Validation Accuracy Not Showing for trainFastR​CNNObjectD​etector

1 回表示 (過去 30 日間)
L
L 2021 年 8 月 7 日
コメント済み: Shivam Singh 2021 年 10 月 11 日
Hi,
I am currently using the following code and I have no idea why the validaton accuracy is not showing.
How to fix it?
valid_label=load('...\valid_label.mat');
dataValid=valid_label.gTruth
[validIMDS, validBXDS]=objectDetectorTrainingData(dataTrain)
validData=combine(validIMDS, validBXDS)
options = trainingOptions('sgdm', ...
'MiniBatchSize', 1, ...
'InitialLearnRate', 1e-3, ...
'MaxEpochs', 10, ...
'CheckpointPath', tempdir,...
'ValidationData',validData, ...
'ValidationFrequency',30, ...j
'Plots','training-progress');
%% Train the model
frcnn S= trainFastRCNNObjectDetector(trainData, lgraph, options, ...
'NegativeOverlapRange',[0 0.3], ...
'PositiveOverlapRange',[0.6 1]);
  1 件のコメント
Shivam Singh
Shivam Singh 2021 年 10 月 11 日
Hello L,
Can you share your dataset (training and validation) and the full code, so that I can inspect clearly?

サインインしてコメントする。

回答 (0 件)

カテゴリ

Help Center および File ExchangeDeep Learning Toolbox についてさらに検索

製品


リリース

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by