAGcmVzdWx0CQABkQAAAAIFAAAADm5ldHdvcmtPdXRwdXRzAAAAAAAAAAAACQAETAAAAAIJAQAAAAxJbnRlZ2VyRW50cnkAAAACAgAAAA5vdXRwdXRMYXllclN1bQkAAZEAAAACBQAAAA5uZXR3b3JrT3V0cHV0cwAAAAAAAAAAAQkABEwAAAACCQEAAAAMSW50ZWdlckVudHJ5AAAAAgIAAAAVaGlkZGVuTGF5ZXJPdXRwdXQxU3VtCQABkQAAAAIFAAAADm5ldHdvcmtPdXRwdXRzAAAAAAAAAAACCQAETAAAAAIJAQAAAAxJbnRlZ2VyRW50cnkAAAACAgAAABVoaWRkZW5MYXllck91dHB1dDJTdW0JAAGRAAAAAgUAAAAObmV0d29ya091dHB1dHMAAAAAAAAAAAMFAAAAA25pbAUAAAAMZGVidWdFbnRyaWVzAAAAANKLcGQ=",
    "height": 3032312,
    "applicationStatus": "succeeded",
    "spentComplexity": 0
}

View: original | compacted
Prev: J63xWbm4vHe8K69nUjTn84N9HuR4NGt6i3fVHent4GN
Next: 6ikdwN6r2UXQdkEA6cxgC4p7TaERghmy8jqXLicqgtRG

Diff: 
Old
New
Differences
1
1
 
{-# STDLIB_VERSION 5 #-}
2
2
 
{-# SCRIPT_TYPE ACCOUNT #-}
3
3
 
{-# CONTENT_TYPE DAPP #-}
4
-
let layer1Weights = [[6004965, 6007324], [4141966, 4142525]]
4
+
let layer1Weights = [[4721113, -5002107], [6226846, -6353789]]
5
5
 
6
-
let layer1Biases = [-2590503, -6356371]
6
+
let layer1Biases = [-2521378, 3389498]
7
7
 
8
-
let layer2Weights = [[8329656, -8971418]]
8
+
let layer2Weights = [[8109936, -7559760]]
9
9
 
10
-
let layer2Biases = [-3811788]
10
+
let layer2Biases = [3490942]
11
11
 
12
12
 
func sigmoid (z,debugPrefix) = {
13
13
 
    let e = 2718281
21
21
 
    }
22
22
 
23
23
 
24
-
func forwardPassLayer1 (input,weights,biases,debugPrefix) = {
25
-
    let sum0 = ((fraction(input[0], weights[0][0], 1000000) + fraction(input[1], weights[0][1], 1000000)) + biases[0])
26
-
    let sum1 = ((fraction(input[0], weights[1][0], 1000000) + fraction(input[1], weights[1][1], 1000000)) + biases[1])
27
-
    let $t011811237 = sigmoid(sum0, (debugPrefix + "L1N0"))
28
-
    let debug0 = $t011811237._1
29
-
    let sig0 = $t011811237._2
30
-
    let $t012421298 = sigmoid(sum1, (debugPrefix + "L1N1"))
31
-
    let debug1 = $t012421298._1
32
-
    let sig1 = $t012421298._2
33
-
    $Tuple2([sig0, sig1], (debug0 ++ debug1))
24
+
func dotProduct (a,b) = {
25
+
    let product0 = fraction(a[0], b[0], 1000000)
26
+
    let product1 = fraction(a[1], b[1], 1000000)
27
+
    (product0 + product1)
34
28
 
    }
35
29
 
36
30
 
37
-
func forwardPassLayer2 (input,weights,biases,debugPrefix) = {
38
-
    let sum0 = ((fraction(input[0], weights[0], 1000000) + fraction(input[1], weights[1], 1000000)) + biases)
39
-
    let sum1 = ((fraction(input[0], weights[0], 1000000) + fraction(input[1], weights[1], 1000000)) + biases)
40
-
    let $t016521708 = sigmoid(sum0, (debugPrefix + "L2N0"))
41
-
    let debug0 = $t016521708._1
42
-
    let sig0 = $t016521708._2
43
-
    let $t017131769 = sigmoid(sum1, (debugPrefix + "L2N1"))
44
-
    let debug1 = $t017131769._1
45
-
    let sig1 = $t017131769._2
46
-
    $Tuple2(sig0, (debug0 ++ debug1))
31
+
func forwardPass (input,weights,biases,layer) = {
32
+
    let sum0 = (dotProduct(input, weights[0]) + biases[0])
33
+
    let sum1 = (dotProduct(input, weights[1]) + biases[1])
34
+
    let $t013311388 = sigmoid(sum0, (layer + "L1N1"))
35
+
    let sigmoidDebug0 = $t013311388._1
36
+
    let sig0 = $t013311388._2
37
+
    let $t013931450 = sigmoid(sum1, (layer + "L1N2"))
38
+
    let sigmoidDebug1 = $t013931450._1
39
+
    let sig1 = $t013931450._2
40
+
    $Tuple2([sig0, sig1, sum0, sum1], (sigmoidDebug0 ++ sigmoidDebug1))
41
+
    }
42
+
43
+
44
+
func xorNeuralNetwork (input1,input2) = {
45
+
    let input = [input1, input2]
46
+
    let $t016281720 = forwardPass(input, layer1Weights, layer1Biases, "HL")
47
+
    let hiddenLayerOutput = $t016281720._1
48
+
    let hiddenDebug = $t016281720._2
49
+
    let $t017251860 = sigmoid((dotProduct([hiddenLayerOutput[0], hiddenLayerOutput[1]], layer2Weights[0]) + layer2Biases[0]), "OL")
50
+
    let outputDebug = $t017251860._1
51
+
    let output = $t017251860._2
52
+
    $Tuple2([output, (dotProduct([hiddenLayerOutput[0], hiddenLayerOutput[1]], layer2Weights[0]) + layer2Biases[0]), hiddenLayerOutput[2], hiddenLayerOutput[3]], (hiddenDebug ++ outputDebug))
47
53
 
    }
48
54
 
49
55
 
50
56
 
@Callable(i)
51
-
func predict (input1,input2) = {
57
+
func predict_original (input1,input2) = {
52
58
 
    let scaledInput1 = if ((input1 == 1))
53
59
 
        then 1000000
54
60
 
        else 0
55
61
 
    let scaledInput2 = if ((input2 == 1))
56
62
 
        then 1000000
57
63
 
        else 0
58
-
    let inputs = [scaledInput1, scaledInput2]
59
-
    let $t020302128 = forwardPassLayer1(inputs, layer1Weights, layer1Biases, "Layer1")
60
-
    let layer1Output = $t020302128._1
61
-
    let debugLayer1 = $t020302128._2
62
-
    let $t021332243 = forwardPassLayer2(layer1Output, layer2Weights[0], layer2Biases[0], "Layer2")
63
-
    let layer2Output = $t021332243._1
64
-
    let debugLayer2 = $t021332243._2
65
-
    (([IntegerEntry("result", layer2Output)] ++ debugLayer1) ++ debugLayer2)
64
+
    let $t022542335 = xorNeuralNetwork(scaledInput1, scaledInput2)
65
+
    let networkOutputs = $t022542335._1
66
+
    let debugEntries = $t022542335._2
67
+
    ([IntegerEntry("result", networkOutputs[0]), IntegerEntry("outputLayerSum", networkOutputs[1]), IntegerEntry("hiddenLayerOutput1Sum", networkOutputs[2]), IntegerEntry("hiddenLayerOutput2Sum", networkOutputs[3])] ++ debugEntries)
66
68
 
    }
67
69
 
68
70
 

Full: 
Old
New
Differences
1
1
 
{-# STDLIB_VERSION 5 #-}
2
2
 
{-# SCRIPT_TYPE ACCOUNT #-}
3
3
 
{-# CONTENT_TYPE DAPP #-}
4
-
let layer1Weights = [[6004965, 6007324], [4141966, 4142525]]
4
+
let layer1Weights = [[4721113, -5002107], [6226846, -6353789]]
5
5
 
6
-
let layer1Biases = [-2590503, -6356371]
6
+
let layer1Biases = [-2521378, 3389498]
7
7
 
8
-
let layer2Weights = [[8329656, -8971418]]
8
+
let layer2Weights = [[8109936, -7559760]]
9
9
 
10
-
let layer2Biases = [-3811788]
10
+
let layer2Biases = [3490942]
11
11
 
12
12
 
func sigmoid (z,debugPrefix) = {
13
13
 
    let e = 2718281
14
14
 
    let base = 1000000
15
15
 
    let positiveZ = if ((0 > z))
16
16
 
        then -(z)
17
17
 
        else z
18
18
 
    let expPart = fraction(e, base, positiveZ)
19
19
 
    let sigValue = fraction(base, base, (base + expPart))
20
20
 
    $Tuple2([IntegerEntry((debugPrefix + "positiveZ"), positiveZ), IntegerEntry((debugPrefix + "expPart"), expPart), IntegerEntry((debugPrefix + "sigValue"), sigValue)], sigValue)
21
21
 
    }
22
22
 
23
23
 
24
-
func forwardPassLayer1 (input,weights,biases,debugPrefix) = {
25
-
    let sum0 = ((fraction(input[0], weights[0][0], 1000000) + fraction(input[1], weights[0][1], 1000000)) + biases[0])
26
-
    let sum1 = ((fraction(input[0], weights[1][0], 1000000) + fraction(input[1], weights[1][1], 1000000)) + biases[1])
27
-
    let $t011811237 = sigmoid(sum0, (debugPrefix + "L1N0"))
28
-
    let debug0 = $t011811237._1
29
-
    let sig0 = $t011811237._2
30
-
    let $t012421298 = sigmoid(sum1, (debugPrefix + "L1N1"))
31
-
    let debug1 = $t012421298._1
32
-
    let sig1 = $t012421298._2
33
-
    $Tuple2([sig0, sig1], (debug0 ++ debug1))
24
+
func dotProduct (a,b) = {
25
+
    let product0 = fraction(a[0], b[0], 1000000)
26
+
    let product1 = fraction(a[1], b[1], 1000000)
27
+
    (product0 + product1)
34
28
 
    }
35
29
 
36
30
 
37
-
func forwardPassLayer2 (input,weights,biases,debugPrefix) = {
38
-
    let sum0 = ((fraction(input[0], weights[0], 1000000) + fraction(input[1], weights[1], 1000000)) + biases)
39
-
    let sum1 = ((fraction(input[0], weights[0], 1000000) + fraction(input[1], weights[1], 1000000)) + biases)
40
-
    let $t016521708 = sigmoid(sum0, (debugPrefix + "L2N0"))
41
-
    let debug0 = $t016521708._1
42
-
    let sig0 = $t016521708._2
43
-
    let $t017131769 = sigmoid(sum1, (debugPrefix + "L2N1"))
44
-
    let debug1 = $t017131769._1
45
-
    let sig1 = $t017131769._2
46
-
    $Tuple2(sig0, (debug0 ++ debug1))
31
+
func forwardPass (input,weights,biases,layer) = {
32
+
    let sum0 = (dotProduct(input, weights[0]) + biases[0])
33
+
    let sum1 = (dotProduct(input, weights[1]) + biases[1])
34
+
    let $t013311388 = sigmoid(sum0, (layer + "L1N1"))
35
+
    let sigmoidDebug0 = $t013311388._1
36
+
    let sig0 = $t013311388._2
37
+
    let $t013931450 = sigmoid(sum1, (layer + "L1N2"))
38
+
    let sigmoidDebug1 = $t013931450._1
39
+
    let sig1 = $t013931450._2
40
+
    $Tuple2([sig0, sig1, sum0, sum1], (sigmoidDebug0 ++ sigmoidDebug1))
41
+
    }
42
+
43
+
44
+
func xorNeuralNetwork (input1,input2) = {
45
+
    let input = [input1, input2]
46
+
    let $t016281720 = forwardPass(input, layer1Weights, layer1Biases, "HL")
47
+
    let hiddenLayerOutput = $t016281720._1
48
+
    let hiddenDebug = $t016281720._2
49
+
    let $t017251860 = sigmoid((dotProduct([hiddenLayerOutput[0], hiddenLayerOutput[1]], layer2Weights[0]) + layer2Biases[0]), "OL")
50
+
    let outputDebug = $t017251860._1
51
+
    let output = $t017251860._2
52
+
    $Tuple2([output, (dotProduct([hiddenLayerOutput[0], hiddenLayerOutput[1]], layer2Weights[0]) + layer2Biases[0]), hiddenLayerOutput[2], hiddenLayerOutput[3]], (hiddenDebug ++ outputDebug))
47
53
 
    }
48
54
 
49
55
 
50
56
 
@Callable(i)
51
-
func predict (input1,input2) = {
57
+
func predict_original (input1,input2) = {
52
58
 
    let scaledInput1 = if ((input1 == 1))
53
59
 
        then 1000000
54
60
 
        else 0
55
61
 
    let scaledInput2 = if ((input2 == 1))
56
62
 
        then 1000000
57
63
 
        else 0
58
-
    let inputs = [scaledInput1, scaledInput2]
59
-
    let $t020302128 = forwardPassLayer1(inputs, layer1Weights, layer1Biases, "Layer1")
60
-
    let layer1Output = $t020302128._1
61
-
    let debugLayer1 = $t020302128._2
62
-
    let $t021332243 = forwardPassLayer2(layer1Output, layer2Weights[0], layer2Biases[0], "Layer2")
63
-
    let layer2Output = $t021332243._1
64
-
    let debugLayer2 = $t021332243._2
65
-
    (([IntegerEntry("result", layer2Output)] ++ debugLayer1) ++ debugLayer2)
64
+
    let $t022542335 = xorNeuralNetwork(scaledInput1, scaledInput2)
65
+
    let networkOutputs = $t022542335._1
66
+
    let debugEntries = $t022542335._2
67
+
    ([IntegerEntry("result", networkOutputs[0]), IntegerEntry("outputLayerSum", networkOutputs[1]), IntegerEntry("hiddenLayerOutput1Sum", networkOutputs[2]), IntegerEntry("hiddenLayerOutput2Sum", networkOutputs[3])] ++ debugEntries)
66
68
 
    }
67
69
 
68
70
 

github/deemru/w8io/6500d08 
23.24 ms ◑