List of TensorFlow Ops that have valid gradients

AccumulatorApplyGradient	Applies a gradient to a given accumulator.
AccumulatorNumAccumulated	Returns the number of gradients aggregated in the given accumulators.
AccumulatorSetGlobalStep	Updates the accumulator with a new value for global_step.
AccumulatorTakeGradient	Extracts the average gradient in the given ConditionalAccumulator.
Add	Returns x + y element-wise.
AudioSummary	Outputs a Summary protocol buffer with audio.
AudioSummaryV2	Outputs a Summary protocol buffer with audio.
BarrierClose	Closes the given barrier.
BarrierIncompleteSize	Computes the number of incomplete elements in the given barrier.
BarrierInsertMany	For each key, assigns the respective value to the specified component.
BarrierReadySize	Computes the number of complete elements in the given barrier.
BarrierTakeMany	Takes the given number of completed elements from a barrier.
DecodeBase64	Decode web-safe base64-encoded strings.
DecodeBmp	Decode the first frame of a BMP-encoded image to a uint8 tensor.
DecodeCSV	Convert CSV records to tensors. Each column maps to one tensor.
DecodeGif	Decode the first frame of a GIF-encoded image to a uint8 tensor.
DecodeJpeg	Decode a JPEG-encoded image to a uint8 tensor.
DecodeJSONExample	Convert JSON-encoded Example records to binary protocol buffer strings.
DecodePng	Decode a PNG-encoded image to a uint8 or uint16 tensor.
DecodeRaw	Reinterpret the bytes of a string as a vector of numbers.
DecodeWav	Decode a 16-bit PCM WAV file to a float tensor.
DeleteSessionTensor	Delete the tensor specified by its handle in the session.
DenseToDenseSetOperation	Applies set operation along last dimension of 2 Tensor inputs.
DenseToSparseSetOperation	Applies set operation along last dimension of Tensor and SparseTensor.
DeserializeManySparse	Deserialize and concatenate SparseTensors from a serialized minibatch.
EncodeBase64	Encode strings into web-safe base64 format.
Equal	Returns the truth value of (x == y) element-wise.
FixedLengthRecordDataset	Creates a dataset that emits the records from one or more binary files.
GetSessionTensor	Get the value of the tensor specified by its handle.
HistogramSummary	Outputs a Summary protocol buffer with a histogram.
ImageSummary	Outputs a Summary protocol buffer with images.
InitializeTable	Table initializer that takes two tensors for keys and values respectively.
InitializeTableFromTextFile	Initializes a table from a text file.
InitializeTableFromTextFileV2	Initializes a table from a text file.
LookupTableExport	Outputs all keys and values in the table.
LookupTableFind	Looks up keys in a table, outputs the corresponding values.
LookupTableImport	Replaces the contents of the table with the specified keys and values.
LookupTableInsert	Updates the table to associates keys with values.
LookupTableSize	Computes the number of elements in the given table.
MatchingFiles	Returns the set of files matching one or more glob patterns.
MergeSummary	Merges summaries.
MergeV2Checkpoints	V2 format specific: merges the metadata files of sharded checkpoints. The
NotEqual	Returns the truth value of (x != y) element-wise.
ParseExample	Transforms a vector of brain.Example protos (as strings) into typed tensors.
ParseSingleSequenceExample	Transforms a scalar brain.SequenceExample proto (as strings) into typed tensors.
ParseTensor	Transforms a serialized tensorflow.TensorProto proto into a Tensor.
QueueClose	Closes the given queue.
QueueDequeue	Dequeues a tuple of one or more tensors from the given queue.
QueueDequeueMany	Dequeues n tuples of one or more tensors from the given queue.
QueueDequeueUpTo	Dequeues n tuples of one or more tensors from the given queue.
QueueEnqueue	Enqueues a tuple of one or more tensors in the given queue.
QueueEnqueueMany	Enqueues zero or more tuples of one or more tensors in the given queue.
QueueSize	Computes the number of elements in the given queue.
ReaderNumRecordsProduced	Returns the number of records this Reader has produced.
ReaderNumWorkUnitsCompleted	Returns the number of work units this Reader has finished processing.
ReaderRead	Returns the next record (key, value pair) produced by a Reader.
ReaderReadUpTo	Returns up to num_records (key, value) pairs produced by a Reader.
ReaderReset	Restore a Reader to its initial clean state.
ReaderRestoreState	Restore a reader to a previously saved state.
ReaderRestoreStateV2	Restore a reader to a previously saved state.
ReaderSerializeState	Produce a string tensor that encodes the state of a Reader.
ReadFile	Reads and outputs the entire contents of the input filename.
ReduceJoin	Joins a string Tensor across the given dimensions.
Restore	Restores a tensor from checkpoint files.
RestoreSlice	Restores a tensor from checkpoint files.
RestoreV2	Restores tensors from a V2 checkpoint.
Reverse	Reverses specific dimensions of a tensor.
Save	Saves the input tensors to disk.
SaveSlices	Saves input tensors slices to disk.
SaveV2	Saves tensors in V2 checkpoint format.
ScalarSummary	Outputs a Summary protocol buffer with scalar values.
SdcaFprint	Computes fingerprints of the input strings.
SetSize	Number of unique elements along last dimension of input set.
ShardedFilename	Generate a sharded filename. The filename is printf formatted as
ShardedFilespec	Generate a glob pattern matching all sharded file names.
SparseAccumulatorApplyGradient	Applies a sparse gradient to a given accumulator.
SparseAccumulatorTakeGradient	Extracts the average sparse gradient in a SparseConditionalAccumulator.
SparseCross	Generates sparse cross from a list of sparse and dense tensors.
SparseToSparseSetOperation	Applies set operation along last dimension of 2 SparseTensor inputs.
StackClose	Delete the stack from its resource container.
StackPop	Pop the element at the top of the stack.
StackPush	Push an element onto the stack.
StringJoin	Joins the strings in the given list of string tensors into one tensor;
StringSplit	Split elements of input based on delimiter into a SparseTensor.
StringToHashBucket	Converts each string in the input Tensor to its hash mod by a number of buckets.
StringToHashBucketFast	Converts each string in the input Tensor to its hash mod by a number of buckets.
StringToHashBucketStrong	Converts each string in the input Tensor to its hash mod by a number of buckets.
StringToNumber	Converts each string in the input Tensor to the specified numeric type.
Substr	Return substrings from Tensor of strings.
TensorArrayClose
TensorArrayCloseV2	Deprecated. Use TensorArrayCloseV3
TensorArrayConcat
TensorArrayConcatV2	Deprecated. Use TensorArrayConcatV3
TensorArrayGather
TensorArrayGatherV2	Deprecated. Use TensorArrayGatherV3
TensorArrayGrad
TensorArrayGradV2	Deprecated. Use TensorArrayGradV3
TensorArrayPack
TensorArrayRead
TensorArrayReadV2	Deprecated. Use TensorArrayReadV3
TensorArrayScatter
TensorArrayScatterV2	Deprecated. Use TensorArrayScatterV3
TensorArraySize
TensorArraySizeV2	Deprecated. Use TensorArraySizeV3
TensorArraySplit
TensorArraySplitV2	Deprecated. Use TensorArraySplitV3
TensorArrayUnpack
TensorArrayWrite
TensorArrayWriteV2	Deprecated. Use TensorArrayGradV3
TextLineDataset	Creates a dataset that emits the lines of one or more text files.
TFRecordDataset	Creates a dataset that emits the records from one or more TFRecord files.
WriteFile	Writes contents to the file at input filename. Creates file if not existing.