Tensorflow Modell sparen: GraphDef kann nicht größer sein als 2 GB

Question

ich die folgende Fehlermeldung bekommen - anscheinend zu der Zeit meines Modell des Sparens

Step = 1799 | Tensorflow Accuracy = 1.0 
Step = 1799 | My Accuracy = 0.0363355780022 
Step = 1800 | Tensorflow Accuracy = 1.0 
Step = 1800 | My Accuracy = 0.0364694929089 
Traceback (most recent call last): 
    File "CNN-LSTM-seg-reg-sigmoid.py", line 290, in <module> 
    saver.save(sess, save_path) 
    File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/training/saver.py", line 1085, in save 
    self.export_meta_graph(meta_graph_filename) 
    File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/training/saver.py", line 1103, in export_meta_graph 
    add_shapes=True), 
    File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2175, in as_graph_def 
    result, _ = self._as_graph_def(from_version, add_shapes) 
    File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 2138, in _as_graph_def 
    raise ValueError("GraphDef cannot be larger than 2GB.") 
ValueError: GraphDef cannot be larger than 2GB. 

Here für tf.constant s achten vorgeschlagen, aber ich habe null Konstanten in meinem Programm. Jedoch sind meine weights und biases wie folgt: tf.Variable(tf.random_normal([32]),name="bc1"). Könnte das ein Problem sein?

Wenn nicht, dann sagt mir this, dass ich irgendwo nach jeder Schleife Iteration zu der Grafik hinzufügen, aber ich bin unsicher, wo es auftritt.

Meine erste Schätzung ist, wenn ich Vorhersagen mache. Ich mache die Prognosen über die Code folgenden ...

# Make prediction 
im = Image.open('/home/volcart/Documents/Data/input_crops/temp data0001.tif') 
batch_x = np.array(im) 
batch_x = batch_x.reshape((1, n_input_x, n_input_y)) 
batch_x = batch_x.astype(float) 
prediction = sess.run(pred, feed_dict={x: batch_x}) 
prediction = tf.sigmoid(prediction.reshape((n_input_x * n_input_y, n_classes))) 
prediction = prediction.eval().reshape((n_input_x, n_input_y, n_classes)) 

Meine zweite Vermutung ist, wenn ich loss und accuracy über die folgende Berechnung: loss, acc = sess.run([cost, accuracy], feed_dict={x: batch_x, y: batch_y})

Mein ganzer Session-Code sieht wie folgt aus:

# Initializing the variables 
init = tf.initialize_all_variables() 
saver = tf.train.Saver() 

gpu_options = tf.GPUOptions() 
config = tf.ConfigProto(gpu_options=gpu_options) 
config.gpu_options.allow_growth = True 

# Launch the graph 
with tf.Session(config=config) as sess: 
    sess.run(init) 
    summary = tf.train.SummaryWriter('/tmp/logdir/', sess.graph) #initialize graph for tensorboard 
    step = 1 
    # Import data 
    data = scroll_data.read_data('/home/volcart/Documents/Data/') 
    # Keep training until reach max iterations 
    while step * batch_size < training_iters: 
     batch_x, batch_y = data.train.next_batch(batch_size) 
     # Run optimization op (backprop) 
     batch_x = batch_x.reshape((batch_size, n_input_x, n_input_y)) 
     batch_y = batch_y.reshape((batch_size, n_input_x, n_input_y)) 
     batch_y = convert_to_2_channel(batch_y, batch_size) 
     sess.run(optimizer, feed_dict={x: batch_x, y: batch_y}) 
     step = step + 1 

     loss, acc = sess.run([cost, accuracy], feed_dict={x: batch_x, 
                  y: batch_y}) 


     # Make prediction 
     im = Image.open('/home/volcart/Documents/Data/input_crops/temp data0001.tif') 
     batch_x = np.array(im) 
     batch_x = batch_x.reshape((1, n_input_x, n_input_y)) 
     batch_x = batch_x.astype(float) 
     prediction = sess.run(pred, feed_dict={x: batch_x}) 
     prediction = tf.sigmoid(prediction.reshape((n_input_x * n_input_y, n_classes))) 
     prediction = prediction.eval().reshape((n_input_x, n_input_y, n_classes)) 

     # Temp arrays are to splice the prediction n_input_x x n_input_y x 2 
      # into 2 matrices n_input_x x n_input_y 
     temp_arr1 = np.empty((n_input_x, n_input_y)) 
     for i in xrange(n_input_x): 
      for j in xrange(n_input_x): 
       for k in xrange(n_classes): 
        if k == 0: 
         temp_arr1[i][j] = 1 - prediction[i][j][k] 

     my_acc = accuracy_custom(temp_arr1,batch_y[0,:,:,0]) 

     print "Step = " + str(step) + " | Tensorflow Accuracy = " + str(acc) 
     print "Step = " + str(step) + " | My Accuracy = " + str(my_acc) 

     if step % 100 == 0: 
      save_path = "/home/volcart/Documents/CNN-LSTM-reg-model/CNN-LSTM-seg-step-" + str(step) + "-model.ckpt" 
      saver.save(sess, save_path) 
      csv_file = "/home/volcart/Documents/CNN-LSTM-reg/CNNLSTMreg-step-" + str(step) + "-accuracy-" + str(my_acc) + ".csv" 
      np.savetxt(csv_file, temp_arr1, delimiter=",") 

Answer 1

Sie wachsen Ihr Diagramm in dieser Zeile:

prediction = tf.sigmoid(prediction.reshape((n_input_x * n_input_y, n_classes))) 

Dies konvertiert Ihr prediction Numpy-Array in den TensorFlow-Konstantenknoten, fügt es in das Diagramm ein und fügt zusätzlich Sigmoid Knoten hinzu.

Sie können Probleme wie diese fangen tf.get_default_graph().finalize() durch Zugabe vor dem Training Schleife