python - Tensorflow report error with the restored training model -


i'm newbie machine learning , tensorflow, using example tutorial source code, model trained , printed accuracy, doesn't include source code export model , variables , import predict new image.

so revised source code export model, , create new python script predict using test data set.

here source code export training model: 

mnist = input_data.read_data_sets(flags.data_dir, one_hot=true) print("run here3") # create model x = tf.placeholder(tf.float32, [none, 784], name="x") w = tf.variable(tf.zeros([784, 10]), name="w") b = tf.variable(tf.zeros([10])) y = tf.matmul(x, w) + b saver = tf.train.saver() sess = tf.interactivesession() ... ignore source code cost function definition , train model #after model trained, save variables , y  tf.add_to_collection('w', w) tf.add_to_collection('b', b) tf.add_to_collection('y', y)  saver.save(sess, 'result')

in new python script, try restore model , re-execute y function 

sess = tf.session() saver = tf.train.import_meta_graph('result.meta') saver.restore(sess, tf.train.latest_checkpoint('./')) w = tf.get_collection('w')[0] b = tf.get_collection('b')[0] y = tf.get_collection('y')[0]   mnist = input_data.read_data_sets('/tmp/tensorflow/mnist/input_data', one_hot=true) img = mnist.test.images[0] x = tf.placeholder(tf.float32, [none, 784]) sess.run(y, feed_dict={x: mnist.test.images})

everything works correctly, w , b value if print them, error while executing last statement(run y function). 

caused op u'x', defined at: file "predict.py", line 58, in <module> tf.app.run(main=main, argv=[sys.argv[0]] + unparsed) file "/users/zhouqi/git/machine-learning/tensorflow/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 44, in run _sys.exit(main(_sys.argv[:1] + flags_passthrough)) file "predict.py", line 25, in main saver = tf.train.import_meta_graph('result.meta') file "/users/zhouqi/git/machine-learning/tensorflow/lib/python2.7/site- packages/tensorflow/python/training/saver.py", line 1566, in    import_meta_graph **kwargs) file "/users/zhouqi/git/machine-learning/tensorflow/lib/python2.7/site-packages/tensorflow/python/framework/meta_graph.py", line 498, in import_scoped_meta_graph producer_op_list=producer_op_list) file "/users/zhouqi/git/machine-learning/tensorflow/lib/python2.7/site-packages/tensorflow/python/framework/importer.py", line 288, in import_graph_def op_def=op_def) file "/users/zhouqi/git/machine-learning/tensorflow/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 2327, in create_op original_op=self._default_original_op, op_def=op_def) file "/users/zhouqi/git/machine-learning/tensorflow/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 1226, in __init__ self._traceback = _extract_stack()  invalidargumenterror (see above traceback): must feed value placeholder tensor 'x' dtype float [[node: x = placeholder[dtype=dt_float, shape=[], _device="/job:localhost/replica:0/task:0/cpu:0"]()]]

it's weird cause use same statement define x, , feed x using same approach while executing y function, don't know why not work?

the problem new placeholder:

x = tf.placeholder(tf.float32, [none, 784])

creating placeholder same name not enough. need same placeholder used when created model. therefore must add x collection in first file:

tf.add_to_collection('x', x)

and load in new file:

x = tf.get_collection('x')[0]

instead of creating new one.


Comments

Post a Comment

Popular posts from this blog

Command prompt result in label. Python 2.7 -

i want put result(output) of shell command label. command works, label shows "0", however, in command prompt, result shown correctly, need showed in label. doing ubuntu. myg1 = button(root, text="rodyti informacija", command=lambda: gauti()) myg1.pack(side=bottom) def gauti(): imti = tekstas.get("1.0", "end-1c") info = subprocess.call("id '{imti}' ".format(imti=imti), shell=true) w = label(root, text= info) w.pack(side = bottom) subprocess.call() returns exit code of process created (in case 0 success). if want text output of process, should instead call subprocess.check_output() .
Read more

javascript - How do I use URL parameters to change link href on page? -

i change href value of link using url parameters. example: default <a id="dlink" href="link-1">link</a> but, when user goes https://url.com/?dlink=link-2 it swaps out link new link <a id="dlink" href="link-2">link</a> first need value url. if you're not using javascript framework can create function using jquery. please check get url parameter jquery or how query string values in js it. after that, element "a" id , set value there. $("#dlink").attr("href", hreffromparam);
Read more

amazon web services - AWS Route53 Trying To Get Site To Resolve To www -

so have domain nameservers pointed aws route 53. site hosted on ec2 instance , reachable resolves site.com, if input www.site.com. i'm curious how site resolve www.site.com. have under route53 domains hosted zone 2 additional record sets outside of ns , soa: a site.com ec2 elastic ip cname www.site.com ec2 public dns i can reach site using both site.com , www.site.com, resolve site.com. there way make if enter either, resolves www.site.com? at first, both should record set. a site.com ec2 elastic ip www.site.com ec2 elastic ip and need redirect setting in web server (nginx, apache, ..) following. (source: how redirect www non-www nginx on centos 7 ) redirect non-www www if want redirect users plain, non-www domain www domain, add server block: server { server_name example.com; return 301 $scheme://www.example.com$request_uri; } save , exit. configures nginx redirect requests "example.com" "www.example.com". not...
Read more