repository
stringclasses
156 values
issue title
stringlengths
1
1.01k
labels
stringclasses
8 values
body
stringlengths
1
270k
tensorflowtensorflow
tf2 tensorarray with multi dimensional tensor write in it will be stack into none shape tensor in autograph mode
Bug
system information os platform and distribution e g linux ubuntu 16 04 linux lz 5 4 0 48 generic 52 18 04 1 ubuntu tensorflow instal from source or binary in conda env pip install tensorflow gpu I tensorflow version use command below v2 3 0 rc2 23 gb36436b087 2 3 0 ...
tensorflowtensorflow
wrong maximum value pass in the positional encoding for the transformer model
Bug
url s with the issue description of issue what need change when create positional encoding vector set we specify maximum position encoding which should be equal to the maximum possible length of the sequence feed into the model so it must be equal to max length which be 40 in your sample it...
tensorflowtensorflow
compile client code with mavx2 lead to eigen error
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 10 14 mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mobile device n a ...
tensorflowtensorflow
tensorflow keras backend dot do not work as expect when second argument be 3 dimensional or high
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow yes but reproduce this bug require very minimal code os platform and distribution e g linux ubuntu 16 04 manjaro linux rolling release tensorflow instal from source...
tensorflowtensorflow
bug in tf
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
tf feature column embed column wrong dtype in lookup variable
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 18 10 mobile device e g iphone 8 pixel 2 samsung galaxy if the issue ha...
tensorflowtensorflow
post quant kera model with upsample2d after post quantize get a false output shape
Bug
system information os platform and distribution e g linux ubuntu 16 04 tensorflow instal from source or binary pip tensorflow version or github sha if from source 2 3 0 command use to run the converter or code if you re use the python api if possible please share a link to ...
tensorflowtensorflow
compile tf2 3 with mkl on osx do not progress beyond first epoch
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 osx 10 15 6 mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mobile device no tensorflow ...
tensorflowtensorflow
detection postprocess operator miss
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 tensorflow instal from source or binary tensorflow version commit sha if source 9aff666a8db689168e2d3aaef17b8a252791ada7 target platform e g arm mbe os arduino nano 33 etc ...
tensorflowtensorflow
tf1 keras model error on loading use tf2 indexerror list index out of range
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 18 04 tensorflow instal from source or binary binary tensorflow version use command below ...
tensorflowtensorflow
spurious tf function retracing warning when develop keras layer in colab
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
add int16 support to the cmsis nn version of softmax
Bug
tensorflow micro the softmax reference kernel have be update with int16 support and the cmsis nn version should to update with int16 support as well
tensorflowtensorflow
tf numpy function do not support symbolic tensor
Bug
system information I have write custom code see below red hat enterprise linux server release 7 7 maipo tensorflow instal from binary pip tensorflow version v2 3 0 rc2 23 gb36436b087 2 3 0 python version 3 6 9 describe the current behavior call tf numpy function w...
tensorflowtensorflow
nnapi delegate bug
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
ethos u scratch tensor not allocate
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 tensorflow instal from source or binary tensorflow version commit sha if source b023b033299a2e88a8f25980e234144657df5fc2 target platform e g arm mbe os arduino nano 33 etc et...
tensorflowtensorflow
duplicate node name in graph one
Bug
hi I have the follow error test the yolov3 example 2020 09 24 11 28 23 492471 I tensorflow core platform cpu feature guard cc 142 your cpu support instruction that this tensorflow binary be not compile to use avx2 traceback most recent call last file c user c81441 appdata local continuum ...
tensorflowtensorflow
tf tensorscatterupdate conversion to tflite
Bug
system information os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary python binary tensorflow version or github sha if from source 2 3 provide the text output from tflite convert error tf tensorscatterupdate op b...
tensorflowtensorflow
keras save model give different accuracy compare to the original model
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
valueerror unknown layer when I be load of the model use tf keras model load model
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow true os platform and distribution e g linux ubuntu 16 04 colab tensorflow version use command below late hi I be face a model loading issue use tf keras model load mod...
tensorflowtensorflow
can not convert centernet keypoint model zoo tf2 model to tflite
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information os platform and distribution e g linux ubuntu 16 04 ubuntu 18 04 4 lts tensorflo...
tensorflowtensorflow
bug in tf math rsqrt
Bug
hello on tf 2 3 about the function tf math rsqrt x name none the documentation say x a tf tensor must be one of the follow type bfloat16 half float32 float64 int32 when run with int32 type import tensorflow as tf x tf constant 2 0 2 tf math rsqrt x ...
tensorflowtensorflow
tf dynamic stitch give wrong shape and output raw memory
Bug
system information linux ubuntu 20 04 tensorflow 2 3 0 instal from binary python version 3 8 2 default jul 16 2020 14 00 26 gcc 9 3 0 no gpu available describe the current behavior when stitch empty partition the output shape be wrong python tf dynamic stitch 1 ...
tensorflowtensorflow
cmsis nn conv unit test case fail for non unity dilation case
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary source tensorflow version commit sha if source 712fd5cfe4a5de420a4226c874664423eba4cb1a target platform e g arm mbe os ...
tensorflowtensorflow
unexpected accuracy output from tf keras model evaluate call on a save and loaded model
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
a commit message should be unit64
Bug
in the commit message diff f0675f2568ff9470bcfc1f2bc79c5386 I find that it say int64 but when I check the api documentation it seem that it should be unit64 please look at parameter description tf2 3 mod tf2 2 mod
tensorflowtensorflow
tf 2 3 concrete function spec issue list index out of range
Bug
in tf 2 3 if we define a model in class it error out with follow error when run predict usr local lib python3 6 dist package tensorflow python eager function py in structured signature check arg type self args kwargs 1778 arg spec kwarg spec self structure input signature 17...
tensorflowtensorflow
gpudelegate produce incorrect result for reduce sum
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
tf lite optimize default hybrid model be not support on tflite micro
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 macos catalina 10 15 6 tensorflow instal from source or binary 2 3 0 tensorflow version commit sha if source target platform e g arm mbe os arduino nano 33 etc esp32 de...
tensorflowtensorflow
different gradient in tf2 when eager mode be enable compare to graph mode
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 18 04 tensorflow instal from source or binary binary tensorflow version use command below ...
tensorflowtensorflow
undocumented notfounderror occur when use tf io gfile glob to retrieve file in a non existent directory
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 chrome os linux 5 4 40 04224 g891a6cce2d44 mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mo...
tensorflowtensorflow
model and layer set to non trainable but weight still adjust
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow custom code os platform and distribution e g linux ubuntu 16 04 google colab tensorflow instal from source or binary google colab tensorflow version use command below goo...
tensorflowtensorflow
should we remove deprecate and ignore field version in graph proto
Bug
we notice that the version field in graph proto have be deprecate since 5 year ago when we try to generate graphdef scala code by sbt protoc from this proto file it produce annoying deprecation warning during the compilation stage should we remove it from tensorflow s codebase l24...
tensorflowtensorflow
crash when attempt to use xla conv with complex input
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 debian 10 4 tensorflow instal from source or binary binary tensorflow version use command below v1 12 1 41557 gae0...
tensorflowtensorflow
usage of intermediate layer cause an exception
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow yes os platform and distribution window 10 tensorflow instal from source or binary binary tensorflow version use command below 2 3 python version 3 7 5 cuda cudnn version 10 ...
tensorflowtensorflow
fail to run on the give interpreter regular tensorflow op be not support by this interpreter make sure you apply link the flex delegate before inference node number 62011 flexsize fail to prepare
Bug
system information os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary pip install tf nightly tensorflow version or github sha if from source 2 4 0 dev20200916 provide the text output from tflite convert for android studi...
tensorflowtensorflow
tflite converter abort dump core in broadcastadd4dslow
Bug
system information os platform and distribution e g linux ubuntu 16 04 linux pop os 20 04 lt tensorflow instal from source or binary binary tensorflow version or github sha if from source 2 3 0 command use to run the converter or code if you re use the python api if poss...
tensorflowtensorflow
rnn can not get save tflite model
Bug
system information os platform and distribution ubuntu 16 04 tensorflow instal from binary tensorflow version tf nightly 2 2 0 command use to run the converter or code if you re use the python api convert pb2 tflite txt the output log from the converter invocation log ...
tensorflowtensorflow
unexpected behaviour for model evaluate inside keras callback
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow yes tensorflow instal from source or binary colab tensorflow version use command below 2 3 0 python version 3 6 9 describe the current behavior inside a keras callb...
tensorflowtensorflow
self attention on word embedding use half precision with mask zero set to true crash train
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 18 04 5 lts tensorflow instal from source or binary binary tensorflow version use command below v1 12 1 415...
tensorflowtensorflow
mobilenetv3 small large not fully quantize after full integer post training quantization run with no error
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 n a use google colaboratory tensorflow version use the command in the template v1 12 1 39569 gcb34190201 2 4 0 dev2020081...
tensorflowtensorflow
textvectorization not work on tpu with tf nightly
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 colab tensorflow version use command below tf nightly 2 4 0 dev20200915 use tpu yes I be try to run textvectorizati...
tensorflowtensorflow
feature request mlir base tflite converter support for 16bit conv2d
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 18 04 1 tensorflow instal from source or binary source tensorflow version ...
tensorflowtensorflow
broadcast not work for divide and divide no nan
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 catalina mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mobile device na tensorf...
tensorflowtensorflow
keras layers api documentation not sort properly
Bug
url s with the issue description of issue what need change very minor issue but the overview list of all layer class be not consistently sort alphabetically though it very much feel like it be intend to e g the gru layer be list above the gaussiandropout layer in the sidebar it be s...
tensorflowtensorflow
converter error for convert save model to tflite
Bug
system information os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary pip install tensorflow tensorflow version or github sha if from source 2 3 0 also instal tf nightly command use to run the converter or code if you re us...
tensorflowtensorflow
runtimeerror quantization not yet support for op
Bug
system information os platform and distribution linux ubuntu 18 04 tensorflow instal from source or binary binary tensorflow version or github sha if from source tf 2 3 0 command use to run the converter or code if you re use the python api def representative datum gen ...
tensorflowtensorflow
sparse tensor error message when apply constraint to dense variable
Bug
system information os platform and distribution ubuntu 18 04 tensorflow instal from source or binary conda tensorflow version I try version 2 2 0 2 3 0 1 15 and 1 14 python version 3 6 and 3 8 3 cuda cudnn version cuda 10 1 and cudnn 7 6 for tf 2 3 0 and 2 2 0 cuda 10 0 and...
tensorflowtensorflow
subclass of tf keras model throw notimplementederror when fit with custom datum
Bug
tf 2 3 0 test on window and in colab there be an error that come out only when certain type of model subclass of tf keras model be combine with certain custom dataloader python generator and subclass of tf keras util sequence yet same loader work ok with simple functional keras model the error be ...
tensorflowtensorflow
syncbatchnormalization have nan loss with channel first format
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 run in docker container host be 18 04 tensorflow instal from source or binary source tensorflow v...
tensorflowtensorflow
custom xception input size broadcast incorrectly
Bug
system information red hat enterprise linux server release 7 7 maipo tensorflow tensorflow gpu 2 1 gpu nvidia v100 python 3 6 10 gcc 7 3 0 cuda 10 1 I think run through slurm the error log valueerror could not broadcast input array from shape 850 550 3 into shape ...
tensorflowtensorflow
just qustion in android java keras model be ok
Bug
hi I m develop android app and my nn in python be keras tf keras with lstm conv2d resnet attention very deep I want to use this nn in java android 1 can I use tf lite for my kera of python I hear tf lite only support some tf model so I m worried 2 and can I convert kera to tensor...
tensorflowtensorflow
int16 softmax be use std vector
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 debian tensorflow instal from source or binary source tensorflow version commit sha if source 35d9474383c9befd99f457031ada977d681742c4 target platform e g arm mbe os arduino nan...
tensorflowtensorflow
tf datum experimental service throw error when use with tpus
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
after several iteration the error input and filter must have the same depth be arise
Bug
I be try to fit my model by use a generator work on my dataset audio file I use a gold cluster slurm to run my program include generator for create my input feature and corresponding target by use yeilde method and safe threading and fit generator for train my model unfortunetly after 388 iteration...
tensorflowtensorflow
tf signal ifft return result with different dtype for subsequent call with the same parameter
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 debian 10 4 tensorflow instal from source or binary binary tensorflow version use command below v1 12 1 40510 ga32...
tensorflowtensorflow
tf function give different output than standard function
Bug
no custom code tf 2 3 I be try to make a backt differentiable so I decide to use tf function s auto differentiation capability despite take a very long time to compile the function finally it get convert into tf op however the loss I get when run a tf version of the function and the python versio...
tensorflowtensorflow
no gradient define for operation raggedtensorfromvariant or no gradient at all
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
tf lite fail conversion with post train int quantization convert fine without runtimeerror mismatch between number of weight maxs and channel
Bug
hey there I ve manage to convert stylegan2 to a tf lite model with your recent advice however now upon try to quantize the model I m get an error the model convert with no issue if quantization be not use it also convert fine with dynamic range quantization only fail if a representative dataset be...
tensorflowtensorflow
cudnn error when pass all mask sequence to rnn layer on gpu
Bug
system information have I write custom code yes os platform and distribution cento 8 tensorflow instal from binary pip tensorflow version 2 3 0 v2 3 0 rc2 23 gb36436b087 python version 3 6 8 cuda cudnn version 10 1 7 gpu model and memory rtx 6000 24 gb de...
tensorflowtensorflow
graph mode failure with mask fed rnn layer within a train step loop on gpu
Bug
system information have I write custom code yes os platform and distribution linux mint 19 1 cento 8 tensorflow instal from binary pip tensorflow version 2 2 0 v2 2 0 rc4 8 g2b96f3662b python version 3 6 9 cuda cudnn version 10 1 7 gpu model and memory quadro p...
tensorflowtensorflow
dense to ragged batch fail with map function implement with py function
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
batch training with model fit not work for all batch size
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 windows tensorflow instal from source or binary binary via anaconda tensorflow version use comma...
tensorflowtensorflow
link doesn t work
Bug
link for tensorflow graph optimization with grappler from url performance comparison do not exist the link point to page which give page not find error
tensorflowtensorflow
reproduce internal ci error via tf lite micro makefile
Bug
tensorflow micro see this comment issuecomment 689199971 for an internal error due to wswitch that we could not reproduce via the tflm makefile the underlying issue there be a miss wswitch in the makefile and an inability to build without dtf lite static memory
tensorflowtensorflow
model predict problem when use one datum at a time
Bug
system information have I write custom code yes os platform and distribution docker tensorflow tensorflow 2 2 0 gpu tensorflow version use command below 2 2 0 python version 3 6 9 cuda cudnn version v10 1 243 gpu model and memory titan x pascal 12go ...
tensorflowtensorflow
pylint incorrectly identify tensorflow public api function in tensorflow 2 2
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow apart from the example below no os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 osx 10 15 6 mobile device e g iphone 8 pixel 2 samsung galaxy if the ...
tensorflowtensorflow
tensorflow2 tutorials actor critic method
Bug
url s with the issue description of issue what need change clear description at the second cell we want to install additional package for visualization maybe should be add sudo apt get update before the line sudo apt get install y xvfb python opengl dev null 2 1 le...
tensorflowtensorflow
same variable name for tf feature column embed column embed tensor in tf 2 3 0
Bug
system information have I write custom code yes os platform and distribution 64 bit linux tensorflow instal from source tensorflow version v2 3 0 rc2 23 gb36436b087 2 3 0 python version 3 6 10 cuda compilation tool release 10 0 v10 0 130 describe the current behavior ...
tensorflowtensorflow
valueerror tensor type variable initializer must either be wrap in an init scope or callable in tpu stragegy
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
conv1d and probablyall other conv layer with dilation rate 1 do not reliably handle change in input size
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
large number be multiply with the embed output
Bug
url s with the issue the url with the issue be description of issue what need change clear description while I be check the transformer code from this tensorflow documentation in the encoder subsection under encoder and decoder section I find that before add posit...
tensorflowtensorflow
microinterpreter tensor size always return zero
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 linux fedora 32 tensorflow instal from source or binary source tensorflow version commit sha if source 238981a91a9b780ab4449829469a71c5d668a273 target platform e g arm mbe os ar...
tensorflowtensorflow
cmsis nn incorrect flag for non dsp mve processor convolution op only
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary source tensorflow version commit sha if source 5a16264ba6f12883726d12d484d4cd61405ddab7 target platform e g arm mbe os ...
tensorflowtensorflow
more detail on the reduce codesize tag use for arc
Bug
tensorflow micro this come up during the review for pr 42020 in this comment discussion r483160984 the tflm team would like to understand well what this reduce codesize tag be do in micro examples micro speech arc emsdp makefile topic of discussion in the current design tag be mos...
tensorflowtensorflow
tensorflow lite micro make it possible to specify optimization level when build with the makefile
Bug
tensorflow micro it would be useful to be able to manually set the optimization level for different build type s from the make command my idea be that the default optimization flag should be as it be now none for build type debug o3 for build type release if the user want to use different optimization...
tensorflowtensorflow
abort core dump tf 2 3 tf lite model convert but crash on invoke
Bug
hey there I m try to convert stylegan2 to tf lite currently there seem to be an issue with invoke the synthesis block of the generator so I ve include a small model here which be just one synthesis block to display the issue convert fine in tf 2 3 with no issue then on attempt to invoke I get the...
tensorflowtensorflow
matrix do not consider sample weight
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow provide below os platform and distribution e g linux ubuntu 16 04 ubuntu mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mobile device tensorflo...
tensorflowtensorflow
non responsive model when build micro speech with cmsis nn
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary source tensorflow version commit sha if source d3cdadd target platform e g arm mbe os arduino nano 33 etc mbe os stm32f...
tensorflowtensorflow
fail to convert sparsetensor to tensor
Bug
I m use tensorflow 2 3 0 and kera 2 4 3 on ubunto 20 04 the code work okay on tensorflow 2 1 0 and kera 2 3 1 class args def init self dataset path dataset train mymodel output my model h5 label output le pickle embedding output embedding pickle image out ...
tensorflowtensorflow
bug when use num parallel call when mapping dataset to tfa function
Bug
as mention over the issue here and advise from other contributor I m create this issue cause use num parallel call tf data experimental autotune inside the map call from my dataset appear to generate a deadlock I ve test with tensorflow version 2 2 and 2 3 and tensorflow addon 0 11 1 and 0 10 0 ...
tensorflowtensorflow
deprecate function setdiff1d still use in the tf source code
Bug
system information I have write custom code linux ubuntu 20 04 tensorflow 2 3 instal use pip python 3 8 2 current unexpected behavior calculate the gradient of the reduce prod function raise this warning warn tensorflow from home prasanth local pythonuserbase lib python3 ...
tensorflowtensorflow
experimental step per execution option discard when load a model from checkpoint
Bug
system information have I write custom code yes os platform and distribution google colab tpu tensorflow version 2 3 python version 3 describe the current behavior I have create and train a model compile with the experimental step per execution option when I load the mod...
tensorflowtensorflow
documentation be become too undecipherable for a meaningful practical work
Bug
the documentation be become too undecipherable for any practical application use tensorflow look at this one how would anyone learn anything for a stream datum from disk not many how tos for datum that don t fit into memory and that page just like a man page for api we end up waste many man hour wi...
tensorflowtensorflow
tflite c java experimental kernel ctc beam search decoder return always buffer length length 1
Bug
system information os platform and distribution e g linux ubuntu 16 04 ubuntu 20 04 mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mobile device pixel 2 tensorflow instal from source or binary source tensorflow version master python version ...
tensorflowtensorflow
unexpected behaviour of tf keras activation relu
Bug
system information tensorflow version v2 3 0 0 gb36436b087 2 3 0 describe the current behavior when pass np nan to tf keras activations relu it return 0 0 this only happen when use gpu the relu activation behave as expect under cpu it return nan describe the expect behavio...
tensorflowtensorflow
uninitialized memory access of per channel param
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 linux ubuntu 16 04 tensorflow instal from source or binary source tensorflow version commit sha if source 5a16264ba6f12883726d12d484d4cd61405ddab7 target platform e g arm mbe os ...
tensorflowtensorflow
convert keras h model to tflite fail with error window fatal exception access violation
Bug
system information os platform and distribution e g linux ubuntu 16 04 tensorflow instal from source or binary tensorflow version or github sha if from source command use to run the converter or code if you re use the python api if possible please share a link to colab jupy...
tensorflowtensorflow
conv3d
Bug
system information os platform and distribution e g linux ubuntu 16 04 tensorflow instal from source or binary tensorflow version or github sha if from source provide the text output from tflite convert some of the operator in the model be not support by the standard tensor...
tensorflowtensorflow
tflite micro resize near neightbour test fail for build type debug
Bug
system information os platform and distribution e g linux ubuntu 16 04 linux ubuntu 18 04 mobile device e g iphone 8 pixel 2 samsung galaxy if the issue happen on mobile device tensorflow instal from source or binary source tensorflow version sha 1 source 44dbace063796e...
tensorflowtensorflow
tf2 can not convert efficientdet d0 to save model tflite
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
cmsis nn conv kernel write outside of allocate memory when num channel 256
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 n a tensorflow instal from source or binary source tensorflow version commit sha if source b36436b087bd8e8701ef51718179037cccdfc26e target platform e g arm mbe os arduino nano 3...
tensorflowtensorflow
autograph unable to process lambda statement tf 2 3 0
Bug
tf 2 0 python c import tensorflow as tf print tf version git version tf version version v2 3 0 rc2 23 gb36436b087 2 3 0 problem description python import tensorflow as tf assert tf version 2 3 0 dataset tf datum dataset from tensor slice 1 2 3 map lam...
tensorflowtensorflow
op require fail at conv op cc 1115
Bug
system information have I write custom code as oppose to use a stock example script provide in tensorflow no use code from cnn tutorial os platform and distribution e g linux ubuntu 16 04 arch linux kernel 5 8 3 arch1 1 tensorflow instal from source or binary python tensorflow opt...
tensorflowtensorflow
memory leak when use tensorflow c api
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
clarify that tf image adjust brightness accept negative value
Bug
url s with the issue description of issue what need change the documentation state that delta should be in the range 0 1 however the delta be add to the image so negative value in the range 1 0 be also valid and useful if the image should be make dark clear descripti...
tensorflowtensorflow
datahandler intermittent invalidargumenterror for large input dimension
Bug
please make sure that this be a bug as per our github policy we only address code doc bug performance issue feature request and build installation issue on github tag bug template system information have I write custom code as oppose to use a stock example script provide in tensorflow ...
tensorflowtensorflow
tflite c api docs seem break
Bug
url s with the issue description of issue what need change the c api reference for tflite be miss many important component on tensorflow org for example the doc for tflite flatbuffermodel and tflite interpreterbuilder be completely go which use to be there the last time I ...
tensorflowtensorflow
add optimize svdf kernel from cmsis nn
Bug
tensorflow micro system information host os platform and distribution e g linux ubuntu 16 04 na tensorflow instal from source or binary na tensorflow version commit sha if source na target platform e g arm mbe os arduino nano 33 etc arm cortex m an optimize vers...
tensorflowtensorflow
calculation of effective scale differ from tflite implementation in quantize kernel
Bug
tensorflow micro describe the problem in the quantize kernel the calculation of the effective scale differ slightly between tflite and tflu we find that in some case this result in single bit difference in output be this difference in implementation intentional l129 l133 l75 l79
tensorflowtensorflow
can not convert predict function of linearregressor
Bug
system information os platform and distribution e g linux ubuntu 16 04 window 10 tensorflow instal from source or binary binary pip tensorflow version or github sha if from source 2 4 0 dev20200824 command use to run the converter or code if you re use the python api i...