no rounded_graph.pb and label.txt

[zhangkom]

no pb and label file.

[dashika]

All model already uploaded to Camera-Classifiers/app/src/main/res/raw/
Why you need pd and label ?

[zhangkom]

the gradle file need two files, as followiing:

// Top-level build file where you can add configuration options common to all sub-projects/modules.

project.ext.ASSET_DIR = projectDir.toString() + '/assets'

assert file(project.ext.ASSET_DIR + "/rounded_graph.pb").exists()
assert file(project.ext.ASSET_DIR + "/retrained_labels.txt").exists()

[zhangkom]

When it work, there is a error in read label file too. I look at the source code and find them , as following:
public TensorFlowImageClassifier(AssetManager assetManager, String modelFilename, String labelFilename, int inputSize, int imageMean, float imageStd, String inputName, String outputName) {

    this.inputName = inputName;
    this.outputName = outputName;

    // Read the label names into memory.
    // TODO(andrewharp): make this handle non-assets.

    BufferedReader br = null;
    try {
        br = new BufferedReader(new InputStreamReader(assetManager.open(labelFilename)));
        String line;
        while ((line = br.readLine()) != null) {
            this.labels.add(line);
        }
        br.close();
    } catch (IOException e) {
        throw new RuntimeException("Problem reading label file!", e);
    }

Do this error not happened in your testing?

[zhangkom]

it worked, thanks