Ytukyg

I'm implementing NDCG cost function for Conv neural network in tensorflow framework. The input data is an array of shape : (1499668, 6, 15, 1), target value is of shape (1499668, 6)

# log 2 with tensorflow

def log2(x):

    num = tf.log(x)

    den = tf.log(tf.constant(2,dtype=num.dtype))

    return num/den



#Create input placeholders

x = tf.placeholder("float", shape=[None, 6,15,1])

y = tf.placeholder("float", shape=[None, n_classes])

target_logs = tf.placeholder("float",shape=[None, n_classes])



#Define convolutional layer

def conv2d(x, W, b, strides=1, reuse=True):

    # Conv2D wrapper, with bias and relu activation

    x = tf.nn.conv2d(x, W, strides=[1, strides, strides, 1], padding='SAME')

    x = tf.nn.bias_add(x, b)

    return tf.nn.relu(x)



#Define Maxpool layer

def maxpool2d(x, k=2):

    return tf.nn.max_pool(x, ksize=[1, k, k, 1], strides=[1, k, k, 1],padding='SAME')



#Define a convolutional neural network function

def conv_net(x, weights, biases):  

    conv1 = conv2d(x, weights['wc1'], biases['bc1'])

    conv1 = maxpool2d(conv1, k=2)



    conv2 = conv2d(conv1, weights['wc2'], biases['bc2'])

    conv2 = maxpool2d(conv2, k=2)



    conv3 = conv2d(conv2, weights['wc3'], biases['bc3'])

    conv3 = maxpool2d(conv3, k=2)



    conv4 = conv2d(conv3, weights['wc4'], biases['bc4'])

    conv4 = maxpool2d(conv4, k=2)



    # Fully connected layer

    # Reshape conv2 output to fit fully connected layer input

    fc1 = tf.reshape(conv4, [-1, weights['wd1'].get_shape().as_list()[0]])

    fc1 = tf.add(tf.matmul(fc1, weights['wd1']), biases['bd1'])

    fc1 = tf.nn.relu(fc1)

    # Output, class prediction 

    out = tf.add(tf.matmul(fc1, weights['out']), biases['out'])

    return out



#Define Loss and Activation functions

pred = conv_net(x, weights, biases)

print(pred.shape)





# Get the indices of sorted predictions

sort_op,sort_indices = tf.nn.top_k(pred,k=6)

sort_op_act,sort_indices_act = tf.nn.top_k(y,k=6)

# Applying the sorting to log values

sort_log_vals = tf.gather(target_logs,sort_indices)





dcg_cost_a = tf.reduce_sum(tf.squared_difference(tf.reduce_sum(tf.divide(y,sort_log_vals),0),tf.reduce_sum(tf.divide(sort_op_act,target_logs),0)))

dcg_cost = tf.reduce_sum(tf.squared_difference(tf.reduce_sum(tf.divide(sort_op,target_logs),0),tf.reduce_sum(tf.divide(sort_op_act,target_logs),0)))

cr_ent_cost_sg = tf.losses.sigmoid_cross_entropy(y,pred)

tot_cost = dcg_cost + cr_ent_cost_sg + dcg_cost_a



#cost = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(logits=pred, labels=y))

#optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(tot_cost)



Optimizer = tf.train.AdamOptimizer()

optim = Optimizer.minimize(loss=dcg_cost)

optim2 = Optimizer.minimize(loss=cr_ent_cost_sg)

optim3 = Optimizer.minimize(loss=tot_cost)



#Evaluate Model

correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))



#calculate accuracy across all the given data and average them out. 

accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))



#Train and Test the Model

with tf.Session() as sess:

    sess.run(init) 

    train_loss = 

    test_loss = 

    train_accuracy = 

    test_accuracy = 

    summary_writer = tf.summary.FileWriter('./Output', sess.graph)

    for i in range(training_iters):

        for batch in range(len(train_np)//batch_size):

            batch_x = train_np[batch*batch_size:min((batch+1)*batch_size,len(train_np))]

            batch_y = train_np_y[batch*batch_size:min((batch+1)*batch_size,len(train_np_y))]

            print(batch_y.shape)

            print(batch_x.shape)

            # Run optimization op and Calculate batch loss and accuracy

            ##opt = sess.run(optimizer, feed_dict={x: batch_x,

            ##                                     y: batch_y})



            num_rows, num_cols = batch_y.shape



            y_logs = np.asarray([np.log2(np.arange(2,8)) for itr in range(0,num_rows)])

            print(y_logs.shape)



            v1,lv = sess.run([optim,dcg_cost],feed_dict={x:batch_x,y:batch_y,target_logs:y_logs})

            print(lv)

            v2,lv2 = sess.run([optim2,cr_ent_cost_sg],feed_dict={x:batch_x,y:batch_y,target_logs:y_logs})

            print(lv2)

            v3,lv3 = sess.run([optim3,tot_cost],feed_dict={x:batch_x,y:batch_y,target_logs:y_logs})

            print (lv+lv2,lv3)

Output is as follows

---------------------------------------------------------------------------

InvalidArgumentError                      Traceback (most recent call last)

/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in _do_call(self, fn, *args)

   1321     try:

-> 1322       return fn(*args)

   1323     except errors.OpError as e:



/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in _run_fn(feed_dict, fetch_list, target_list, options, run_metadata)

   1306       return self._call_tf_sessionrun(

-> 1307           options, feed_dict, fetch_list, target_list, run_metadata)

   1308 



/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in _call_tf_sessionrun(self, options, feed_dict, fetch_list, target_list, run_metadata)

   1408           self._session, options, feed_dict, fetch_list, target_list,

-> 1409           run_metadata)

   1410     else:



InvalidArgumentError: Incompatible shapes: [10000,6] vs. [10000,6,6]

     [[Node: truediv_44 = RealDiv[T=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_Placeholder_2_0_1, GatherV2_11)]]



During handling of the above exception, another exception occurred:



InvalidArgumentError                      Traceback (most recent call last)

<ipython-input-85-33963d934f68> in <module>()

     26             v2,lv2 = sess.run([optim2,cr_ent_cost_sg],feed_dict={x:batch_x,y:batch_y,target_logs:y_logs})

     27             print(lv2)

---> 28             v3,lv3 = sess.run([optim3,tot_cost],feed_dict={x:batch_x,y:batch_y,target_logs:y_logs})

     29             print (lv+lv2,lv3)

     30 



/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in run(self, fetches, feed_dict, options, run_metadata)

    898     try:

    899       result = self._run(None, fetches, feed_dict, options_ptr,

--> 900                          run_metadata_ptr)

    901       if run_metadata:

    902         proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)



/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in _run(self, handle, fetches, feed_dict, options, run_metadata)

   1133     if final_fetches or final_targets or (handle and feed_dict_tensor):

   1134       results = self._do_run(handle, final_targets, final_fetches,

-> 1135                              feed_dict_tensor, options, run_metadata)

   1136     else:

   1137       results = 



/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)

   1314     if handle is None:

   1315       return self._do_call(_run_fn, feeds, fetches, targets, options,

-> 1316                            run_metadata)

   1317     else:

   1318       return self._do_call(_prun_fn, handle, feeds, fetches)



/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/client/session.py in _do_call(self, fn, *args)

   1333         except KeyError:

   1334           pass

-> 1335       raise type(e)(node_def, op, message)

   1336 

   1337   def _extend_graph(self):



InvalidArgumentError: Incompatible shapes: [10000,6] vs. [10000,6,6]

     [[Node: truediv_44 = RealDiv[T=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_Placeholder_2_0_1, GatherV2_11)]]



Caused by op 'truediv_44', defined at:

  File "/usr/local/anaconda/lib/python3.6/runpy.py", line 193, in _run_module_as_main

    "__main__", mod_spec)

  File "/usr/local/anaconda/lib/python3.6/runpy.py", line 85, in _run_code

    exec(code, run_globals)

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel_launcher.py", line 16, in <module>

    app.launch_new_instance()

  File "/usr/local/anaconda/lib/python3.6/site-packages/traitlets/config/application.py", line 658, in launch_instance

    app.start()

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel/kernelapp.py", line 477, in start

    ioloop.IOLoop.instance().start()

  File "/usr/local/anaconda/lib/python3.6/site-packages/zmq/eventloop/ioloop.py", line 177, in start

    super(ZMQIOLoop, self).start()

  File "/usr/local/anaconda/lib/python3.6/site-packages/tornado/ioloop.py", line 888, in start

    handler_func(fd_obj, events)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tornado/stack_context.py", line 277, in null_wrapper

    return fn(*args, **kwargs)

  File "/usr/local/anaconda/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py", line 440, in _handle_events

    self._handle_recv()

  File "/usr/local/anaconda/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py", line 472, in _handle_recv

    self._run_callback(callback, msg)

  File "/usr/local/anaconda/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py", line 414, in _run_callback

    callback(*args, **kwargs)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tornado/stack_context.py", line 277, in null_wrapper

    return fn(*args, **kwargs)

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel/kernelbase.py", line 283, in dispatcher

    return self.dispatch_shell(stream, msg)

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel/kernelbase.py", line 235, in dispatch_shell

    handler(stream, idents, msg)

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel/kernelbase.py", line 399, in execute_request

    user_expressions, allow_stdin)

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel/ipkernel.py", line 196, in do_execute

    res = shell.run_cell(code, store_history=store_history, silent=silent)

  File "/usr/local/anaconda/lib/python3.6/site-packages/ipykernel/zmqshell.py", line 533, in run_cell

    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)

  File "/usr/local/anaconda/lib/python3.6/site-packages/IPython/core/interactiveshell.py", line 2698, in run_cell

    interactivity=interactivity, compiler=compiler, result=result)

  File "/usr/local/anaconda/lib/python3.6/site-packages/IPython/core/interactiveshell.py", line 2802, in run_ast_nodes

    if self.run_code(code, result):

  File "/usr/local/anaconda/lib/python3.6/site-packages/IPython/core/interactiveshell.py", line 2862, in run_code

    exec(code_obj, self.user_global_ns, self.user_ns)

  File "<ipython-input-72-9bbef16c3cc3>", line 13, in <module>

    dcg_cost_a = tf.reduce_sum(tf.squared_difference(tf.reduce_sum(tf.divide(y,sort_log_vals),0),tf.reduce_sum(tf.divide(sort_op_act,target_logs),0)))

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py", line 198, in divide

    return x / y

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py", line 847, in binary_op_wrapper

    return func(x, y, name=name)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/math_ops.py", line 955, in _truediv_python3

    return gen_math_ops.real_div(x, y, name=name)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/ops/gen_math_ops.py", line 5704, in real_div

    "RealDiv", x=x, y=y, name=name)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py", line 787, in _apply_op_helper

    op_def=op_def)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 3414, in create_op

    op_def=op_def)

  File "/usr/local/anaconda/lib/python3.6/site-packages/tensorflow/python/framework/ops.py", line 1740, in __init__

    self._traceback = self._graph._extract_stack()  # pylint: disable=protected-access



InvalidArgumentError (see above for traceback): Incompatible shapes: [10000,6] vs. [10000,6,6]

     [[Node: truediv_44 = RealDiv[T=DT_FLOAT, _device="/job:localhost/replica:0/task:0/device:CPU:0"](_arg_Placeholder_2_0_1, GatherV2_11)]]

The script errors out at : v3,lv3 = sess.run([optim3,tot_cost],feed_dict={x:batch_x,y:batch_y,target_logs:y_logs})
the previous computations (v1,lv1 and v2,lv2) works fine with the same dictionary inputs. Not quite sure what is causing this shape incompatibility. Appreciate any input.