python - Tensorboard not corrrectly logging precision and recall metrics -


i using tensorboard log cross-entropy , accuracy successfully, precision , recall graphs wrong when logged tf.metrics.recall , tf.metrics.precicion.

my problem 6 class classification problem. know manual calculations outside of tensorboard precision should ~99% @ 80% recall, graphs recorded tensorboard show flat 16% precision @ 100% recall.

the precision stat logged curiously close 1/6th corresponds random selection of 6 output nodes.

the code below:

    def classifier_graph(x, y, learning_rate=0.1):             """                     build graph classification, given our input layer x ,                     output layer, y.             """              tf.name_scope('classifier'):                     tf.name_scope('model'):                             w = tf.variable(tf.zeros([xdim, ydim]), name='w')                             b = tf.variable(tf.zeros([ydim]), name='b')                             y_ = tf.matmul(x, w) + b                      tf.name_scope('cross_entropy'):                             diff = tf.nn.softmax_cross_entropy_with_logits(labels=y, logits=y_)                             cross_entropy = tf.reduce_mean(diff)                             summary = tf.summary.scalar('cross_entropy', cross_entropy)                      tf.name_scope('train'):                             #cross_entropy = tf.reduce_mean(-tf.reduce_sum(y * tf.log(y_), reduction_indices=[1]), name='cross_en$                             train_step = tf.train.gradientdescentoptimizer(learning_rate).minimize(cross_entropy)                             #minimise cross_entropy via gd                      tf.name_scope('accuracy'):                             tf.name_scope('correct_prediction'):                                     correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))                             tf.name_scope('accuracy'):                                     accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))                                     tf.summary.scalar('accuracy', accuracy)                       tf.name_scope('metrics'):                             _, recall = tf.metrics.recall(y, y_ )                             _, precision = tf.metrics.precision(y, y_)                              v_rec = tf.summary.scalar('recall', recall)                             v_prec = tf.summary.scalar('precision', precision)              metrics = tf.summary.merge_all()              return [w, b, y_, cross_entropy, train_step, metrics]        def train_classifier(insamples, outsamples, batch_size, iterations, feature_set_index=1, model=none, device="/gpu:0"):          x = tf.placeholder(tf.float32, [none, xdim], name='x') # none indications arbitrary first dimension         y = tf.placeholder(tf.float32, [none, ydim], name='y')          w, b, y_, cross_entropy, train_step, metrics  = classifier_graph(x, y)          tf.session(config=config) sess, tf.device(device):              init = tf.global_variables_initializer()             init_l = tf.local_variables_initializer()              sess.run(init)             sess.run(init_l)              file_writer = tf.summary.filewriter(logdir, tf.get_default_graph())              all_classifier_results, all_models, all_err, all_recall, all_precision = [],[],[],[],[]              t = 0             while t < iterations:                 batch_x, batch_y = batch_feed(insamples, batch_size, feature_set_index)                 t += 1                 _, err, metrics_str, = sess.run([train_step, cross_entropy, metrics], feed_dict={x: batch_x, y: batch_y })                  all_err.append(err)                  file_writer.add_summary(metrics_str,t)          return 'done'

any ideas on why might be? thanks. x, y , y_ numpy arrays.


Comments

Post a Comment

Popular posts from this blog

android - InAppBilling registering BroadcastReceiver in AndroidManifest -

i'm implementing inapp billing in android app. works well, however, i'm trying decouple broadcast receiver activity manifest. suggestion in android's trivialdrive sample: // important: dynamically register broadcast messages updated purchases. // register receiver here instead of <receiver> in manifest // because call getpurchases() @ startup, therefore can ignore // broadcasts sent while app isn't running. // note: registering listener in activity bad idea, done here // because sample. regardless, receiver must registered after // iabhelper setup, before first call getpurchases(). currently there's class extends broadcastreceiver : public class iabbroadcastreceiver extends broadcastreceiver { /** * intent action receiver should filter for. */ public static final string action = "com.android.vending.billing.purchases_updated"; private final iabbroadcastlistener mlistener; public iabbroadcastreceiver(iabbroadcastlistener listener) { m...
Read more

python Tkinter Capturing keyboard events save as one single string -

is possible store keyboard events 1 string? code below stores 1 char , prints it. card reader or bar code reader, contains collection of character/ string not 1 character @ time. goal save char pressed text variable. from tkinter import * root = tk() def key(event): text= event.char text+= event.char print ("pressed", text) def callback(event): frame.focus_set() print ("clicked at", event.x, event.y) frame = frame(root, width=100, height=100) frame.bind("<key>", key) frame.bind("<button-1>", callback) frame.pack() root.mainloop() currently, creating text variable , everytime key() function gets called, , text stores last character typed. you can define text module level variable , , use module level text inside key function - text = '' def key(event): global text text+= event.char print("pressed", text)
Read more

sql server - Why does Linq-to-SQL add unnecessary COUNT()? -

here tables: create table customer ( cid int primary key, name varchar(32) ) create table ord ( oid int primary key, cid int foreign key references customer, address varchar(20) ) here's linq-to-sql statement: var aaa = c in db.customer select new { c, o = c.ord.tolist() }; and here's inexplicable query generated linq-to-sql: select [t0].[cid] [cid], [t0].[name] [name], [t1].[oid] [oid], [t1].[cid] [cid2], [t1].[address] [address], (select count(*) [dbo].[ord] [t2] [t2].[cid] = [t0].[cid]) [value] [dbo].[customer] [t0] left outer join [dbo].[ord] [t1] on [t1].[cid] = [t0].[cid] order [t0].[cid], [t1].[oid] i want understand how rid of count(*) part. it's entirely uncalled for! linq needs count(*) in order determine number of entries returned o = c.ord.tolist() .
Read more