509 lines (486 loc) · 33.5 KB

TensorFlow Python reference documentation

Building Graphs:
- add_to_collection
- as_dtype
- bytes
- control_dependencies
- convert_to_tensor
- convert_to_tensor_or_indexed_slices
- device
- Dimension
- DType
- get_collection
- get_collection_ref
- get_default_graph
- get_seed
- Graph
- GraphKeys
- import_graph_def
- load_file_system_library
- load_op_library
- name_scope
- NoGradient
- op_scope
- Operation
- register_tensor_conversion_function
- RegisterGradient
- RegisterShape
- reset_default_graph
- Tensor
- TensorShape
Constants, Sequences, and Random Values:
- constant
- fill
- linspace
- ones
- ones_like
- random_crop
- random_normal
- random_shuffle
- random_uniform
- range
- set_random_seed
- truncated_normal
- zeros
- zeros_like
Variables:
- all_variables
- assert_variables_initialized
- assign
- assign_add
- assign_sub
- constant_initializer
- count_up_to
- device
- get_checkpoint_state
- get_variable
- get_variable_scope
- IndexedSlices
- initialize_all_variables
- initialize_local_variables
- initialize_variables
- is_variable_initialized
- latest_checkpoint
- local_variables
- make_template
- moving_average_variables
- no_regularizer
- ones_initializer
- random_normal_initializer
- random_uniform_initializer
- Saver
- scatter_add
- scatter_sub
- scatter_update
- sparse_mask
- trainable_variables
- truncated_normal_initializer
- uniform_unit_scaling_initializer
- update_checkpoint_state
- Variable
- variable_op_scope
- variable_scope
- VariableScope
- zeros_initializer
Tensor Transformations:
- bitcast
- boolean_mask
- cast
- concat
- depth_to_space
- dynamic_partition
- dynamic_stitch
- expand_dims
- gather
- gather_nd
- one_hot
- pack
- pad
- rank
- reshape
- reverse
- reverse_sequence
- saturate_cast
- shape
- shape_n
- size
- slice
- space_to_depth
- split
- squeeze
- string_to_number
- tile
- to_bfloat16
- to_double
- to_float
- to_int32
- to_int64
- transpose
- unique_with_counts
- unpack
Math:
- abs
- accumulate_n
- add
- add_n
- argmax
- argmin
- batch_cholesky
- batch_fft
- batch_fft2d
- batch_fft3d
- batch_ifft
- batch_ifft2d
- batch_ifft3d
- batch_matmul
- batch_matrix_determinant
- batch_matrix_diag
- batch_matrix_diag_part
- batch_matrix_inverse
- batch_matrix_solve
- batch_matrix_solve_ls
- batch_matrix_triangular_solve
- batch_self_adjoint_eig
- ceil
- cholesky
- complex
- complex_abs
- conj
- cos
- cross
- diag
- diag_part
- digamma
- div
- edit_distance
- erf
- erfc
- exp
- fft
- fft2d
- fft3d
- floor
- floordiv
- ifft
- ifft2d
- ifft3d
- igamma
- igammac
- imag
- inv
- invert_permutation
- lbeta
- lgamma
- listdiff
- log
- matmul
- matrix_determinant
- matrix_inverse
- matrix_solve
- matrix_solve_ls
- matrix_triangular_solve
- maximum
- minimum
- mod
- mul
- neg
- pow
- real
- reduce_all
- reduce_any
- reduce_max
- reduce_mean
- reduce_min
- reduce_prod
- reduce_sum
- round
- rsqrt
- scalar_mul
- segment_max
- segment_mean
- segment_min
- segment_prod
- segment_sum
- self_adjoint_eig
- sign
- sin
- sparse_segment_mean
- sparse_segment_sqrt_n
- sparse_segment_sqrt_n_grad
- sparse_segment_sum
- sqrt
- square
- squared_difference
- sub
- trace
- transpose
- truediv
- unique
- unsorted_segment_sum
- where
Strings:
- reduce_join
- string_to_hash_bucket
Histograms:
- histogram_fixed_width
Control Flow:
- add_check_numerics_ops
- Assert
- case
- check_numerics
- cond
- count_up_to
- equal
- greater
- greater_equal
- group
- identity
- is_finite
- is_inf
- is_nan
- less
- less_equal
- logical_and
- logical_not
- logical_or
- logical_xor
- no_op
- not_equal
- Print
- select
- tuple
- verify_tensor_all_finite
- where
- while_loop
Higher Order Functions:
- foldl
- foldr
- map_fn
- scan
Tensor Handle Operations:
- delete_session_tensor
- get_session_handle
- get_session_tensor
Images:
- adjust_brightness
- adjust_contrast
- adjust_hue
- adjust_saturation
- central_crop
- convert_image_dtype
- crop_to_bounding_box
- decode_jpeg
- decode_png
- draw_bounding_boxes
- encode_jpeg
- encode_png
- extract_glimpse
- flip_left_right
- flip_up_down
- grayscale_to_rgb
- hsv_to_rgb
- pad_to_bounding_box
- per_image_whitening
- random_brightness
- random_contrast
- random_flip_left_right
- random_flip_up_down
- random_hue
- random_saturation
- resize_area
- resize_bicubic
- resize_bilinear
- resize_image_with_crop_or_pad
- resize_images
- resize_nearest_neighbor
- rgb_to_grayscale
- rgb_to_hsv
- sample_distorted_bounding_box
- transpose_image
Sparse Tensors:
- shape
- sparse_add
- sparse_concat
- sparse_fill_empty_rows
- sparse_merge
- sparse_reorder
- sparse_retain
- sparse_split
- sparse_tensor_dense_matmul
- sparse_tensor_to_dense
- sparse_to_dense
- sparse_to_indicator
- SparseTensor
- SparseTensorValue
Inputs and Readers:
- batch
- batch_join
- decode_csv
- decode_json_example
- decode_raw
- FIFOQueue
- FixedLenFeature
- FixedLengthRecordReader
- FixedLenSequenceFeature
- IdentityReader
- input_producer
- limit_epochs
- match_filenames_once
- matching_files
- parse_example
- parse_single_example
- placeholder
- placeholder_with_default
- QueueBase
- RandomShuffleQueue
- range_input_producer
- read_file
- ReaderBase
- shuffle_batch
- shuffle_batch_join
- size
- slice_input_producer
- string_input_producer
- TextLineReader
- TFRecordReader
- VarLenFeature
- WholeFileReader
Data IO (Python functions):
- tf_record_iterator
- TFRecordWriter
Neural Network:
- avg_pool
- batch_normalization
- bias_add
- compute_accidental_hits
- conv2d
- conv2d_transpose
- depthwise_conv2d
- depthwise_conv2d_native
- dropout
- elu
- embedding_lookup
- embedding_lookup_sparse
- fixed_unigram_candidate_sampler
- in_top_k
- l2_loss
- l2_normalize
- learned_unigram_candidate_sampler
- local_response_normalization
- log_softmax
- log_uniform_candidate_sampler
- max_pool
- max_pool_with_argmax
- moments
- nce_loss
- normalize_moments
- relu
- relu6
- sampled_softmax_loss
- separable_conv2d
- sigmoid
- sigmoid_cross_entropy_with_logits
- softmax
- softmax_cross_entropy_with_logits
- softplus
- softsign
- sparse_softmax_cross_entropy_with_logits
- sufficient_statistics
- tanh
- top_k
- uniform_candidate_sampler
- weighted_cross_entropy_with_logits
Running Graphs:
- AbortedError
- AlreadyExistsError
- CancelledError
- DataLossError
- DeadlineExceededError
- FailedPreconditionError
- get_default_session
- InteractiveSession
- InternalError
- InvalidArgumentError
- NotFoundError
- OpError
- OutOfRangeError
- PermissionDeniedError
- ResourceExhaustedError
- Session
- UnauthenticatedError
- UnavailableError
- UnimplementedError
- UnknownError
Training:
- AdadeltaOptimizer
- AdagradOptimizer
- AdamOptimizer
- add_queue_runner
- AggregationMethod
- clip_by_average_norm
- clip_by_global_norm
- clip_by_norm
- clip_by_value
- ClusterSpec
- Coordinator
- exponential_decay
- ExponentialMovingAverage
- export_meta_graph
- FtrlOptimizer
- generate_checkpoint_state_proto
- global_norm
- global_step
- GradientDescentOptimizer
- gradients
- histogram_summary
- image_summary
- import_meta_graph
- LooperThread
- merge_all_summaries
- merge_summary
- MomentumOptimizer
- Optimizer
- QueueRunner
- replica_device_setter
- RMSPropOptimizer
- scalar_summary
- Server
- SessionManager
- start_queue_runners
- stop_gradient
- summary_iterator
- SummaryWriter
- Supervisor
- write_graph
- zero_fraction
Wraps python functions:
- py_func
Testing:
- assert_equal_graph_def
- compute_gradient
- compute_gradient_error
- get_temp_dir
- is_built_with_cuda
- main
Layers (contrib):
- apply_regularization
- convolution2d
- fully_connected
- l1_regularizer
- l2_regularizer
- make_all
- optimize_loss
- sum_regularizer
- summarize_activation
- summarize_activations
- summarize_collection
- summarize_tensor
- summarize_tensors
- xavier_initializer
- xavier_initializer_conv2d
Utilities (contrib):
- constant_value
- make_ndarray
- make_tensor_proto
- stripped_op_list_for_graph