repository stringclasses 156
values | issue title stringlengths 1 1.01k ⌀ | labels stringclasses 8
values | body stringlengths 1 270k ⌀ |
|---|---|---|---|
tensorflowtensorflow | top k doesn t order low index first if two element be equal | Bug | bug if two element be equal top k be not order they by index value reproduce the error here be the command I be run import tensorflow as tf sess tf session sess run tf nn top k tf convert to tensor 0 0 0 0 k 3 and here be the output for tensorflow 1 3 0 rc0... |
tensorflowtensorflow | the fixedlenfeature of parse example | Bug | view api doc the description maybe wrong each fixedlenfeature df map to a tensor of the specify type or tf float32 if not specify and shape serialize size df shape but the example show for dense result in two serialized example feature feature key age val... |
tensorflowtensorflow | window bazel build fail undeclared inclusion | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary source tensorflow version use command below 2... |
tensorflowtensorflow | error when use dataset map with num thread none | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes code attach below os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary binary tensorflow version use co... |
tensorflowtensorflow | training op test assertion failure on ubuntu 16 04 | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary source tensorflow version use command below ... |
tensorflowtensorflow | incorrect tfbinaryurl for instal with anaconda on linux | Bug | there be incorrect tfbinaryurl at tensorflow org in case of instal with anaconda in linux the example show the follow command for python 2 7 tensorflow pip install ignore instal upgrade but this be for python 3 4 so correct command be below tensorflow pip install ignore instal up... |
tensorflowtensorflow | memory leak in java api when use gpu | Bug | system information custom code os cento 7 tensorflow instal from source or binary binary tensorflow version use command below n a python version n a bazel version if compile from source n a cuda cudnn version 8 0 gpu model and mem... |
tensorflowtensorflow | a potential null pointer deference bug in graphpropertie class | Bug | dear developer I be study the code of tensorflow and find a potential null pointer dereference bug in tensorflow core grappler cost graph property cc the problem the function updateenter at line 70 of tensorflow core grappler cost graph property cc call the getcontext function of the shaper... |
tensorflowtensorflow | tensorflow python kernel test denormal test test failure on ppc64le error array be not equal | Bug | system information os platform and distribution e g linux ubuntu 16 04 ubuntu 16 04 ppc64le tensorflow instal from source or binary instal from source tensorflow version use command below v1 2 1 0 gb4957ff 1 2 1 python version py... |
tensorflowtensorflow | can t import graph contain mutablehashtable | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 linux mint 18 tensorflow instal from source or binary binary pip tensorflow version use command belo... |
tensorflowtensorflow | conv2d transpose produce different result on gpu | Bug | system information cat etc issue linux st 4 2 0 42 generic 49 14 04 1 ubuntu smp we d jun 29 20 22 11 utc 2016 x86 64 x86 64 x86 64 gnu linux version 14 04 4 lts trusty tahr version i d 14 04 be we in docker ... |
tensorflowtensorflow | error in doc | Bug | the doc here for bound box have an error the paragraph reading for example if an image be 100 x 200 pixel and the bounding box be 0 1 0 2 0 5 0 9 the bottom left and upper right coordinate of the bounding box will be 10 40 to 50 180 would well go something like for... |
tensorflowtensorflow | tf variable initializer seem break | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 ubuntu 14 04 tensorflow instal from source or binary the nightly wheel tensorflow version use command... |
tensorflowtensorflow | can t import graph contain batch sequence with state | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 linux mint 18 tensorflow instal from source or binary binary pip tensorflow version use command belo... |
tensorflowtensorflow | no mention of how to use custom runconfig with estimator in the estimator tutorial | Bug | on the tutorial of create estimator use tf contrib learn there doesn t seem to be any mention of how to create your own runconfig object in order to specify the configuration for an estimator run the configuration in particular I want to find be on how to write summary after custom sized step I eventually find it i... |
tensorflowtensorflow | no opkernel be register to support op rfft for cpu run on android | Bug | I m run tensorflow on android and get this error cause by java lang illegalargumentexception no opkernel be register to support op rfft with these attrs register device cpu register kernel I m use the master branch w... |
tensorflowtensorflow | contrib data dataset doc issue with dataset map tf py func in 1 3 0rc0 | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary 1 3 0rc0 python version 3 6 cuda cudnn version... |
tensorflowtensorflow | propose new github for tensorflow notebook | Bug | I would like to propose that we make a separate repository in github for jupyter notebooks I think it should be separate from the main tensorflow github so that people can check it out without have to check out all of tensorflow I ve create such a github and will work on it by myself for now but I think event... |
tensorflowtensorflow | doc window instruction need update | Bug | any way to contribute to the installation instruction for window I deal with a lot of people have installation instruction on stack overflow etc and there be a couple of thing on the install page that could do with be correct clarify or change happy to do these myself if you d like I ll list they below ... |
tensorflowtensorflow | tf contrib data dataset filter documentation break | Bug | with the old input pipeline function in tf train batch you could specify the allow small final batch parameter which would allow or disallow a small final batch with the new input pipeline function in tf contrib data the batch function allow a small final batch by default and to the good of my knowle... |
tensorflowtensorflow | import tensorflow break numpy linalg matrix power | Bug | 10771 system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 arch linux tensorflow instal from source or binary source tensorflow version use command below... |
tensorflowtensorflow | error in installation instruction | Bug | in instruction specify at validateyourinstallation it state for example the follow command install the cpu only version of tensorflow for python 2 7 tensorflow pip install ignore instal upgrade however it be for python 3 4 not 2 7 |
tensorflowtensorflow | xla crash on wasserstein gan | Bug | system information os platform and distribution e g linux ubuntu 16 04 ubuntu 16 04 tensorflow instal from source or binary tensorflow instal from source tensorflow version use command below v1 2 0 1382 g708cbaf 1 2 0 python version python 3 5 2 bazel vers... |
tensorflowtensorflow | fetch datum in distribute tensorflow have too much latency | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow the above be a simple benchmark which test the overhead of distribute tf it fetch a configurable sized variable from the parameter server and do a matmul on the worker it also do a matm... |
tensorflowtensorflow | error on tf contrib training stratify sample | Bug | I make a small example to illustrate which make some synthetic datum with unbalanced class and try to take balanced sample from it import tensorflow as tf from tensorflow python framework import op from tensorflow python framework import dtype batch size 10 datum a 9990 b 10 label... |
tensorflowtensorflow | miss definition in get start with tensorflow guide | Bug | system information macos 10 12 5 instal from binary tensorflow v1 2 0 1751 g43a819e13 1 2 1 python 3 6 1 bazel 0 5 2 homebrew describe the problem the custom model section of the getting start with tensorflow guide doesn t define eval input fn I run across the error wh... |
tensorflowtensorflow | why use x be dict to check y | Bug | in tensorflow contrib learn python learn learn io datum feeder py on l325 self y none if y be none else dict k check array v v dtype for k v in list y item if x be dict else check array y y dtype I happen to have a implementation where x be a dict but y be a numpy arr... |
tensorflowtensorflow | hang when fitting tensorflow learn model which contain cross sparse column with column and the datum be load with pandas input fn | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 run on databrick gpu cluster ubuntu 16 04 1 lts tensorflow instal from source or binary pip instal tensorflo... |
tensorflowtensorflow | go sigabrt when execute the same node more than once | Bug | problem in go when we pass the same node to the fetch list more then once sigabrt be raise source code log go package poc test import fmt tf github com tensorflow tensorflow tensorflow go github com tensorflow tensorflow tensorflow go op ... |
tensorflowtensorflow | correction to get start | Bug | hello I would like to submit that there be a programming bug on line 29 of the custom model tutorial within get start with tensorflow the line should read input fn tf contrib learn io numpy input fn x x train y train batch size 4 num epoch 1000 good regard jeff |
tensorflowtensorflow | tf nn elu incorrect second derivative | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 linux ubuntu 14 04 tensorflow instal from source or binary binary tensorflow version use command below ... |
tensorflowtensorflow | installation fail with virtualenv error the executable be not function tensorflow | Bug | operating system macos sierra 10 12 3 the version of virtualenv be 15 1 0 issue like these admindemacbook air admin virtualenv system site package tensorflow use base prefix anaconda new python executable in user admin tensorflow bin python dyld library not load loader path... |
tensorflowtensorflow | wrong path for retrain py | Bug | python tensorflow tensorflow example image retrain retrain py bottleneck dir bottleneck model dir inception summary dir training summary long output graph retrain graph pb output label retrain label txt image dir flower photo the path be wrong for retrain py with 40... |
tensorflowtensorflow | major performance hit when run two process on same gpu | Bug | I ve observe a major performance hit when run two identical model on the same gpu with allow growth set to true so that each process which only need a small fraction of the gpu memory 1 gb 11 gb be use when allow growth be set to true when run a single model on a single process it consistently finish go t... |
tensorflowtensorflow | tfdbg do not work with coordinator queuerunner | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 linux mint 18 tensorflow instal from source or binary binary pip tensorflow version use command below... |
tensorflowtensorflow | map func of tf contrib data dataset map get dict key instead of value when the nest structure of dataset be dict | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary source tensorflow version use command below ... |
tensorflowtensorflow | bug tf version 1 2 dropout layer reject tensor type rate | Bug | version 1 2 the follow code python tf contrib layer dropout tf zero 100 10 keep prob tf placeholder with default 0 9 dp or python tf layer dropout tf zero 100 10 1 tf placeholder with default 0 9 dp will fail with follow error txt ... |
tensorflowtensorflow | tf addgradient incorrectly report add have no gradient | Bug | system information os x with a cpu libtensorflow so build off master 2336cdf describe the problem tf addgradient report no gradient define for op add please see for instruction on how to add c gradient when ask to compute dz dx for z x y see code example ... |
tensorflowtensorflow | tensorflow 1 2 filewriter need to be create after tf text summary op | Bug | the follow work and create a tf text summary which I can find via tensorboard import tensorflow as tf sess tf interactivesession summary op tf summary text config config tf convert to tensor hello world summary writer tf summary filewriter tmp tensorboard sess graph text ... |
tensorflowtensorflow | tf summary filewriter doesn t support unicode path | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes in fact I use code from denny britz s cnn tutorial line 116 os platform and distribution e g linux ubuntu 16 04 window 10 x86 64 locale russian ... |
tensorflowtensorflow | tf summary comparison operator overloading | Bug | so I be try to use a comparison operation with a tf summary object it do not throw an error but the result be incorrect I couldn t find any help for this on the official documentation here be my logs pdb type loss 0 pdb type loss 1 pdb loss value tag c... |
tensorflowtensorflow | tensorboard histogram dashboard link break | Bug | the link to the tensorboard histogram dashboard documentation be break 404 if that section be currently in the process of be write perhaps add a dummy page say so |
tensorflowtensorflow | performance contirb seq2seq attention wrapper slow due to use matmul instead of reduce sum | Bug | tf version 1 2 0 rc0 contirb seq2seq attention wrapper be great it make use attention much easy however I find use attention wrapper will be much slow then tf version 1 0 after some experiment I find it be due to use matmul instead of reduce sum from attetntion wrapper py 731 expand ali... |
tensorflowtensorflow | gpu cpu memcpy fail error or internalerror c2c fft fail error when use fft2d | Bug | the problem I write a python script that use fft2d from tensorflow you can find the python script attach testfft2d py use gpu in this script I first create a convolution apply to a tensor and than compute a loss between a reference input tensor and a variable one with ftt2d and ifft2d operation wh... |
tensorflowtensorflow | stride behaviour different between caffe and tensorflow | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 ubuntu 14 04 tensorflow instal from source or binary binary pip install tensorflow version use com... |
tensorflowtensorflow | all the graph have an operation retval that make the train loop slow | Bug | I have try to train the model in tf cnn benchmark py and cifar10 multi gpu train py all the timeline result like the image bellow there be an operation name retval in the end of training loop but as we see long idle time be leave in the timeline how can I make the training fast by get rid of t... |
tensorflowtensorflow | tf nn max pool wrong doc | Bug | system information not applicable describe the problem api state that ksize have length 4 the size of window for each dimension of the input tensor however value be a 4 d tensor so doesn t this mean that ksize should be length 4 same for stride dig into maxpoole op cc show that... |
tensorflowtensorflow | tensorflow contrib keras python keras model throw error for a valid keras code | Bug | system information cat etc issue linux parikshit xps l322x 4 4 0 79 generic 100 ubuntu smp we d may 17 19 58 14 utc 2017 x86 64 x86 64 x86 64 gnu linux version 16 04 2 lts xenial xerus version i d 16 04 version codename xenial be we i... |
tensorflowtensorflow | poor scalability of tensorflow multigpu training on a single machine performance bug | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow os platform and distribution e g linux ubuntu 16 04 debian 3 16 7 tensorflow instal from source or binary source hash b1e174e tensorflow version use comma... |
tensorflowtensorflow | tf image resize bicubic produce artifact in output image | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes see below os platform and distribution e g linux ubuntu 16 04 all platform tensorflow instal from source or binary source tensorflow version use command... |
tensorflowtensorflow | documentation cifar10 input py be reference but miss on master | Bug | describe the problem file cifar10 input py be reference in documentation but be miss on master documentation piece r0 7 version |
tensorflowtensorflow | break link for cifar 10 tutorial | Bug | on the main page of tensorflow tutorial for cifar 10 I e model input url the link in the code organisation code organization section be all break to be precise the on click they github page say page not find please remove this bug asap |
tensorflowtensorflow | darwin support for mkl build configure script | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 mac os x 10 12 4 darwin tensorflow instal from source or binary source git tag 1 2 0 rc 2 tensorf... |
tensorflowtensorflow | inner tf device inherit device index when use wildcard index | Bug | tf version v1 2 0 rc0 735 gf48673b about one week ago python with tf device gpu 7 with tf device cpu print tf constant 0 device device cpu 7 workaround python with tf device gpu 7 with tf device none tf device cpu print t... |
tensorflowtensorflow | strange performance sparse tensor matmul in kernel test | Bug | sorry for the previous issue accord to the document here tensorflow python sparse tensor dense matmul op test benchmark a sparse m k with nonzero value between 1 and 80 b dense k n when I run sparse tensor dense matmul op test py with large n m k I get nnz n ... |
tensorflowtensorflow | run session use c api be significantly slow than use python | Bug | system information os platform and distribution linux 13 1 bottle tensorflow instal from source tensorflow version 1 1 0 git version v1 1 0 1 g10ec24a compiler version v1 1 0 1 g10ec24a compiler c suse linux 4 8 1 20130909 gcc 4 8 branch revision 202388 ... |
tensorflowtensorflow | memoryerror when freeze large model | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 ubuntu 16 04 tensorflow instal from source or binary source tensorflow versio... |
tensorflowtensorflow | tf train scaffold miss global step attribute | Bug | accord to the docstring of tf train scaffold there be a global step attribute with the follow description global step a tensor contain the global step counter pick from and store into the global step collection in the graph by default see l84 the problem be that no such att... |
tensorflowtensorflow | undocumented change in share variable from version 1 0 to 1 1 | Bug | system information custom code a minimal reproducible example provide below linux ubuntu 14 04 tensorflow instal from binary use pip tensorflow version 1 0 1 and 1 1 0 cuda 8 0 cudnn 5 1 geforce gtx 1080 problem variable scope and sharing work d... |
tensorflowtensorflow | clarify installation doc for window 32 bit | Bug | quick feature doc request I ve notice a few people on stack overflow ask about run tensorflow on 32 bit machine I m aware that some people have have some success as describe by mrry s post here s o tensorflow on 32 bit linux one example of a confused user try to use window s o how to ... |
tensorflowtensorflow | bug beamsearchdecoder should not assume that when time 0 beam will be full | Bug | score flat control flow op cond time 0 lambda array op reshape score batch size 1 lambda score 0 num available beam control flow op cond time 0 lambda math op reduce prod score shape 1 lambda math op reduce prod score... |
tensorflowtensorflow | recordinput documentation in api doc site | Bug | recordinput be in the tf 1 2 code l1602 but I don t see it in the tf 1 2 api doc should it be present |
tensorflowtensorflow | feature request a guide for create custom deepdream model | Bug | the current deepdream guide locate here use the inception5h model from what I can tell it do not appear very straightforward in term of how to fine tune the model in order to create different deepdream hallucination from a custom datum set it also do not appear to be relatively easy to change the model that ... |
tensorflowtensorflow | memory leak | Bug | I have a memory leak with tensorflow I refer to to address my issue and I follow the advice of the answer that seem to have solve the problem however it do not work here in order to recreate the memory leak I have create a simple example first I use this function that I get here to check the... |
tensorflowtensorflow | contrib learn python learn io datum feeder py y be dict instead of x be dict at line 325 | Bug | version r1 2 I be try to create a regressor with x as dict and y as ndarray get error that at line mention in title say ndarray object as no attribute item when I check the code it look like x be dict y be dict isinstance x dict y be not none and isinstance y dict if isinstanc... |
tensorflowtensorflow | doc arg scope override define in path | Bug | I find that some op have incorrect define in path in the contrib layer doc example avg pool2d batch norm bias add conv2d conv2d in plane conv2d transpose dropout flatten fully connect layer norm one hot encoding separable conv2d softmax it always use tensorflow contrib ... |
tensorflowtensorflow | lack support of qint32 in tf nn tanh | Bug | accord to the doc the tanh operation support float point input as well as fix point input of type qint32 however in the late master a typeerror raise when run follow code import tensorflow as tf sess tf interactivesession x tf constant 1 2 3 dtype tf float32 from te... |
tensorflowtensorflow | tensorflowdebugger do not dump stack pack concat nod | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 linux ubuntu 14 04 tensorflow instal from source or binary binary tensorflow version use command bel... |
tensorflowtensorflow | c online documentation codeblock not format | Bug | there be some codeblock in c documentation write with github style fence markdown that be not render as eg screenshot of a page with doc not format as code it look like something be go wrong with the site generation that when translate markdown it do not pickup these block in t... |
tensorflowtensorflow | can not run bazel build inception download and preprocess imagenet | Bug | it always give error no such target inception download and preprocess imagenet target inception download and preprocess imagenet not declare in package define by home hank tensorflow build the build file be empty |
tensorflowtensorflow | how to know what command and type be support in io build | Bug | accord to this comment issuecomment 302817142 and other related issue currently there be some command and type that io user can t load from frozen pb graph so we could use any tf api in python but not in io it be hard to guess what python api will not be support in io so be there any documentation or ... |
tensorflowtensorflow | tf contrib distribution undocumented behaviour in multinomial and categorical | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 macos sierra 10 12 3 tensorflow instal from source or binary binary via pip tensorflow version use c... |
tensorflowtensorflow | tensorflow 1 2 0rc0 strange behavior with benchmark script | Bug | I m run script by tfboyd and I m observe strange behavior and crash 1 there s a lot of message like these below these message happen in both standalone and distribute run 2017 05 22 17 21 00 589871 I tensorflow core common runtime gpu gpu device cc 996 create tensorflow device gpu 3 ... |
tensorflowtensorflow | warn for get start tensorflow core tutorial a custom model | Bug | system information have I use stock example script provide in tensorflow from os platform and distribution ubuntu 16 04 2 lts gnu linux 4 4 0 78 generic x86 64 tensorflow instal from binary tensorflow version v1 1 0 rc0 61 g1ec6ed5 1 1 0 cuda cudnn version not u... |
tensorflowtensorflow | tf estimator vs contrib very unclear difference and lot of deprecation | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 windows ubuntu tensorflow instal from source or binary binary tensorflow version use command below ... |
tensorflowtensorflow | issue break or outdate link link for nlp tutorial | Bug | the second link in the highlight session of the webpage vector representation be break or outdate this be the link that s not work 404 from github |
tensorflowtensorflow | valueerror refuse to perform an overparameterized separable convolution | Bug | please go to stack overflow for help and support if you open a github issue here be our policy 1 it must be a bug or a feature request 2 the form below must be fill out here s why we have that policy tensorflow developer respond to issue we want to focus on work that benefit the whol... |
tensorflowtensorflow | contrib layers avg pool2d raise warning when serialize metagraph | Bug | system information os platform and distribution ubuntu 14 04 lt tensorflow instal from source tensorflow version v1 1 0 rc2 1015 gf2047a3 1 1 0 rc2 bazel version 0 4 5 cuda cudnn version 8 0 5 1 5 gpu model and memory maxwell titan x 12 gib... |
tensorflowtensorflow | go sigsegv when use int32 instead of int64 and miss error in resize function | Bug | problem in go some operation cause a sigsegv when use an int32 instead of an int64 and I have reason to believe that the same will happen when use float instead of double and vice versa the resize operation don t define the output shape correctly when the input be not a batch ... |
tensorflowtensorflow | wrong hyperlink in web page tutorial | Bug | there be a wrong html link in text in the tutorial vector representation of word at address the text be in the first paragraph after the bullet point in the highlight section at the top of the page the link for tensorflow model tutorial embed word2vec py be wrong it should point to the f... |
tensorflowtensorflow | session run method s feed dict argument implicitly convert type | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 osx 10 12 3 tensorflow instal from source or binary binary tensorflow version use command below 1... |
tensorflowtensorflow | xla abort core dump | Bug | os ubuntu linux 16 04 tensorflow compile from source tensorflow version r1 1 bazel version 0 4 5 cuda cudnn version 8 0 5 1 gpu model memory titanx 12 gb after turn on xla jit compile tf fail with a core dump 2017 05 14 14 50 38 673877 I tensorflow core common runtime gpu gpu ... |
tensorflowtensorflow | www tensorflow org version incorrectly list 0 12 as the most recent stable branch | Bug | say that 0 12 be most recent stable branch which be cause new user to install the old version issuecomment 301211285 |
tensorflowtensorflow | distribute runtime leak memory on window | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow no os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary binary tensorflow version use command below b unk... |
tensorflowtensorflow | unexpected error at contrib seq2seq s beamsearchdecoder | Bug | system information os platform and distribution e g linux ubuntu 16 04 macos tensorflow instal from source or binary source tensorflow version use command below v1 1 0 rc2 773 g7fa0cf39f bazel version if compile from source 0 4 5 cuda cudnn version ... |
tensorflowtensorflow | tf abs isn t document to handle complex but it do appear to work as tf complex ab use to | Bug | issue 7405 be a bug file that tf complex ab be remove in 1 0 at the bottom it say that tf abs now do that work but the doc for tf abs only mention float I confirm that tf abs do in fact do as the comment on the issue describe l225 please incorporate the info from complex ab into ... |
tensorflowtensorflow | c api exception | Bug | hi the c api method tf graphgettensornumdim throw an exception with message node x be not find in the graph even for valid node that be in the graph this tend to happen with the output of particular op e g reshape and matmul I think it may have to do with shape inference after look at the c api i... |
tensorflowtensorflow | tensorflow consume much more memory than expect | Bug | my model have four cpu variable 500 m 3 tf int32 500 m tf float32 500 m tf float32 ftrl accumulate slot 500 m tf float32 ftrl linear slot expect memory consumption should be 500 m 6 4 12 g however tensorflow use 20 g memory when I increase 500 m to 1b total memory usa... |
tensorflowtensorflow | not a jpeg issue | Bug | bazel bin tensorflow example image retrain retrain image dir training image look for image in non human look for image in human create bottleneck at tmp bottleneck non human datum negative jpeg cr night 512x384 cr night 512x384 rcr m26 a10 d2005 04 07 t22 38 wn jpg txt 2017 05 09 01 56 48 8900... |
tensorflowtensorflow | memory leak when implement rnn attention decoder | Bug | system information cat etc issue linux quad 4 4 0 72 generic 93 ubuntu smp fri mar 31 14 07 41 utc 2017 x86 64 x86 64 x86 64 gnu linux version 16 04 lts xenial xerus version i d 16 04 be we in docker ... |
tensorflowtensorflow | large page fault cause slow performance while use gpu | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary binary use pip install tensorflow gpu 1 0 1 ... |
tensorflowtensorflow | segmentation fault on help tf train sequenceexample | Bug | I be try to try and test the tf train sequenceexample class but it wasn t in the official doc and use help crash python do anyone know how to use it or how to solve this issue |
tensorflowtensorflow | optimize compile tensorflow use almost 2x more memory than non optimize binary | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary both tensorflow version use command below ... |
tensorflowtensorflow | more description for optimizer adagrad adadelta ftrl | Bug | be it necessary to add more description for optimizer such as adagrad adadelta ftrl and so on just as what adam do since these be several quite new optimizer and I think it s well to show user more detail about these optimizer so that they can understand why do these optimizer work well than sgd in some situ... |
tensorflowtensorflow | miss image in | Bug | system information n a describe the problem include an image but this do not exist gather png doesn t seem to exist anywhere in the tf repository |
tensorflowtensorflow | installation instruction for conda install tensorflow in the root environment | Bug | the instruction installinganaconda say to 1 create a new empty environment 2 activate it 3 install tensorflow via pip but pip be not instal in the new environment so the third command will call the first pip inside the path system variable that usually be the pip instal in the root ... |
tensorflowtensorflow | tensorboard graph vizualisation crash with chrome safari | Bug | server system information dockerfile image from tensorflow tensorflow late gpu run pip install tensorflow upgrade run pip install tensorflow gpu upgrade cat etc issue linux 479a65b403e2 4 ... |
tensorflowtensorflow | convolutional neural network tutorial problem | Bug | I be learn about convolutional neural network from the tutorial inside it there be a link for get the code but it be not work how can I get the code |
tensorflowtensorflow | sigsegv with sparse add and broadcast | Bug | system information have I write custom code as oppose to use a stock example script provide in tensorflow yes enclose below os platform and distribution e g linux ubuntu 16 04 ubuntu 16 04 tensorflow instal from source or binary binary via pip tensorflow version... |
tensorflowtensorflow | be batch norm param argument miss in the depthwise convolution 2d layer implementation | Bug | this be with reference to l1850 the line say if batch norm param be none but there be no batch norm param argument include in the function and it doesn t seem that batch norm be implement within the function as an option be the batch norm function include in the regularizer function or have it not be... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.