Scribd Logo
Upload

Welcome to Scribd!

  • 10 views

    Modeldefault - CKPT Modelgradientdescent - CKPT Modelbatch200.Ckpt

    Uploaded by

    stark
    The document describes the configuration and results of several neural network models trained on a dataset. It provides information on the number of steps, batch size, convolutional and max pool layers, optimizer, learning rate, hidden layers and dropout rate for each model. It also shows the accuracy at various steps of training and on test datasets for each model. The models differ in terms of batch size, number of layers, number of steps and hidden layer configuration. The best performing models achieved over 98% accuracy on the online test set.

    Copyright:

    © All Rights Reserved
    Download as XLSX, PDF, TXT or read online from Scribd

    Modeldefault - CKPT Modelgradientdescent - CKPT Modelbatch200.Ckpt

    Uploaded by

    stark
    10 views3 pages
    The document describes the configuration and results of several neural network models trained on a dataset. It provides information on the number of steps, batch size, convolutional and max pool layers, optimizer, learning rate, hidden layers and dropout rate for each model. It also shows the accuracy at various steps of training and on test datasets for each model. The models differ in terms of batch size, number of layers, number of steps and hidden layer configuration. The best performing models achieved over 98% accuracy on the online test set.

    Original Title

    Model_tune_log

    Copyright

    © © All Rights Reserved

    Available Formats

    XLSX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document describes the configuration and results of several neural network models trained on a dataset. It provides information on the number of steps, batch size, convolutional and max pool layers, optimizer, learning rate, hidden layers and dropout rate for each model. It also shows the accuracy at various steps of training and on test datasets for each model. The models differ in terms of batch size, number of layers, number of steps and hidden layer configuration. The best performing models achieved over 98% accuracy on the online test set.

    Copyright:

    © All Rights Reserved
    Download as XLSX, PDF, TXT or read online from Scribd
    Download as xlsx, pdf, or txt
    10 views3 pages

    Modeldefault - CKPT Modelgradientdescent - CKPT Modelbatch200.Ckpt

    Uploaded by

    stark
    The document describes the configuration and results of several neural network models trained on a dataset. It provides information on the number of steps, batch size, convolutional and max pool layers, optimizer, learning rate, hidden layers and dropout rate for each model. It also shows the accuracy at various steps of training and on test datasets for each model. The models differ in terms of batch size, number of layers, number of steps and hidden layer configuration. The best performing models achieved over 98% accuracy on the online test set.

    Copyright:

    © All Rights Reserved
    Download as XLSX, PDF, TXT or read online from Scribd
    Download as xlsx, pdf, or txt
    You are on page 1of 3

    modelDefault.ckpt modelGradientDescent.ckpt modelBatch200.

    ckpt
    Steps 3000 3000 3000
    BatchSize 50 50 300
    Convlution 1,1 1,1 1,1
    KennelSize 2,2 2,2 2,2
    MaxPool 2,2 2,2 2,2
    Optimizer AdamOptimizer GradientDescentOptimizer AdamOptimizer
    Learning Rate 0.0001 0.50 0.0001
    Layer 1 featur 32 32 32
    Layer 1 patch 5 5 5
    Layer 2 featur 64 64 64
    Layer 2 patch 5 5 5
    Hidden 1024 1024 1024
    Dropoffrate 0.5 0.5 0.5
    Layer 3 features
    Layer 3 patch

    Training & result


    0 steps 0.88 0.12 0.1333
    100 0.94 0.12 0.9267
    200 0.88 0.14 0.8933
    300 0.96 0.12 0.9533
    400 0.96 0.1 0.9633
    1000 0.96 0.1 0.98
    1500 0.96 0.14 0.98
    2000 0.96 0.08 0.98
    2500 1 0.08 1
    3000 0.98 0.04 0.98
    5000
    10000
    15000
    20000
    ONLINE_TEST 0.9784 0.1135 0.9882
    HW_TEST 0.4923 0.06 0.6153

    Comment
    modelHiddenLayer.ckpt modelSteps.ckpt model3Layer.ckpt
    3000 20000 3000
    100 50 100
    1,1 1,1 1,1
    2,2 2,2 2,2
    2,2 2,2 2,2
    AdamOptimizer AdamOptimizer AdamOptimizer
    0.0010 0.0001 0.0010
    32 32 16
    5 5 5
    64 64 32
    5 5 5
    100 1024 100
    0.5 0.5 0.5
    64
    5
    layer3_maxPool = 1,1

    0.11 0.1 0.08


    0.92 0.86 0.94
    0.96 0.92 0.96
    0.98 0.8 0.96
    0.94 0.98 0.95
    0.98 0.96 0.98
    0.98 1 0.99
    1 0.92 1
    0.99 1 0.99
    0.99 1 1
    0.98 1
    1 1
    1 1
    1 1
    0.9921 0.9932 0.9928
    0.7282 0.7179 0.8974

    good 3k steps, good


    model3Layer2.ckpt modelHiddenLayerSuper.ckpt
    3000 20000
    100 100
    1,1 1,1
    2,2 2,2
    2,2 2,2
    AdamOptimizer AdamOptimizer
    0.0010 0.0010
    16 32
    5 5
    32 64
    5 5
    100 100
    0.5 0.5
    64
    5
    layer2_maxPool = 1,1

    0.07 0.12
    0.95 0.93
    0.97 0.97
    0.98 0.97
    0.96 0.96
    0.98 0.99
    0.99 0.96
    0.99 0.99
    0.99 0.99
    1 1
    1
    1
    0.99
    1
    0.9929 0.9939
    0.8205 0.8461

    ok

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy