InvalidArgumentError error when evaluating new data in tensorflow

Question

I am trying to predict the classification of a new instance of data using the feed forward DNN graph in tensorflow.

This is a continuation of this thread, which was solved ValueError when trying to evaluate a new data instance in tensorflow

The code is:

import tensorflow as tf
import pandas as pd

dataframe = pd.read_csv("jfkspxstrain.csv") # Let's have Pandas load our dataset as a dataframe
dataframe = dataframe.drop(["Field6", "Field9", "rowid"], axis=1) # Remove columns we don't care about
dataframe.loc[:, ("y2")] = dataframe["y1"] == 0           # y2 is the negation of y1
dataframe.loc[:, ("y2")] = dataframe["y2"].astype(int)    # Turn TRUE/FALSE values into 1/0
trainX = dataframe.loc[:, ['Field2', 'Field3', 'Field4', 'Field5', 'Field7', 'Field8', 'Field10']].as_matrix()
trainY = dataframe.loc[:, ["y1", 'y2']].as_matrix()

dataframe = pd.read_csv("jfkspxstest.csv") # Let's have Pandas load our dataset as a dataframe
dataframe = dataframe.drop(["Field6", "Field9", "rowid"], axis=1) # Remove columns we don't care about
dataframe.loc[:, ("y2")] = dataframe["y1"] == 0           # y2 is the negation of y1
dataframe.loc[:, ("y2")] = dataframe["y2"].astype(int)    # Turn TRUE/FALSE values into 1/0
testX = dataframe.loc[:, ['Field2', 'Field3', 'Field4', 'Field5', 'Field7', 'Field8', 'Field10']].as_matrix()
testY = dataframe.loc[:, ["y1", 'y2']].as_matrix()

n_nodes_hl1 = 10
n_nodes_hl2 = 10
n_nodes_hl3 = 10

n_classes = 2
batch_size = 1

x = tf.placeholder('float',[None, 7])
y = tf.placeholder('float')

def neural_network_model(data):
    hidden_1_layer = {'weights':tf.Variable(tf.random_normal([7, n_nodes_hl1])),
                      'biases':tf.Variable(tf.random_normal([n_nodes_hl1]))}

    hidden_2_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl1, n_nodes_hl2])),
                      'biases':tf.Variable(tf.random_normal([n_nodes_hl2]))}

    hidden_3_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl2, n_nodes_hl3])),
                      'biases':tf.Variable(tf.random_normal([n_nodes_hl3]))}

    output_layer = {'weights':tf.Variable(tf.random_normal([n_nodes_hl1, n_classes])),
                      'biases':tf.Variable(tf.random_normal([n_classes]))}
    l1 = tf.add(tf.matmul(data, hidden_1_layer['weights']), hidden_1_layer['biases'])
    l1 = tf.nn.relu(l1)

    l2 = tf.add(tf.matmul(l1, hidden_2_layer['weights']), hidden_2_layer['biases'])
    l2 = tf.nn.relu(l2)

    l3 = tf.add(tf.matmul(l2, hidden_3_layer['weights']), hidden_3_layer['biases'])
    l3 = tf.nn.relu(l3)

    output = tf.matmul(l3, output_layer['weights']) + output_layer['biases']    
    return output

def train_neural_network(x):
    prediction = neural_network_model(x)
    cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(prediction,y))
    optimizer = tf.train.AdamOptimizer().minimize(cost)

    hm_epochs = 5

    with tf.Session() as sess:
        sess.run(tf.initialize_all_variables())

        for epoch in range(hm_epochs):
            epoch_loss = .1
            for _ in range(399):
                _, c = sess.run([optimizer, cost], feed_dict = {x: trainX, y: trainY})
                epoch_loss += c
            print('Epoch', epoch, 'completed out of', hm_epochs, 'loss:', epoch_loss)

        correct = tf.equal(tf.argmax(prediction,1), tf.argmax(y,1))
        accuracy = tf.reduce_mean(tf.cast(correct, 'float'))
        print('Accuracy:',accuracy.eval({x: testX, y: testY}))
        classification = y.eval(feed_dict={x: [[51.0,10.0,71.0,65.0,5.0,70.0,30.06]]})
        print (classification)
train_neural_network(x)

The error is:

InvalidArgumentError (see above for traceback): You must feed a value for placeholder tensor 'Placeholder_1' with dtype float
     [[Node: Placeholder_1 = Placeholder[dtype=DT_FLOAT, shape=[], _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

on this line

classification = y.eval(feed_dict={x: [[51.0,10.0,71.0,65.0,5.0,70.0,30.06]]})

Code and data is here: https://github.com/jhsmith12345/tensorflow

I see that this is asking for float values, which I thought I had fed it. Any help is appreciated!

Should it be classification = prediction.eval(...) instead of classification = y.eval(...) ? y is a tf.placeholder as targets of the neural network. — Jenny
– Jenny, Commented Feb 12, 2017 at 8:49

The Puternerd · Accepted Answer · 2017-02-12 21:35:04Z

So assuming that y is your target data, during training you need to provide a value for the target data so that the error can be calculated. The line classification = y.eval is obtaining a prediction, so there should be no requirement to provide training data - so should be classification = prediction.eval ...

 def train_neural_network(x):
  prediction = neural_network_model(x)
  cost =tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(prediction,y))
optimizer = tf.train.AdamOptimizer().minimize(cost)

  hm_epochs = 5

  with tf.Session() as sess:
    sess.run(tf.initialize_all_variables())

    for epoch in range(hm_epochs):
        epoch_loss = .1
        for _ in range(399):
            _, c = sess.run([optimizer, cost], feed_dict = {x: trainX, y: trainY})
            epoch_loss += c
        print('Epoch', epoch, 'completed out of', hm_epochs, 'loss:', epoch_loss)

    correct = tf.equal(tf.argmax(prediction,1), tf.argmax(y,1))
    accuracy = tf.reduce_mean(tf.cast(correct, 'float'))
    print('Accuracy:',accuracy.eval({x: testX, y: testY}))
    classification = prediction.eval(feed_dict={x: [[51.0,10.0,71.0,65.0,5.0,70.0,30.06]]})
    print (classification)

 train_neural_network(x)

Collectives™ on Stack Overflow

InvalidArgumentError error when evaluating new data in tensorflow

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related