tensorflow / tensorflow

21 Pull requests merged by 14 people

• [ROCm] Update profiler support for rocm

  #50388 merged Sep 2, 2021

• [PluggableDevice] add DEVICE_DEFAULT for collective/bcast ops

  #50609 merged Sep 1, 2021

• [XLA] Enable pointwise row vectorization for small row.

  #50756 merged Sep 1, 2021

• [PluggableDevice] add DEVICE_DEFAULT for debug/stage ops

  #50610 merged Sep 1, 2021

• Fix crash with tf.summary.create_file_writer when non-scalar values are passed

  #51715 merged Aug 31, 2021

• Add use_default_shell_env = True to all ctx.actions.run_shell rules

  #44549 merged Aug 31, 2021

• Fix typo `tpu_system_medata`

  #51660 merged Aug 31, 2021

• Add GPU kernel for SparseConcat

  #51391 merged Aug 31, 2021

• [XLA] Add back: lift bitcast PR

  #51580 merged Aug 30, 2021

• [PluggableDevice] add DEVICE_DEFAULT for session/transpose ops

  #50623 merged Aug 30, 2021

• Add GpuSelectFlagged helper function and test

  #51696 merged Aug 30, 2021

• Fix crash with tf.range when start is large/negative

  #51711 merged Aug 30, 2021

• Add missing deps for Legalize to Linalg

  #51735 merged Aug 30, 2021

• [Intel oneDNN] update curl version to 7.78.0

  #51661 merged Aug 30, 2021

• Add `--config=rbe_lite_linux` in .bazelrc

  #51567 merged Aug 30, 2021

• Add back "PR #49173: [Crash fix] Fix cudaMallocAsync crashes."

  #50961 merged Aug 30, 2021

• Fix TF serving build failure on s390x caused by aws lib target

  #51656 merged Aug 27, 2021

• Flatbuffer_conversion changes for Slice

  #51670 merged Aug 27, 2021

• replace x86intrin.h to immintrin.h

  #51553 merged Aug 27, 2021

• [TF:TRT] Enable dynamic batch dim for Conv2dBackpropInput

  #51468 merged Aug 26, 2021

• Fix `DeprecationWarning: the imp module is deprecated in favour of importlib`

  #51663 merged Aug 26, 2021

23 Pull requests opened by 16 people

• [INTEL MKL] Add Conv3D fusions

  #51702 opened Aug 26, 2021

• The following upstream commits caused a 15% performance degradation o…

  #51703 opened Aug 26, 2021

• [PluggableDevice] Add the option to force memory growth

  #51705 opened Aug 27, 2021

• [tflite] fix default of xnnpack_delegate_provider

  #51706 opened Aug 27, 2021

• Fix crash of tf.image.pad_to_bounding_box with large input value.

  #51717 opened Aug 27, 2021

• [Intel MKL] Fuse BiasAdd and Gelu with MatMul

  #51722 opened Aug 27, 2021

• Fix tf.math.segment_max/min/mean/sun/prod crashes(aborts) when segment_ids is large

  #51733 opened Aug 28, 2021

• I have reservations

  #51739 opened Aug 30, 2021

• "ZeroDivisionError: integer division or modulo by zero" while backpropagating #51653

  #51747 opened Aug 30, 2021

• [tf.data] graduate choose_from_datasets API from experimental to tf.data.Dataset

  #51753 opened Aug 30, 2021

• [INTEL MKL] Skip CUDA specific remapper tests

  #51754 opened Aug 30, 2021

• [PluggableDevice] Add range checks to TF_ExpectedOutputDataType in non-debug builds

  #51758 opened Aug 31, 2021

• [PluggableDevice] Add TF_IsHostMemoryInput and TF_IsHostMemoryOutput to the C API

  #51759 opened Aug 31, 2021

• [PluggableDevice] add int32 DEVICE_DEFAULT register for Unpack/Sum ops

  #51761 opened Aug 31, 2021

• [PluggableDevice] add int32 DEVICE_DEFAULT for Switch/ParallelDynamicStitch

  #51762 opened Aug 31, 2021

• Updated log_poisson_loss() function

  #51765 opened Aug 31, 2021

• [PluggableDevice] add int32 DEVICE_DEFAULT for BroadcastTo/Pack/StridedSlice

  #51777 opened Sep 1, 2021

• Tflite GPU OpenGL Delegate Allow HWC Const Add

  #51781 opened Sep 1, 2021

• `distutils` is deprecated in Python 3.10 #51776

  #51782 opened Sep 1, 2021

• [TFLite] Add INT8 and INT16x8 support to NON_MAX_SUPPRESSION_V4 / V5 operators

  #51785 opened Sep 1, 2021

• fix op does not dominate this use after mhlo-fusion pass

  #51788 opened Sep 1, 2021

• [PluggableDevice] Add TF_OpkernelConstruction_GetAttrNamesList

  #51791 opened Sep 2, 2021

• [TFLite] Change rounding mode of the quantized RESIZE_BILINEAR op for coherence with MultiplyByQuantizedMultiplier

  #51794 opened Sep 2, 2021

128 Issues closed by 44 people

• tf.contrib.training.HParams domain interval support

  #26436 closed Sep 2, 2021

• LayerNormalization crashes on empty inputs when run on CPU

  #46366 closed Sep 2, 2021

• Segmentation fault in tf.image.combined_non_max_suppression

  #42421 closed Sep 2, 2021

• DepthwiseConv1D: implementation, check and warning message

  #36935 closed Sep 2, 2021

• Can't find GPUs after updating tensorflow to 2.6.0 (with pip) in conda environment

  #51684 closed Sep 2, 2021

• How does TFLite model and TFLite-quantized model converted?

  #49119 closed Sep 2, 2021

• TensorFlow master build fails on s390x due to boringssl issue

  #51770 closed Sep 2, 2021

• Bug with optimizer_v2.OptimizerV2.set_weights method

  #50859 closed Sep 2, 2021

• Repeat weights in dense layer / Forum registration not possible

  #51784 closed Sep 2, 2021

• Crash in sess.run on CPU after restoring model from checkpoint.

  #29486 closed Sep 2, 2021

• gpu doesn't report error when id is actually out of range in tf.nn.embedding_lookup(embed, ids)

  #27033 closed Sep 2, 2021

• KeyError: 'ExperimentalFunctionBufferingResource' in Tf >= 1.13 (mkl)

  #27752 closed Sep 2, 2021

• What architecture do block3 and stack3 from tf.python.keras.applications.resnet correspond to?

  #51550 closed Sep 2, 2021

• esrgan failed on image over 50x50 pixels : error output tensor

  #51551 closed Sep 2, 2021

• tensorflow saved model to tf dialect mlir

  #51233 closed Sep 2, 2021

• Failed to build v2.3.0 from source in debug mode

  #42427 closed Sep 2, 2021

• Cannot run GPU on tensorflow Docker Image with Tensorflow 2.3.1

  #45011 closed Sep 1, 2021

• Tensorflow 2.3.1 is not compatible with cuda 10.2 on Windows

  #44622 closed Sep 1, 2021

• C API Release

  #39174 closed Sep 1, 2021

• Setting of Number of Threads for TFLite Interpreter API

  #50822 closed Sep 1, 2021

• Missing headers (TensorFlow C, 2.4.0, Windows)

  #46343 closed Sep 1, 2021

• Error while running sample cpp program in tensorflow

  #24123 closed Sep 1, 2021

• tfdbg memory leak

  #24513 closed Sep 1, 2021

• TFDBG Segmentation Fault

  #24817 closed Sep 1, 2021

• ValueError: No gradients provided for any variable

  #51489 closed Sep 1, 2021

• metric_learning的损失函数如何应用在检索中，目前实现中都存在传labels给到度量损失函数

  #51546 closed Sep 1, 2021

• MLIR tests fail to build with non-system gcc and later glibc

  #50873 closed Sep 1, 2021

• C- How to draw bounding-box using tensorflow c_api

  #25480 closed Sep 1, 2021

• tf.control_dependencies() not work in a custom_gradient function

  #19168 closed Sep 1, 2021

• Tensorflow crashes when running the profiler with CUDA 11

  #46036 closed Sep 1, 2021

• Results are different on MacOS for fixed data

  #20308 closed Sep 1, 2021

• Inconsistency In CategoricalCrossentropy Calculation

  #51720 closed Sep 1, 2021

• Unable to build micro_speech for bluepill target (region `FLASH' overflowed by 9656 bytes)

  #30433 closed Sep 1, 2021

• Can not Exporting a GraphDef from file

  #30851 closed Sep 1, 2021

• google.protobuf.json_format.MessageToDict incorrectly decodes bytes_list

  #31250 closed Sep 1, 2021

• set_weights crashes in custom train_step() method

  #47675 closed Sep 1, 2021

• tensorflow.data.experimental.load not working inside tensorflow.data.Dataset().flatmap

  #50594 closed Sep 1, 2021

• tf.GradientTape.gradients() does not support graph control flow operations

  #51534 closed Sep 1, 2021

• Building tests fails on POWER

  #45228 closed Sep 1, 2021

• SavedModel imported in ML.NET throws vacuous error

  #51779 closed Sep 1, 2021

• Build error: Undeclared inclusions of various numpy header files

  #26157 closed Sep 1, 2021

• make_early_stopping_hook will always return _MultiWorkerEarlyStoppingHook, no matter what strategy is used

  #51525 closed Sep 1, 2021

• Internal error: Tried to take gradients (or similar) of a variable without handle data

  #51535 closed Sep 1, 2021

• When use keras + TensorBoard with parameter server strategy, error on worker is raised and core is dumped.

  #51768 closed Sep 1, 2021

• tf.summary.create_file_writer aborts

  #46909 closed Aug 31, 2021

• TFLM: add run-time check that statically allocated buffers are sufficiently large

  #38156 closed Aug 31, 2021

• implicit conversion from 'int' to 'float' changes value from INT_MAX to INT_MAX+1

  #38106 closed Aug 31, 2021

• google.protobuf.message.DecodeError: Error parsing message when using tf.function

  #51589 closed Aug 31, 2021

• Brand new Pipenv environment can't install tensorflow?

  #50370 closed Aug 31, 2021

• same issue with CUDA=10.0 CUDNN=7.4 Tensorflow=1.14.0

  #51482 closed Aug 31, 2021

• Providing a Prebuilt Binary for TFLite Benchmark

  #26499 closed Aug 31, 2021

• Separable Conv2D allows dilation_rate!=1 and strides !=1

  #51710 closed Aug 31, 2021

• ABSL_HAVE_ADDRESS_SANITIZER use in Tensorflow headers breaks build for MSVC 16.11 with /fsanitize=address

  #51685 closed Aug 31, 2021

• ValueError: Attempt to convert a value (None) with an unsupported type (<class 'NoneType'>) to a Tensor.

  #51376 closed Aug 31, 2021

• does lstm only running on CPU? Can it delegated to Apple Neural Engine(ANE) by CoreML delegate ?

  #51674 closed Aug 31, 2021

• NameError:

  #51745 closed Aug 31, 2021

• How to collect code coverage of C++ code when running Python tests?

  #51091 closed Aug 31, 2021

• Question about custom convolution layer

  #51519 closed Aug 31, 2021

• Inconsistency in Gradients Calculation of max_pooling and reduce_max

  #51682 closed Aug 31, 2021

• Tensorflow profiler crashes when using string categorical layer

  #41244 closed Aug 31, 2021

• SavedModelBundle.getSignatures() does not return outputs in correct order.

  #35150 closed Aug 31, 2021

• Support on TPU for tf.contrib.framework.sort

  #19517 closed Aug 31, 2021

• How to modify ckpt value

  #51713 closed Aug 31, 2021

• tf.ragged.range and tf.range crash (abort) when `start` is large

  #46899 closed Aug 30, 2021

• tf.keras.backend.arange crash (abort) when start is large

  #46889 closed Aug 30, 2021

• Dilated and causal convolutions on microcontrollers

  #48567 closed Aug 30, 2021

• Incorrect predictions while exporting keras model to android

  #26068 closed Aug 30, 2021

• Add usage examples for tf.audio APIs

  #35079 closed Aug 30, 2021

• ERROR: /home/user/.cache/bazel/_bazel_user/b4774fbdb8542988b4e302c9e073f145/external/com_google_absl/absl/types/BUILD.bazel:190:1: C++ compilation of rule '@com_google_absl//absl/types:bad_variant_access' failed (Exit 1) on benchmark_model tool build

  #26882 closed Aug 30, 2021

• Build issue of "cannot find symbol class Fill where T is a type-variable: T extends Object declared in class Zeros" still persists, when building Bazel.

  #25239 closed Aug 30, 2021

• bazel build aot armeabi-v7a so failed

  #25275 closed Aug 30, 2021

• map_fn can't process sparse tensor in graph mode

  #50554 closed Aug 30, 2021

• In TFnighty TFLite Interpreter missing setUseNNAPI "wrapper" method to call Interpreter.Options.setUseNNAPI

  #44023 closed Aug 30, 2021

• saved_model bug

  #51448 closed Aug 30, 2021

• Internal Error with TF_GPU_ALLOCATOR=cuda_malloc_async

  #50669 closed Aug 30, 2021

• tf.identity cannot support a tuple value

  #51440 closed Aug 30, 2021

• Apple M1 - TF Master - Build Error: Symbol not found

  #51494 closed Aug 30, 2021

• Reshape output tensor in Tensorflow lite.

  #51740 closed Aug 30, 2021

• Debundled protobuf broken

  #37861 closed Aug 30, 2021

• tf.math.confusion_matrix InvalidArgumentError about incompatible shapes

  #51077 closed Aug 30, 2021

• import error

  #51425 closed Aug 30, 2021

• tensorflow-aarch64 with keras

  #15416 closed Aug 30, 2021

• CMake build fails v1.13.1 fatal error: tensorflow/stream_executor/dnn.pb.h: No such file or directory #include "tensorflow/stream_executor/dnn.pb.h"

  #29042 closed Aug 30, 2021

• CAN'T BUILD TENSORFLOW

  #23142 closed Aug 30, 2021

• numpy.dtype size changed, may indicate binary incompatibility

  #25934 closed Aug 30, 2021

• How to use StreamingFilesDataset

  #51700 closed Aug 29, 2021

• NotFoundError: No algorithm worked! when using Conv2D

  #43174 closed Aug 29, 2021

• Unable to generate train.record, but for text.record it works fine

  #51729 closed Aug 29, 2021

• [tf2.6] Model saving error for a customized model and loss function

  #51665 closed Aug 29, 2021

• TF RNN subclass model called twice although sequence length is one

  #51731 closed Aug 28, 2021

• tf.nn.softmax_cross_entropy_with_logits and tf.keras.backend.categorical_crossentropy crash(abort)

  #46908 closed Aug 28, 2021

• Getting non-deterministic results on TF1.9

  #51492 closed Aug 28, 2021

• dlopen/dlclose a so file which contain tensorflow static library cause memory leak

  #51441 closed Aug 28, 2021

• micro: port op SLICE from lite

  #51669 closed Aug 27, 2021

• ESP32 LyraT: Audio data/I2S issue

  #45185 closed Aug 27, 2021

• NameError: name 'trainer' is not defined

  #51721 closed Aug 27, 2021

• How to install Tensorflow on AIX7.2 server

  #46759 closed Aug 27, 2021

• Symmetric convolved with symmetric returns non-symmetric

  #51718 closed Aug 27, 2021

• the moving_mean and moving_variance in batchnormalization going to very big after 20 epoch

  #44371 closed Aug 27, 2021

• tf.stop_gradient sometimes does not work as expected when called from a @tf.function

  #51473 closed Aug 27, 2021

• 07502269190

  #51707 closed Aug 27, 2021

• How to convert savedmodel to frozen model(.pb)

  #51486 closed Aug 27, 2021

• TFLite Conversion Error: Element Shape Required to be 1D

  #50942 closed Aug 27, 2021

• Adding Object-Detection / Segmentation applications to tf.keras.applications

  #50120 closed Aug 27, 2021

• Fails building example projects for ESP micro

  #48570 closed Aug 27, 2021

• Fail to converted Tensorflow models to TensorFlow Lite

  #50660 closed Aug 27, 2021

• [tf2.6] Loaded subclassed-Model fails to be retrained

  #51712 closed Aug 27, 2021

• TF2.6 How to avoid UnimplementedError error for non-deterministic ops?

  #51466 closed Aug 27, 2021

• high vram usage when setting architecture on rtx3090?

  #51458 closed Aug 27, 2021

• Error while trying to convert TF-Hub MRCNN model to tflite

  #51708 closed Aug 27, 2021

• ERROR: Could not find a version that satisfies the requirement tensorflow==1.15 (from versions: none)

  #51356 closed Aug 27, 2021

• save_model and export categorical_column_with_vocabulary_file

  #51347 closed Aug 27, 2021

• Training Mask R Cnn

  #51694 closed Aug 27, 2021

• Loading checkpoint by skipping the head layer

  #51420 closed Aug 27, 2021

• tensorflow distribute trainning error

  #51427 closed Aug 27, 2021

• ImportError: cannot import name 'build_info' when using command import tensorflow as tf

  #20692 closed Aug 27, 2021

• Conv2DTranspose in Hexagon delegate crash

  #51093 closed Aug 27, 2021

• Mask R-CNN on TFLite (Input and Output tensors Help)

  #51679 closed Aug 27, 2021

• NameError:

  #51690 closed Aug 27, 2021

• `tf.GradientTape.batch_jacobian` fails when slicing tensor with `int64` index

  #51528 closed Aug 26, 2021

• TypeError: EndVector() takes 1 positional argument but 2 were given

  #51590 closed Aug 26, 2021

• libhdfs3 instead of libhdfs

  #16919 closed Aug 26, 2021

• DeprecationWarning: the imp module is deprecated in favour of importlib

  #51631 closed Aug 26, 2021

• DeprecationWarning: the imp module is deprecated

  #31412 closed Aug 26, 2021

• Conversion fails at model loading

  #51430 closed Aug 26, 2021

• AttributeError: 'Operation' object has no attribute '_c_op'

  #51446 closed Aug 26, 2021

• Output size same as input size only if strides = 1

  #48137 closed Aug 26, 2021

• Activity Regularizer not working with quantization aware training (QAT)

  #51680 closed Aug 26, 2021

50 Issues opened by 46 people

• how to use SparseTensor with ctc loss where the label's length varies within one batch

  #51802 opened Sep 2, 2021

• When loading a saved model with multiple outputs, outputs order is not deterministic

  #51801 opened Sep 2, 2021

• Android GPU delegate gives same output from intermediate layers output

  #51800 opened Sep 2, 2021

• Compiling TF 2.6 in debug mode on WIndows env.

  #51799 opened Sep 2, 2021

• tensorflow2 memory leak when using tf.function

  #51798 opened Sep 2, 2021

• Interpreter invocation error after input tensor resizing

  #51797 opened Sep 2, 2021

• Attributes section in Keras Model (documentation) is not generated properly

  #51796 opened Sep 2, 2021

• tf.GradientTape not working properly.

  #51795 opened Sep 2, 2021

• Unable to restore the model with shared ops using from_config()

  #51793 opened Sep 2, 2021

• weighted_moments produces NaNs when weights are all zeros

  #51792 opened Sep 2, 2021

• Transfer learning ResNet mode performance drop in recent versions

  #51790 opened Sep 2, 2021

• tf.nn.max_pool_with_argmax to support multiple dimensions (ND)

  #51787 opened Sep 1, 2021

• Type Error: __init__() got an unexpected keyword argument 'epochs'

  #51786 opened Sep 1, 2021

• TypeError: 'NoneType' object is not callable

  #51783 opened Sep 1, 2021

• 2021-09-01 16:39:11.050592: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operatiWARNING:tensorflow:AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x000001A3554913A0> and will run it as-is. Please report this to the TensorFlow team. To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert

  #51780 opened Sep 1, 2021

• When I check the TensorBoard for a Worker of ParameterServer traning, I can't see any send/recv or push/pull op to the PS? Is this normal?

  #51778 opened Sep 1, 2021

• Building from source failed on Python 3.10

  #51776 opened Sep 1, 2021

• Sizing problem with BatchDataset in Tensorflow 2.3.1

  #51774 opened Aug 31, 2021

• tensorboard no runs found for PyTorch 1.9 in VSCode in CentOS 7

  #51773 opened Aug 31, 2021

• AttributeError: 'NoneType' object has no attribute 'label_map'

  #51772 opened Aug 31, 2021

• tf.vectorized_map fails on ragged tensors

  #51771 opened Aug 31, 2021

• Saving a NSL model

  #51769 opened Aug 31, 2021

• Conflict happened when using both tensorboard and Nsight System

  #51767 opened Aug 31, 2021

• Issue about using Nsight System on Tensorflow2.6.0

  #51766 opened Aug 31, 2021

• How is TensorFlow compatible with Python2 and Python3 ？

  #51763 opened Aug 31, 2021

• tf.distribute.MirroredStrategy does not work via srun (SLURM) causes NCCL error

  #51760 opened Aug 31, 2021

• mips toolchain compile tflite2.6 error

  #51757 opened Aug 31, 2021

• [tflite conversion] Invalid file dropped for invalid shape without error

  #51756 opened Aug 31, 2021

• Import tensor error

  #51755 opened Aug 31, 2021

• Tensorflow Lite C API + GPU delegate crash on exit

  #51751 opened Aug 30, 2021

• [AArch64] error: inlining failed in call to 'always_inline' 'uint8x16_t vaeseq_u8(uint8x16_t, uint8x16_t)': target specific option mismatch

  #51750 opened Aug 30, 2021

• Error loading saved model

  #51746 opened Aug 30, 2021

• NameError:

  #51744 opened Aug 30, 2021

• typing-extensions package is conflicted

  #51743 opened Aug 30, 2021

• TransformGraphWithStringInputs is available in tf2.0 however I am planning to jump to tf2.2 or tf2.3 but I couldn't find "TransformGraphWithStringInputs" in these 2 versions. Any suggestions ?

  #51742 opened Aug 30, 2021

• CAST: Not supported Cast case.

  #51741 opened Aug 30, 2021

• Importing layers from keras instead of tensorflow.keras causes erroneous model.summary() behavior

  #51738 opened Aug 29, 2021

• Failed to build minimal example of Tensorflow Lite on Qualcomm robotic kit rb5

  #51737 opened Aug 29, 2021

• Image transformations require SciPy. Install SciPy.

  #51736 opened Aug 29, 2021

• Failed to get compute capability major for device: UNKNOWN ERROR (1); 0

  #51730 opened Aug 28, 2021

• More freedom in Dtype of indices of sparse.SparseTensors

  #51728 opened Aug 28, 2021

• Unable to save custom model using save_model, even though documentation states it is possible

  #51727 opened Aug 28, 2021

• Several links to the github code are broken in https://www.tensorflow.org/

  #51726 opened Aug 28, 2021

• How can I increase throughput for multi sessions/models on single GPU of online inference？

  #51725 opened Aug 28, 2021

• GPU not found when using TF 2.5.0 in Google Colab

  #51724 opened Aug 28, 2021

• Build stuck at Linking _pywrap_tensorflow_internal.so step

  #51723 opened Aug 28, 2021

• TensorRT model outputs/performance keeps changing across conversions

  #51719 opened Aug 27, 2021

• Grappler graph optimizations may add nondeterministic behavior

  #51716 opened Aug 27, 2021

• '@local_config_cc//:toolchain' does not contain a toolchain for cpu 'arm64-v8a'

  #51709 opened Aug 27, 2021

• Tensorflow not installing correctly

  #51704 opened Aug 26, 2021

354 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [TF:TRT] Enable dynamic shape for "slice" and "strided slice" operations.

  #51475 commented on Sep 1, 2021 • 47 new comments

• Compatibility with NumPy 1.20+

  #48935 commented on Sep 2, 2021 • 20 new comments

• Implementation of a Copy Removal Pass.

  #51630 commented on Sep 1, 2021 • 14 new comments

• [PluggableDevice] Add pluggable profiler C Interface support

  #51647 commented on Sep 2, 2021 • 12 new comments

• "ZeroDivisionError: integer division or modulo by zero" while backpropagating

  #51653 commented on Sep 1, 2021 • 6 new comments

• "group_by_reducer" scales badly with more images

  #51491 commented on Sep 1, 2021 • 5 new comments

• Document XLA core functionality

  #51634 commented on Sep 1, 2021 • 5 new comments

• [ROCm]: Updating Matrix Solve Op to use rocSolver

  #51423 commented on Sep 1, 2021 • 5 new comments

• Error while converting mrcnn to tf dialect

  #51691 commented on Aug 31, 2021 • 4 new comments

• AttributeError: module 'keras.engine' has no attribute 'Layer'

  #51594 commented on Aug 31, 2021 • 4 new comments

• OSError: [Errno 9] Bad file descriptor raised on program exit

  #50487 commented on Aug 31, 2021 • 4 new comments

• TFLite for conv1D and DepthwiseConv1D

  #51692 commented on Sep 1, 2021 • 4 new comments

• FIx random gamma and space to depth tests.

  #50813 commented on Sep 2, 2021 • 4 new comments

• Remove AWS files as s3 support is now in modular file system

  #51032 commented on Sep 1, 2021 • 4 new comments

• tf.profiler.experimental.Trace only works with name is "TraceContext"

  #50440 commented on Aug 26, 2021 • 3 new comments

• WARNING:absl:Found untraced functions such as lstm_cell_2_layer_call_fn, lstm_cell_2_layer_call_and_return_conditional_losses, lstm_cell_2_layer_call_fn, lstm_cell_2_layer_call_and_return_conditional_losses, lstm_cell_2_layer_call_and_return_conditional_losses while saving (showing 5 of 5). These functions will not be directly callable after loading.

  #47554 commented on Aug 30, 2021 • 3 new comments

• Problem with mean OpenGL since 2.5 version

  #51429 commented on Aug 30, 2021 • 3 new comments

• Failed when upgrading grpc to 1.36.0

  #50989 commented on Aug 31, 2021 • 3 new comments

• Missing _keras_logits from model.predict(x) leads to difference in loss evaluation

  #51574 commented on Sep 1, 2021 • 3 new comments

• TfLite Generating the project issue

  #51701 commented on Sep 1, 2021 • 3 new comments

• tensorflow-gpu Docker Image - Python TensorRT Import Fails

  #51619 commented on Sep 1, 2021 • 3 new comments

• Bogus cache keys in Network._output_shape_cache can cause wrong output shapes (and gradual but unbounded memory growth)

  #34827 commented on Sep 2, 2021 • 3 new comments

• [TF:TRT] Reorganize testing utilities.

  #51235 commented on Sep 2, 2021 • 3 new comments

• [ROCM] enable some multigpu tests

  #51470 commented on Sep 1, 2021 • 3 new comments

• TFLite Failed to build gpu delegate on Linux

  #38336 commented on Aug 27, 2021 • 2 new comments

• Custom loss function is not working

  #43650 commented on Aug 27, 2021 • 2 new comments

• tf.keras.models.save_model not saving the probabilistic_model

  #50774 commented on Aug 28, 2021 • 2 new comments

• model.layers returns empty list

  #51645 commented on Aug 28, 2021 • 2 new comments

• TFLite quant model run on the android phone have no speed up than float model

  #40914 commented on Aug 30, 2021 • 2 new comments

• Adding Tab press for showing available methods and autocompletion for objects

  #39142 commented on Aug 30, 2021 • 2 new comments

• XNNPack delegate undefined reference

  #42147 commented on Aug 31, 2021 • 2 new comments

• tf.math.xdivy decorated with @tf.function returns 0 when the input is large

  #51643 commented on Aug 31, 2021 • 2 new comments

• How to remove unused operators from tensorflowlite_c.so to reduce it's size?

  #44090 commented on Aug 31, 2021 • 2 new comments

• Fail to find the dnn implementation while using recurrent layers

  #45248 commented on Sep 1, 2021 • 2 new comments

• [C++][FATAL]Calling FreezeSavedModel in cc/tools/saved_model.h causes protobuf::FatalException

  #45943 commented on Sep 1, 2021 • 2 new comments

• An easy way to update tensors that complies with numpy syntax

  #51582 commented on Sep 1, 2021 • 2 new comments

• script exits after starting first epoch

  #51483 commented on Sep 1, 2021 • 2 new comments

• py_function doesn't work with functions with no return values and inputs as keras.Input

  #51677 commented on Sep 1, 2021 • 2 new comments

• Inference is slower on Densenet121 when using XLA

  #51179 commented on Sep 2, 2021 • 2 new comments

• Remove GCC_HOST_COMPILER_PREFIX as it may be out of sync with GCC_HOST_COMPILER_PATH

  #39263 commented on Sep 2, 2021 • 2 new comments

• Java Module support

  #39378 commented on Sep 2, 2021 • 2 new comments

• InvalidArgumentError with `Conv1D` which has the same `dilation_rate` as the batch size, using `tf.keras.Model.fit`, on TPU

  #43690 commented on Sep 2, 2021 • 2 new comments

• model.evaluate() does not yield the same accuracy as computing it manually using a for-loop

  #50881 commented on Sep 2, 2021 • 2 new comments

• tf.image.resize/resize_with_crop_or_pad/pad_to_bounding_box/extract_glimpse crash(abort)

  #46890 commented on Sep 2, 2021 • 2 new comments

• Addition of TanhExp activation function - issue #45929

  #47566 commented on Aug 30, 2021 • 2 new comments

• [WIP] Initial support for sparse labels on confusion-matrix metrics

  #48122 commented on Aug 30, 2021 • 2 new comments

• Added Lite option to EfficientNets.

  #48400 commented on Aug 28, 2021 • 2 new comments

• Allow `keras.utils.Sequence` sub-classes to use sparse/ragged tensors

  #49872 commented on Aug 29, 2021 • 2 new comments

• [TFLite] Update MultiplyByQuantizedMultiplier to use single-rounding instead of double-rounding

  #50290 commented on Aug 31, 2021 • 2 new comments

• [XLA] Make replay_computation accept input HLO file with multiple module.

  #51558 commented on Aug 31, 2021 • 2 new comments

• Installing tensorflow 2 on raspberry pi 3 model B

  #47142 commented on Aug 26, 2021 • 1 new comment

• ./configure [--help|-h] does not work

  #16518 commented on Aug 26, 2021 • 1 new comment

• Request for better installation instructions for TF on Ampere GPUs!

  #47066 commented on Aug 26, 2021 • 1 new comment

• please update the release with lib file assets

  #47065 commented on Aug 26, 2021 • 1 new comment

• TF 2.4.1 with python 3.8.7 (apple silicon with rosetta) terminated by SIGILL on import

  #47041 commented on Aug 26, 2021 • 1 new comment

• AttributeError: type object 'IteratorBase' has no attribute 'from_structure'

  #50871 commented on Aug 26, 2021 • 1 new comment

• How can I make multiple sessions share a graph? Not found: Container localhost does not exist

  #51568 commented on Aug 26, 2021 • 1 new comment

• undeclared identifier 'ABSL_FALLTHROUGH_INTENDED'

  #39590 commented on Aug 26, 2021 • 1 new comment

• CMP0002 error when building TensorFlow cc unit tests

  #17267 commented on Aug 26, 2021 • 1 new comment

• no such package '@androidsdk//com.android.support'

  #39648 commented on Aug 26, 2021 • 1 new comment

• No registered 'EagerPyFunc' OpKernel for XLA_GPU_JIT devices compatible with node {{node EagerPyFunc}}){{node EagerPyFunc}}

  #51493 commented on Aug 26, 2021 • 1 new comment

• Cannot split windowed data when batching is used in tf.data.Dataset

  #51084 commented on Aug 26, 2021 • 1 new comment

• Could not find device for node (`reduce_max` of complex)

  #51502 commented on Aug 26, 2021 • 1 new comment

• How the parameters of decay_rate & decay_steps are taken into consideration while computing InverseTimeDecay()

  #46071 commented on Aug 27, 2021 • 1 new comment

• tensorflow.lib missing some symbols on linking in a application

  #39407 commented on Aug 27, 2021 • 1 new comment

• "Node 'Const' is not unique" Using C++ api on windows to freeze graph

  #29907 commented on Aug 27, 2021 • 1 new comment

• Hwloc mirror - download issue

  #30483 commented on Aug 27, 2021 • 1 new comment

• Build failure in tensorflow/tensorflow:devel docker container on MacOS Catalina 10.15.4

  #39328 commented on Aug 27, 2021 • 1 new comment

• tensortflow-lite installation and packaging

  #48575 commented on Aug 27, 2021 • 1 new comment

• layer.output raises AttributeError because inbound nodes lost after call to activation function

  #34834 commented on Aug 27, 2021 • 1 new comment

• RuntimeError: tensorflow/lite/kernels/conv.cc:349 input->dims->data[3] != filter->dims->data[3] (64 != 2)Node number 12 (CONV_2D) failed to prepare.

  #51570 commented on Aug 27, 2021 • 1 new comment

• precision and recall values kept unchanged for some training steps.

  #19138 commented on Aug 27, 2021 • 1 new comment

• “Cloud TPU” console spam on every TensorFlow import

  #50423 commented on Aug 27, 2021 • 1 new comment

• about the Dropout

  #51593 commented on Aug 27, 2021 • 1 new comment

• Thread hang when setting inter_op_parallelism_threads=1

  #38969 commented on Aug 27, 2021 • 1 new comment

• C-API input tensor type of all run functions

  #39026 commented on Aug 27, 2021 • 1 new comment

• CUDA compiling include file path failure inside tensorflow python installation directory in Windows 10

  #39829 commented on Aug 27, 2021 • 1 new comment

• error when installing for go from source.

  #42632 commented on Aug 27, 2021 • 1 new comment

• CUDA 11.0 Build with tensorflow master on Win 10 : Link error

  #42620 commented on Aug 27, 2021 • 1 new comment

• [TF Lite] How to use XNNPACK delegate on Windows?

  #47007 commented on Aug 27, 2021 • 1 new comment

• version 2.3.0 build error: user_ops_gen_cc: undefined symbol

  #42221 commented on Aug 27, 2021 • 1 new comment

• Build with clang support is failing

  #46976 commented on Aug 27, 2021 • 1 new comment

• C++ compilation of rule failed - Error tensorflow building from source

  #46786 commented on Aug 27, 2021 • 1 new comment

• tensorflow 2.3.0: target 'grpc++_public_hdrs' not declared in package when build from source

  #41953 commented on Aug 27, 2021 • 1 new comment

• Some **function** symbols are exported as DATA from prebuilt Windows CPU tensorflow.dll

  #46730 commented on Aug 27, 2021 • 1 new comment

• tf.data.experimental.ignore_errors

  #27022 commented on Aug 27, 2021 • 1 new comment

• Issue in building Tensorflow library on Windows machine

  #23342 commented on Aug 27, 2021 • 1 new comment

• macOS Target //tensorflow/tools/pip_package:build_pip_package failed to build in debug mode

  #46566 commented on Aug 27, 2021 • 1 new comment

• Unable to selectively build TensorFlow Lite with Docker

  #46514 commented on Aug 27, 2021 • 1 new comment

• Can't easily write custom C++ ops which update resource variables

  #27899 commented on Aug 27, 2021 • 1 new comment

• Error: "Failed to connect to all addresses", TPU on Colab

  #51579 commented on Aug 27, 2021 • 1 new comment

• ios demo app version error

  #51543 commented on Aug 27, 2021 • 1 new comment

• How to wrap a CuPy function inside a function decorated by tf.function

  #51642 commented on Aug 27, 2021 • 1 new comment

• mixed_float16 + example code = horrible performance on a GeForce GTX 1660 Ti

  #44191 commented on Aug 27, 2021 • 1 new comment

• TF25 inconsistency between eager and graph mode

  #50728 commented on Aug 27, 2021 • 1 new comment

• How to build tensorflow when making small changes?

  #46972 commented on Aug 27, 2021 • 1 new comment

• Inconsistent prediction by model using Attention layer

  #40911 commented on Aug 28, 2021 • 1 new comment

• tf.math.segment_max/min/mean/sun/prod crashes(aborts) when segment_ids is large

  #46888 commented on Aug 28, 2021 • 1 new comment

• tf.signal.inverse_stft does not reconstructs the original signal

  #36616 commented on Aug 29, 2021 • 1 new comment

• Initializing TFLite model causes crash on iOS

  #51547 commented on Aug 30, 2021 • 1 new comment

• Cannot load SavedModel when model trained with tfa.optimizers.NovoGrad: ValueError: Shapes (3, 8) and () are incompatible

  #45557 commented on Aug 30, 2021 • 1 new comment

• Not found: No registered '_FusedMatMul' OpKernel for CPU devices compatible with node

  #27671 commented on Aug 30, 2021 • 1 new comment

• failed to embed python code that import tensorflow in c++

  #20561 commented on Aug 30, 2021 • 1 new comment

• TFLite GPU works slower than CPU

  #30169 commented on Aug 30, 2021 • 1 new comment

• Tensorflow 1.8 graph destructor hanging indefinitely

  #20600 commented on Aug 30, 2021 • 1 new comment

• use SPICE tf-lite model in Swift but get error "Provided data count 64000 must match the required count 4"

  #51621 commented on Aug 30, 2021 • 1 new comment

• tf.experimental.dlpack.from_dlpack() does not support negative strides

  #51598 commented on Aug 30, 2021 • 1 new comment

• error

  #51620 commented on Aug 30, 2021 • 1 new comment

• Determinism OP bug on TF 2.6.0

  #51596 commented on Aug 30, 2021 • 1 new comment

• A name conflict bug caused by LoweringFunctionalOps pass

  #50393 commented on Aug 30, 2021 • 1 new comment

• Feature Request: GPUOptions for Go binding

  #22926 commented on Aug 30, 2021 • 1 new comment

• Debugger V2 not working. Invalid argument: DebugNumericSummaryV2Op requires tensor_id to be less than or equal to (2^53). Given tensor_id:26

  #43608 commented on Aug 30, 2021 • 1 new comment

• calback pyfunc_18 is not found

  #51623 commented on Aug 30, 2021 • 1 new comment

• How can I clear GPU memory in tensorflow 2?

  #36465 commented on Aug 30, 2021 • 1 new comment

• Introduce ability to clear GPU memory in Tensorflow 2

  #48545 commented on Aug 30, 2021 • 1 new comment

• Unable to find implementation of Select TensorFlow operators

  #30799 commented on Aug 30, 2021 • 1 new comment

• TanhExp activation function

  #45929 commented on Aug 30, 2021 • 1 new comment

• TF Lite GPU delegate - Selection of GPU device

  #44332 commented on Aug 30, 2021 • 1 new comment

• Densenet (Floating point models) reported accuracy is much lower than the original paper

  #44032 commented on Aug 30, 2021 • 1 new comment

• What is the minimum subset of cuda toolkit to install for tensorflow gpu to work?

  #38824 commented on Aug 30, 2021 • 1 new comment

• How to load a tflite model directly from a link or aws s3?

  #42323 commented on Aug 30, 2021 • 1 new comment

• Generate C++ code from TensorFlow Lite metadata

  #38740 commented on Aug 30, 2021 • 1 new comment

• C API for TensorFlow Lite for Microcontrollers (micro)?

  #38536 commented on Aug 30, 2021 • 1 new comment

• Using libtensorflow-core.a to enable training on Android

  #31829 commented on Aug 30, 2021 • 1 new comment

• Multiple duplicate tflite android example projects

  #30087 commented on Aug 30, 2021 • 1 new comment

• libtensorflowlite.so usage is not documented

  #28105 commented on Aug 30, 2021 • 1 new comment

• No way to use string type with Lite C APIs.

  #27223 commented on Aug 30, 2021 • 1 new comment

• Tensorflow-lite segmentation model does not work with NNAPI on Android 9.0

  #27031 commented on Aug 30, 2021 • 1 new comment

• lite: non-standard variable-length arrays

  #25215 commented on Aug 30, 2021 • 1 new comment

• TFLiteConverter unable to convert object detection model from export_tflite_ssd_graph.py

  #24910 commented on Aug 30, 2021 • 1 new comment

• Self-contained source code package of tensorflow

  #11149 commented on Aug 30, 2021 • 1 new comment

• Build TensorFlowLite C API with GPU delegate with CMake

  #50940 commented on Aug 30, 2021 • 1 new comment

• [building tensorflow with bazel ] Error:C++ compilation of rule '@boringssl//:crypto' failed.

  #12579 commented on Aug 30, 2021 • 1 new comment

• TF2.0 - Cannot convert EagerTensor to datatype and casting

  #34848 commented on Aug 30, 2021 • 1 new comment

• type dependent output

  #51349 commented on Aug 30, 2021 • 1 new comment

• There should be No expansion ReLU on MobileNet v2 bottleneck

  #50835 commented on Aug 30, 2021 • 1 new comment

• [Tflite] Unable to convert model to tflite with uneven input dimension

  #26435 commented on Aug 31, 2021 • 1 new comment

• TensorFlow Lite Android nightly runs inference x2 faster than latest version 1.13.1

  #26385 commented on Aug 31, 2021 • 1 new comment

• tf.image.extract_glimpse crashes with negative input

  #51618 commented on Aug 31, 2021 • 1 new comment

• RTX30 series: unsupported compute compability for CUDA 9.0

  #51088 commented on Aug 31, 2021 • 1 new comment

• Question about tf.compat.v1.metrics.auc

  #51407 commented on Aug 31, 2021 • 1 new comment

• Improve testing infrastructure to prevent breaking open source builds

  #26776 commented on Aug 31, 2021 • 1 new comment

• Named dictionary inputs and outputs for tf.keras.Model

  #34114 commented on Aug 31, 2021 • 1 new comment

• Problems setting up Tensorflow on M1 Mac

  #51403 commented on Aug 31, 2021 • 1 new comment

• RuntimeError: Given shapes, [1, 23], ] and [1, 10], are not broadcastable.Node number 1392 (SELECT_V2) failed to prepare.

  #51602 commented on Aug 31, 2021 • 1 new comment

• Graph Transform Tool unable to build in TF source r1.5?

  #16205 commented on Aug 31, 2021 • 1 new comment

• Build fails on latest master with clang and CUDA

  #32323 commented on Aug 31, 2021 • 1 new comment

• Please modify cuda-bin in cuda_configure.bzl to copy just needed files instead of directory

  #32244 commented on Aug 31, 2021 • 1 new comment

• tensorflow 1.14 build failed

  #31434 commented on Aug 31, 2021 • 1 new comment

• [TF 2.0 API Docs] tf.dynamic_partition

  #29854 commented on Aug 31, 2021 • 1 new comment

• does pip wheels support hdfs on default?

  #30981 commented on Aug 31, 2021 • 1 new comment

• Segfault in custom op with TensorFlow 1.14, Python 3.6 and GCC 5.

  #30956 commented on Aug 31, 2021 • 1 new comment

• Tensorflow 2.0 - Build From Source not creating Python Wrappers such as _pywrap_tensorflow_internal.lib

  #30839 commented on Aug 31, 2021 • 1 new comment

• tf.keras.layers.ConvLSTM2D crashes when kernel_size contains zero

  #51625 commented on Aug 31, 2021 • 1 new comment

• Segmentation fault when using NCCL in MultiWorkerMirroredStrategy

  #50926 commented on Aug 31, 2021 • 1 new comment

• How to limit the cpu core number in Tensorflow 2.0

  #34415 commented on Aug 31, 2021 • 1 new comment

• Arduino tensorflow

  #43020 commented on Aug 31, 2021 • 1 new comment

• tfio.audio.AudioIOTensor doesn't prepare data in time in loop.

  #43604 commented on Aug 31, 2021 • 1 new comment

• TensorRT Converter *Core Dumped*

  #50417 commented on Aug 31, 2021 • 1 new comment

• tf.execute_volume_patches misses one dimension when used with placeholders and None

  #34898 commented on Aug 31, 2021 • 1 new comment

• whether tflite can not support the two operator ( MaxPool2D and QUANTIZE) in TensorFlow 2.0.0 when running the mode in embed freeRTOS system?

  #34608 commented on Aug 31, 2021 • 1 new comment

• tf.keras computes incorrect loss values with Masking

  #34491 commented on Aug 31, 2021 • 1 new comment

• Master build failed with user local specific GCC

  #38300 commented on Aug 31, 2021 • 1 new comment

• Build Tensorflow with triSYCL?

  #38737 commented on Aug 31, 2021 • 1 new comment

• 2.1.0 build error due to pybind11

  #37695 commented on Aug 31, 2021 • 1 new comment

• Unable to successfully build a devel-cpu.Dockerfile for arm32v7/arm64v8

  #37985 commented on Aug 31, 2021 • 1 new comment

• /bin/sh: 1: python: not found error when generate projects

  #38070 commented on Aug 31, 2021 • 1 new comment

• Does a Docker container need to use the same exact libcuda version as the host's driver?

  #51641 commented on Aug 31, 2021 • 1 new comment

• bug in rsqrt

  #51043 commented on Aug 31, 2021 • 1 new comment

• Support resizing images on TPU using `tf.image.resize` when `size` is not a compile-time constant

  #51693 commented on Aug 31, 2021 • 1 new comment

• KeyError: "Failed to add concrete function b'__inference_train_14141' to object based saved model as it captures tf.Tensor(<unprintable>, shape=(), dtype=resource) tensor

  #51587 commented on Aug 31, 2021 • 1 new comment

• Different results for add_loss() within TimeDistributed for eager vs. compiled execution of keras Layer

  #46244 commented on Sep 1, 2021 • 1 new comment

• Android Studio CPU Profiler

  #51389 commented on Sep 1, 2021 • 1 new comment

• Can't set empty layer name

  #47131 commented on Sep 1, 2021 • 1 new comment

• Failed to load the native TensorFlow runtime.

  #51638 commented on Sep 1, 2021 • 1 new comment

• What is the format setting used in the repository?

  #51649 commented on Sep 1, 2021 • 1 new comment

• [TF 2.0] Custom objects

  #33387 commented on Sep 1, 2021 • 1 new comment

• My libnnappi_delegate.so is debug or release

  #51652 commented on Sep 1, 2021 • 1 new comment

• InvalidArgumentError in Model Maker Object Detection Tutorial

  #51576 commented on Sep 1, 2021 • 1 new comment

• Apple M1 - tf.sort only sorts up to 16 values for float32

  #51499 commented on Sep 1, 2021 • 1 new comment

• Compilation Problems

  #45405 commented on Sep 1, 2021 • 1 new comment

• tflite 2.4 build benchmark_model error on ubuntu 16.04 (cmake)

  #45290 commented on Sep 1, 2021 • 1 new comment

• Check failed: dim_size >= 1 (0 vs. 1)

  #34257 commented on Sep 1, 2021 • 1 new comment

• need some Ops in RNN

  #33239 commented on Sep 1, 2021 • 1 new comment

• GPU Profiling: MemoryProfile do not contain memory events.

  #48146 commented on Sep 1, 2021 • 1 new comment

• Loading a saved multi-input model

  #39013 commented on Sep 1, 2021 • 1 new comment

• TensorflowLite gives wrong results when use GPU delegate

  #35378 commented on Sep 1, 2021 • 1 new comment

• Object detection producing incorrect results on mobile (ios)

  #35205 commented on Sep 1, 2021 • 1 new comment

• toco_from_protos: not found - breaking

  #34698 commented on Sep 1, 2021 • 1 new comment

• Inconsistent loading of saved model

  #35002 commented on Sep 1, 2021 • 1 new comment

• Tensorflow 2.3.0 Bazel 3.1.0 compile ERROR

  #45205 commented on Sep 1, 2021 • 1 new comment

• tensorflow-lite-gpu.aar built from source is much slower than downloaded prebuilt library

  #35170 commented on Sep 1, 2021 • 1 new comment

• Windows build failed - Internal compiler error Visual Studio 2017 - FAILED: Build did NOT complete successfully

  #35180 commented on Sep 1, 2021 • 1 new comment

• -D_GLIBCXX_USE_CXX11_ABI=1 increases a lot RAM usage

  #35160 commented on Sep 1, 2021 • 1 new comment

• Build error due to missing dependency declarations in double-conversion

  #35014 commented on Sep 1, 2021 • 1 new comment

• freeze_graph returns with error

  #21670 commented on Sep 1, 2021 • 1 new comment

• Build failure when building Tensorflow MLIR

  #51601 commented on Sep 1, 2021 • 1 new comment

• terminate called after throwing an instance of 'std::out_of_range' error when call made to tf.contrib.tensor_forest.random_forest.TensorForestEstimator.predict()

  #11058 commented on Sep 1, 2021 • 1 new comment

• Infinity mask breaks gradient

  #11756 commented on Sep 1, 2021 • 1 new comment

• Distributed TensorFlow got error message with the MPI collective Ops units test

  #18092 commented on Sep 1, 2021 • 1 new comment

• ImportError: libcublas.so.9.0: cannot open shared object file: No such file or directory

  #18237 commented on Sep 1, 2021 • 1 new comment

• Can't build tensorflow from source with docker image

  #39947 commented on Sep 1, 2021 • 1 new comment

• Training a Model

  #49464 commented on Sep 1, 2021 • 1 new comment

• Faild Build 2.5.0 Windows

  #49836 commented on Sep 1, 2021 • 1 new comment

• terminate called after throwing an instance of 'std::bad_alloc' what(): std::bad_alloc

  #49975 commented on Sep 1, 2021 • 1 new comment

• With lazy_import tensorflow cannot be imported

  #33316 commented on Sep 1, 2021 • 1 new comment

• failed to alloc X bytes unified memory; result: CUDA_ERROR_OUT_OF_MEMORY: out of memory

  #51673 commented on Sep 1, 2021 • 1 new comment

• An application using tensorflow_cc.dll crashes whenever the Tensorflow API is invoked.

  #33267 commented on Sep 1, 2021 • 1 new comment

• TFLite allocate tensors fails: (CONCATENATION) failed to prepare after input shape resize

  #34345 commented on Sep 1, 2021 • 1 new comment

• tensorflow-gpu==2.6 don't run on cuda by default

  #51644 commented on Sep 1, 2021 • 1 new comment

• Size and CombinedNMS Op support request for tflite

  #34693 commented on Sep 1, 2021 • 1 new comment

• Inconsistency in Gradients Calculation of reduce_max

  #51681 commented on Sep 1, 2021 • 1 new comment

• Built TF 1.13.1 - libcublas.so.10.1: cannot open shared object file with classify_image.py

  #27155 commented on Sep 1, 2021 • 1 new comment

• Terribly slow conditional updates

  #26841 commented on Sep 1, 2021 • 1 new comment

• flatbuffers download:

  #47485 commented on Sep 1, 2021 • 1 new comment

• tf-nightly-gpu builds from February are crashing on 'tf.keras model.fit()' for Nvidia RTX 3090; January builds performance is slower than 2080ti on : 2Layer Bidirectional LSTM + 3 Dense Layers

  #47473 commented on Sep 1, 2021 • 1 new comment

• I have to install TF 1.8!

  #37094 commented on Sep 1, 2021 • 1 new comment

• Fit failed in TPU when model contains conv_transpose

  #50567 commented on Sep 1, 2021 • 1 new comment

• TF-Nightly Broke for Multi GPU training

  #51676 commented on Sep 1, 2021 • 1 new comment

• Recurrent layers fail on (at least some) Linuxes where use_bias=True, succeeds on Macos

  #48195 commented on Sep 2, 2021 • 1 new comment

• Default values of mean and standard deviation for TFLite RNN Models.

  #51456 commented on Sep 2, 2021 • 1 new comment

• [Question/Feature request] How to stack variable length tensors in a TensorArray?

  #20870 commented on Sep 2, 2021 • 1 new comment

• BUG: Initial values are still random, even both the graph-level and the operation seed are set

  #21432 commented on Sep 2, 2021 • 1 new comment

• GPU Support - Shared Object Exclusion

  #29835 commented on Sep 2, 2021 • 1 new comment

• Error loading a TensorRT optimised graph

  #28854 commented on Sep 2, 2021 • 1 new comment

• Error: Not in GZIP format

  #35491 commented on Sep 2, 2021 • 1 new comment

• Object Detection API: Unable to train a quantization aware Faster RCNN + Resnet50 object detector

  #29736 commented on Sep 2, 2021 • 1 new comment

• Tflite results dont match corresponding Tensorflow for Mobilenet.

  #29693 commented on Sep 2, 2021 • 1 new comment

• tf.keras.layers.LocallyConnected2D crashes with implementation mode = 2 and negative strides

  #51624 commented on Sep 2, 2021 • 1 new comment

• TFLite GPU Delegates: Reshape fails to reshape input matrix with batch >1

  #29490 commented on Sep 2, 2021 • 1 new comment

• Node.js (JavaScript) TensorFlow Lite Wrapper API

  #29236 commented on Sep 2, 2021 • 1 new comment

• tensorflowlib for 2.0

  #35405 commented on Sep 2, 2021 • 1 new comment

• [CPU] Best CPU Build found with MKL config in bazel build BUT with MKL Disabled in script, especially on LSTM, can't undestand why

  #38858 commented on Sep 2, 2021 • 1 new comment

• Adding some Boosted tree ops to the 'allowed' list

  #50801 commented on Sep 2, 2021 • 1 new comment

• Exception Error while converting Keras LSTM model to TFlite model

  #51380 commented on Sep 2, 2021 • 1 new comment

• AndroidRuntime: Node number 1 (SPLIT) failed to prepare.

  #35053 commented on Sep 2, 2021 • 1 new comment

• kernel error

  #51699 commented on Sep 2, 2021 • 1 new comment

• Googleapis no longer usable as a system lib with ignored com_github_googleapis_googleapis

  #42361 commented on Sep 2, 2021 • 1 new comment

• kernel shutdown, error

  #51698 commented on Sep 2, 2021 • 1 new comment

• tensorflow/go for Windows?

  #15869 commented on Sep 2, 2021 • 1 new comment

• Performance issue with 3070

  #44936 commented on Sep 2, 2021 • 1 new comment

• The LSTM does not generate reproducible results, but GRU does

  #18323 commented on Sep 2, 2021 • 1 new comment

• RAM memory issue in Xavier

  #26359 commented on Sep 2, 2021 • 1 new comment

• DNNClassifier estimator train shows unsupported feed type error occasionally

  #25133 commented on Sep 2, 2021 • 1 new comment

• Hang on out of memory error

  #37029 commented on Sep 2, 2021 • 1 new comment

• TensorRT's create_inference_graph() produced output with too big size

  #24789 commented on Sep 2, 2021 • 1 new comment

• Added ResNext models

  #37146 commented on Sep 1, 2021 • 1 new comment

• TFLu: Add UInt8 support for logistic operator

  #46873 commented on Aug 26, 2021 • 1 new comment

• [GO] Memory leak when loading in a string source in encodeTensorWithSlices

  #48051 commented on Aug 26, 2021 • 1 new comment

• [TF:TRT] ConvertSplit for dynamic shapes.

  #48246 commented on Aug 26, 2021 • 1 new comment

• When events are specified as root events by user, they should be treated as root events in trace view.

  #48922 commented on Aug 26, 2021 • 1 new comment

• Issue tensorflow#33840 in tf.image.ssim_multiscale

  #48938 commented on Aug 26, 2021 • 1 new comment

• Half pixel centers

  #49042 commented on Aug 26, 2021 • 1 new comment

• Made rounding in convert_image_dtype for numbers close to zero

  #49868 commented on Sep 1, 2021 • 1 new comment

• Extend the algorithm selector to consider a list of allowed and disallowed tactics

  #50408 commented on Aug 31, 2021 • 1 new comment

• [TFLite] Quantized int8/16 support for TILE operator

  #50424 commented on Aug 31, 2021 • 1 new comment

• support fixed64 in features parser .

  #50451 commented on Aug 31, 2021 • 1 new comment

• [PluggableDevice] register default device for wrapdataset ops

  #50580 commented on Sep 2, 2021 • 1 new comment

• Update errors_impl.py

  #50672 commented on Sep 1, 2021 • 1 new comment

• dockerfiles: pin jedi to allow ipython tab completion

  #50754 commented on Sep 2, 2021 • 1 new comment

• Move code shared between lite and micro to elementwise_portable.h

  #50836 commented on Sep 1, 2021 • 1 new comment

• Support per-channel quantized INT8 weights unpacking in XNNPACK delegate

  #50875 commented on Aug 27, 2021 • 1 new comment

• [TF-TRT] Refactoring Top Level __init__ file for TF-TRT

  #50884 commented on Sep 1, 2021 • 1 new comment

• [XLA:GPU] BatchNorm lowering overhaul for cudnn fast fp16 kernels and act+side input fusion

  #50914 commented on Sep 1, 2021 • 1 new comment

• [PluggableDevice] fix While segment fault on PluggableDevice

  #50916 commented on Aug 31, 2021 • 1 new comment

• [oneDNN] Improving Graph Rewrite Performance

  #50932 commented on Sep 1, 2021 • 1 new comment

• Crash fix cudaMallocAsync usage of TF_CUDA_MALLOC_ASYNC_SUPPORTED_PREALLOC.

  #50962 commented on Sep 1, 2021 • 1 new comment

• Refactor Slice reference op for port to TFLM

  #51672 commented on Sep 2, 2021 • 1 new comment

• For loop in "call" function not working with Autograph/tf.range

  #51585 commented on Aug 26, 2021 • 0 new comments

• ValueError: `tape` is required when a `Tensor` loss is passed.

  #51622 commented on Aug 26, 2021 • 0 new comments

• DepthwiseConv2D is slower than Conv2D

  #46217 commented on Aug 27, 2021 • 0 new comments

• Customize continuous string tensor type for feature hashing on GPU.

  #48520 commented on Aug 27, 2021 • 0 new comments

• The Phi definition of GELU is incorrect

  #50510 commented on Aug 27, 2021 • 0 new comments

• experimental_connect_to_cluster hangs

  #34564 commented on Aug 27, 2021 • 0 new comments

• Different inference values tensorflow vs tflite model

  #42346 commented on Aug 27, 2021 • 0 new comments

• MaybeLockVariableInputMutexesInOrder() from training_op_helpers.h is broken

  #51686 commented on Aug 27, 2021 • 0 new comments

• tfa.activations.snake - wrong function in docs

  #50737 commented on Aug 27, 2021 • 0 new comments

• Tensorflow estimator returns the same answer

  #50872 commented on Aug 27, 2021 • 0 new comments

• tf.data.Dataset.from_generator converts input argument types implicitly instead of just forwarding

  #35944 commented on Aug 27, 2021 • 0 new comments

• google.protobuf.message.DecodeError: Error parsing message when using tf.function

  #51588 commented on Aug 27, 2021 • 0 new comments

• Nan gradient when calling tf.gradients() more than 1 time in graph mode for TF Dormand-Prince solver

  #51688 commented on Aug 27, 2021 • 0 new comments

• Grammatical Mistakes in the documentation of tf.keras.Model.load_weights

  #50536 commented on Aug 27, 2021 • 0 new comments

• use the stream builder to invoke FFT library fail in custom op

  #20142 commented on Aug 29, 2021 • 0 new comments

• mkl_cpu_allocator.h is not compiled under windows anymoe

  #15690 commented on Aug 30, 2021 • 0 new comments

• Huge CPU, GPU output diff after updates to fully_connected GpuDelegate gl kernel.

  #42294 commented on Aug 30, 2021 • 0 new comments

• how to reduce libtensorflowlite.so size

  #48068 commented on Aug 30, 2021 • 0 new comments

• win10 build tensorflow2.4.1,ERROR: C:/tensorflow/tensorflow/python/util/BUILD:609:27: C++ compilation of rule '//tensorflow/python/util:fast_module_type.so' failed (Exit 2): python.exe failed: error executing command

  #48040 commented on Aug 30, 2021 • 0 new comments

• TensorFlow Lite Op Request

  #29887 commented on Aug 30, 2021 • 0 new comments

• DSP HiFi4 support (delegate) for TF-Lite

  #47217 commented on Aug 30, 2021 • 0 new comments

• tflite custom Bazel build for wasm target

  #46359 commented on Aug 30, 2021 • 0 new comments

• Support strings as a first-class input/output data type in Swift

  #41131 commented on Aug 30, 2021 • 0 new comments

• when support tflite op: SparseFillEmptyRows

  #40427 commented on Aug 30, 2021 • 0 new comments

• Roadmap for publishing TFLite runtime on PyPi?

  #38642 commented on Aug 30, 2021 • 0 new comments

• Tensorflow lite NnApiDelegate crashes on Pixel 3a XL

  #36645 commented on Aug 30, 2021 • 0 new comments

• Android TFlite inconsistent performance when app is not in focus.

  #35283 commented on Aug 30, 2021 • 0 new comments

• tf.signal.mfccs_from_log_mel_spectrograms caused internal error in converted TensorFlow Lite model

  #34326 commented on Aug 30, 2021 • 0 new comments

• Need tf.signal.rfft op in TFLite

  #27303 commented on Aug 30, 2021 • 0 new comments

• googletest.h used in open source project

  #26671 commented on Aug 30, 2021 • 0 new comments

• Two roundings in MultiplyByQuantizedMultiplier(TFLite) leads to inconsistent result with tensorflow

  #25087 commented on Aug 30, 2021 • 0 new comments

• Android tfLite Shared STL support

  #22531 commented on Aug 30, 2021 • 0 new comments

• Address missing TensorFlow operations to TFLite:

  #21526 commented on Aug 30, 2021 • 0 new comments

• Feature request: Mask R-CNN support on TensorFlow Lite

  #19293 commented on Aug 30, 2021 • 0 new comments

• Cannot install tensorflow go

  #36495 commented on Aug 31, 2021 • 0 new comments

• Layer_by_layer Boosting Tree

  #51549 commented on Aug 31, 2021 • 0 new comments

• Branch r2.0 Failed Source Build on macOS Mojave

  #31247 commented on Aug 31, 2021 • 0 new comments

• Adam implementation differs from paper (applies bias B_2 correction to \epsilon)

  #35102 commented on Aug 31, 2021 • 0 new comments

• Cannot load model checkpoints: Two checkpoint references resolved to different objects

  #51640 commented on Aug 31, 2021 • 0 new comments

• [Find header file after build manually tensor flow lite]

  #29686 commented on Sep 1, 2021 • 0 new comments

• Memory leak using C_API

  #29217 commented on Sep 1, 2021 • 0 new comments

• [Translation] Translation for Korean

  #28994 commented on Sep 1, 2021 • 0 new comments

• INTERNAL ERROR reported while trying to apply GpuDelegate to tflite

  #28902 commented on Sep 1, 2021 • 0 new comments

• Can anyone build a optimized tensorflow-gpu-v2.0 based on cuda10.0 and cudnn7.6(centos7) for me?

  #35543 commented on Sep 1, 2021 • 0 new comments

• windows build failed with master branch

  #35977 commented on Sep 1, 2021 • 0 new comments

• Possible bug with session handle as feed dict input

  #51344 commented on Sep 1, 2021 • 0 new comments

• SVD handles small singular values poorly

  #26842 commented on Sep 1, 2021 • 0 new comments

• [TF 2.0 API Docs] tf.audio.encode_wav

  #28238 commented on Sep 1, 2021 • 0 new comments

• Memory leak activated when setting a seed in TensorFlow 2 (2.0 & 2.1) and using eager execution.

  #37252 commented on Sep 1, 2021 • 0 new comments

• TF Real Op Not Supported for TFLite when generating MFCCs

  #33897 commented on Sep 2, 2021 • 0 new comments

• [TF 2.0] tf.summary centralized step management

  #26406 commented on Sep 2, 2021 • 0 new comments

• Incorrect flops calculation for operations with complex numbers

  #27266 commented on Sep 2, 2021 • 0 new comments

• compute_mask not called when custom layer does not alter inputs and mask is not None

  #32322 commented on Sep 2, 2021 • 0 new comments

• Bazel build does not pick up correct compiler include paths

  #33975 commented on Sep 2, 2021 • 0 new comments

• tf.quantization.fake_quant_with_min_max_vars(_gradient) crash(abort)

  #46910 commented on Sep 2, 2021 • 0 new comments

• [Intel MKL] unit test for bfloat16 on matmul.

  #42949 commented on Aug 30, 2021 • 0 new comments

• [Intel MKL] Fuse group of primitive ops for batch normalization to Fu…

  #44950 commented on Aug 26, 2021 • 0 new comments

• [INTEL MKL] Convert necessary PlaceholderWithDefault or Placeholder to Constant f…

  #45822 commented on Aug 26, 2021 • 0 new comments

• [INTEL MKL] Refactor mkl_relu_ops files, and add accuracy and performace UT.

  #46457 commented on Aug 26, 2021 • 0 new comments

• [Intel MKL] Fix shape inference for Quantized{Conv2D/DepthwiseConv2D}-like operations

  #46779 commented on Aug 26, 2021 • 0 new comments

• Custom os.path.join that is aware of TF filesystems

  #48125 commented on Aug 31, 2021 • 0 new comments

• Map XLA opaque type to a scalar TensorShape in XLAShapeToTensorShape.

  #48233 commented on Aug 26, 2021 • 0 new comments

• Add elastic training for workers in PSv2

  #48601 commented on Aug 27, 2021 • 0 new comments

• Update ICU to 69.1

  #49026 commented on Aug 26, 2021 • 0 new comments

• Ensure default types are accurate: fix `int` given for `float` args

  #49974 commented on Aug 30, 2021 • 0 new comments

• XLA persistent compilation cache

  #50071 commented on Aug 27, 2021 • 0 new comments

• [WIP][MLIR][DISC] Add device type info to memref's address space in lmhlo dialect

  #50365 commented on Aug 26, 2021 • 0 new comments

• Fix TransferLiteralFromDevice bug which appears in Tensorflow Master

  #50485 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice]add default device for function ops

  #50578 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] register default devices for functional ops

  #50581 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for datasets ops

  #50600 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for Mutex/DestroyResource/Sequence ops

  #50601 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for python ops

  #50602 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for batch_kernel/concat/fact/partioned_function ops

  #50603 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for map_stage ops

  #50604 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for iterator/optional ops

  #50605 commented on Aug 31, 2021 • 0 new comments

• [PluggableDevice] add DEVICE_DEFAULT for scopedallocator/shape ops

  #50608 commented on Aug 31, 2021 • 0 new comments

• Documentation issue read me.

  #50729 commented on Sep 1, 2021 • 0 new comments

• [TFLite] Add int8 and int16x8 support for the ROUND operator

  #50786 commented on Sep 1, 2021 • 0 new comments

• Fix numerical stability issue in binary search

  #50855 commented on Sep 2, 2021 • 0 new comments

• Remove hdfs as support has been moved to modular file systems.

  #51033 commented on Aug 31, 2021 • 0 new comments

• Enable resource update aliasing in auto-clustering

  #51353 commented on Aug 30, 2021 • 0 new comments

• TF Lite unit (kernel) tests CMake build fix

  #51406 commented on Sep 1, 2021 • 0 new comments

• add int8 support for tf.tile

  #51480 commented on Aug 31, 2021 • 0 new comments

• [TF:TRT] Replace c asserts with CHECK macro

  #51498 commented on Aug 30, 2021 • 0 new comments

• TF-TRT Set binding dimensions even if tensor is not an execution tensor

  #51697 commented on Aug 30, 2021 • 0 new comments