Tensorflow Invalid Argument: Assertation Failed [Label IDs must < n_classes]

Question

I get an error implementing a DNNClassifier in Tensorflow 1.3.0 with Python 2.7. I got the sample code from the Tensorflow tf.estimator Quickstart Tutorial and I want to run it with my own dataset: 3D coordinates and 10 different classes (int labels). Here is my implementation:

#!/usr/bin/env python
# -*- coding: utf-8 -*-

def ReadLabels(file):
    #load the labels from test file here
    labelFile = open(file, "r")
    Label = labelFile.readlines();
    returnL = [[Label[i][j+1] for j in range(len(Label[0])-3)] for i in range(len(Label))]
    returnLint = list();
    for i in range(len(returnL)):
        tmp = ''
        for j in range(len(returnL[0])):
            tmp += str(returnL[i][j])
        returnLint.append(int(tmp))
    return returnL, returnLint

def NumpyReadBin(file,numcols,type):
    #load the data from binary file here
    import numpy as np
    trainData = np.fromfile(file,dtype=type)
    numrows = len(trainData)/numcols
    #print trainData[0:100]
    result = [[trainData[i+j*numcols] for i in range(numcols)] for j in range(numrows)]
    return result

def TensorflowDNN():
    #load sample dataset
    trainData = NumpyReadBin('data/TrainingData.dat',3,'float32')
    valData = NumpyReadBin('data/ValidationData.dat',3,'float32')
    testData = NumpyReadBin('data/TestingData.dat',3,'float32')
    #load sample labels
    trainL, trainLint = ReadLabels('data/TrainingLabels.txt')
    validateL, validateLint = ReadLabels('data/ValidationLabels.txt')
    testL, testLint = ReadLabels('data/TestingLabels.txt')

    import tensorflow as tf
    import numpy as np

    #get unique labels
    uniqueTrain = set()
    for l in trainLint:
        uniqueTrain.add(l)
    uniqueTrain = list(uniqueTrain)
    numClasses = len(uniqueTrain)
    numDims = len(trainData[0])

    #All features have real-value data
    feature_columns = [tf.feature_column.numeric_column("x", shape=[3])]

    # Build 3 layer DNN with 10, 20, 10 units respectively.
    classifier = tf.estimator.DNNClassifier(feature_columns=feature_columns,
                                              hidden_units=[10, 20, 10],
                                              n_classes=numClasses,
                                              model_dir="../Classification/tmp")

    # Define training inputs
    train_input_fn = tf.estimator.inputs.numpy_input_fn(
                                                x={"x": np.array(trainData)},y=np.array(trainLint),
                                                num_epochs = None, shuffle = True)

    #Train the model
    classifier.train(input_fn = train_input_fn, steps = 2000)

    #Define Validation inputs
    val_input_fn = tf.estimator.inputs.numpy_input_fn(
                                                x={"x": np.array(valData)},y=np.array(validateLint),
                                                num_epochs = 1, shuffle = False)

    # Evaluate accuracy.
    accuracy_score = classifier.evaluate(input_fn=val_input_fn)["accuracy"]
    print("\nTest Accuracy: {0:f}\n".format(accuracy_score))

if __name__ == '__main__':
    TensorflowDNN()

The Functions RedLabels(...) and NumpyReadBin(...) are loading my saved dataset in tensors. Since the labels are integer numbers that I read from a text file the function is a bit weird, but what I get in the end is an array with integers from tese labels: [11, 12, 21, 22, 23, 31, 32, 33, 41, 42].

However I am not able to classify anything, because upon calling classifier.train(input_fn = train_input_fn, steps = 2000), I get the following error:

...Traceback and stuff like that...
InvalidArgumentError (see above for traceback): assertion failed: [Label IDs must < n_classes] [Condition x < y did not hold element-wise:x (dnn/head/labels:0) = ] [[21][32][42]...] [y (dnn/head/assert_range/Const:0) = ] [10]
[[Node: dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert = Assert[T=[DT_STRING, DT_STRING, DT_INT64, DT_STRING, DT_INT64], summarize=3, _device="/job:localhost/replica:0/task:0/cpu:0"](dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert/Switch/_117, dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert/data_0, dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert/data_1, dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert/Switch_1/_119, dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert/data_3, dnn/head/assert_range/assert_less/Assert/AssertGuard/Assert/Switch_2/_121)]]

Did anyone come across this error before or has an idea how to solve it? I guess it is somehow complaining about the number of classes/format of labels in my dataset, but I know that trainLint contains 10 different classlabels, and that is the value of numClasses. Could it be the format of my trainLint array?

Answer 1

So the solution as Ishant Mrinal pointed out:

Tensorflow expects the integers from 0 up to the number of classes as class labels (range(0, num_classes)), not "arbitrary" numbers like in my case. Thanks!:)

...The other option I just came across is to add a label_vocabulary to the classifier-definition:

classifier = tf.estimator.DNNClassifier(feature_columns=feature_columns,
                                          hidden_units=[10, 20, 10],
                                          n_classes=numClasses,
                                          model_dir=saveAt,
                                          label_vocabulary=uniqueTrain)

With this option I can define the labels like I had before, converted to strings.