io.py 23.0 KB
Newer Older
1
#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserved.
D
dzhwinter 已提交
2
#
D
dzhwinter 已提交
3 4 5
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
D
dzhwinter 已提交
6
#
D
dzhwinter 已提交
7
#     http://www.apache.org/licenses/LICENSE-2.0
D
dzhwinter 已提交
8
#
D
dzhwinter 已提交
9 10 11 12 13
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
F
fengjiayi 已提交
14
import contextlib
D
dzhwinter 已提交
15

Y
Yu Yang 已提交
16
from .. import core
T
typhoonzero 已提交
17
from ..framework import convert_np_dtype_to_dtype_, default_main_program, default_startup_program, Program
Y
Yu Yang 已提交
18
from ..unique_name import generate as unique_name
T
WIP  
typhoonzero 已提交
19 20
from control_flow import BlockGuard
from ..layer_helper import LayerHelper
Y
Refine  
Yu Yang 已提交
21
from ..executor import global_scope
Y
yuyang18 已提交
22
from layer_function_generator import generate_layer_fn, templatedoc
Y
Yu Yang 已提交
23

Y
Yu Yang 已提交
24 25
__all__ = [
    'data', 'BlockGuardServ', 'ListenAndServ', 'Send', 'open_recordio_file',
F
fengjiayi 已提交
26
    'open_files', 'read_file', 'shuffle', 'batch', 'double_buffer',
Y
yuyang18 已提交
27
    'random_data_generator', 'Preprocessor', 'load'
Y
Yu Yang 已提交
28
]
Y
Yu Yang 已提交
29 30 31 32 33 34 35 36 37 38


def data(name,
         shape,
         append_batch_size=True,
         dtype='float32',
         lod_level=0,
         type=core.VarDesc.VarType.LOD_TENSOR,
         stop_gradient=True):
    """
K
kavyasrinet 已提交
39
    **Data Layer**
Y
Yu Yang 已提交
40

K
kavyasrinet 已提交
41
    This function takes in the input and based on whether data has
C
caoying03 已提交
42
    to be returned back as a minibatch, it creates the global variable by using
Y
Yu Yang 已提交
43
    the helper functions. The global variables can be accessed by all the
C
caoying03 已提交
44
    following operators in the graph.
Y
Yu Yang 已提交
45 46 47 48

    All the input variables of this function are passed in as local variables
    to the LayerHelper constructor.

K
kavyasrinet 已提交
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64
    Args:
       name(str): The name/alias of the function
       shape(list): Tuple declaring the shape.
       append_batch_size(bool): Whether or not to append the data as a batch.
       dtype(int|float): The type of data : float32, float_16, int etc
       type(VarType): The output type. By default it is LOD_TENSOR.
       lod_level(int): The LoD Level. 0 means the input data is not a sequence.
       stop_gradient(bool): A boolean that mentions whether gradient should flow.

    Returns:
        Variable: The global variable that gives access to the data.

    Examples:
        .. code-block:: python

          data = fluid.layers.data(name='x', shape=[784], dtype='float32')
Y
Yu Yang 已提交
65 66 67 68 69 70 71 72 73 74 75 76 77
    """
    helper = LayerHelper('data', **locals())
    shape = list(shape)
    for i in xrange(len(shape)):
        if shape[i] is None:
            shape[i] = -1
            append_batch_size = False
        elif shape[i] < 0:
            append_batch_size = False

    if append_batch_size:
        shape = [-1] + shape  # append batch size as -1

Y
Yu Yang 已提交
78
    data_var = helper.create_global_variable(
Y
Yu Yang 已提交
79 80 81 82 83
        name=name,
        shape=shape,
        dtype=dtype,
        type=type,
        stop_gradient=stop_gradient,
F
fengjiayi 已提交
84 85
        lod_level=lod_level,
        is_data=True)
Y
Yu Yang 已提交
86
    return data_var
T
typhoonzero 已提交
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117


class BlockGuardServ(BlockGuard):
    """
    BlockGuardServ class.

    BlockGuardServ class is used to create an op with a block in a program.
    """

    def __init__(self, server):
        if not (isinstance(server, ListenAndServ)):
            raise TypeError("BlockGuardServ takes a ListenAndServ")
        super(BlockGuardServ, self).__init__(server.helper.main_program)
        self.server = server

    def __exit__(self, exc_type, exc_val, exc_tb):
        if exc_type is not None:
            return False

        self.server.complete_op()
        return super(BlockGuardServ, self).__exit__(exc_type, exc_val, exc_tb)


class ListenAndServ(object):
    """
    ListenAndServ class.

    ListenAndServ class is used to wrap listen_and_serv op to create a server
    which can receive variables from clients and run a block.
    """

Y
Yancey1989 已提交
118
    def __init__(self, endpoint, inputs, fan_in=1, optimizer_mode=True):
119
        self.helper = LayerHelper("listen_and_serv")
Y
Yancey1989 已提交
120
        self.inputs = inputs
T
typhoonzero 已提交
121 122 123
        self.outputs = []
        self.endpoint = endpoint
        self.fan_in = fan_in
T
typhoonzero 已提交
124 125
        # FIXME(typhoonzero): add optimizer_mode is stupid, should make it more
        # general.
T
WIP  
typhoonzero 已提交
126
        self.optimizer_mode = optimizer_mode
T
typhoonzero 已提交
127 128 129 130 131 132 133 134 135 136 137 138 139

    def do(self):
        return BlockGuardServ(self)

    def get_params_and_grads(self):
        main_program = self.helper.main_program
        current_block = main_program.current_block()
        parent_block = self.parent_block()
        # params and grads in the same order.
        params = list()
        grads = list()
        for op in current_block.ops:
            # FIXME(typhoonzero): op.inputs is None if it's cloned.
T
WIP  
typhoonzero 已提交
140 141 142 143 144 145 146 147
            if self.optimizer_mode:
                if "Grad" in op.inputs and "Param" in op.inputs:
                    params.append(op.inputs["Param"].name)
                    grads.append(op.inputs["Grad"].name)
            else:
                # simple recv mode, recv operators inputs.
                for iname in op.input_names:
                    for in_var_name in op.input(iname):
T
typhoonzero 已提交
148 149
                        params.append(parent_block.var(in_var_name))
                        grads.append(parent_block.var(in_var_name))
T
typhoonzero 已提交
150 151 152

        return params, grads

T
typhoonzero 已提交
153 154 155 156 157 158 159
    def parent_block(self):
        prog = self.helper.main_program
        parent_idx = prog.current_block().parent_idx
        assert parent_idx >= 0
        parent_block = prog.block(parent_idx)
        return parent_block

T
typhoonzero 已提交
160 161 162 163
    def complete_op(self):
        main_program = self.helper.main_program
        current_block = main_program.current_block()
        parent_block = self.parent_block()
T
typhoonzero 已提交
164
        empty_block = Program().global_block()
T
typhoonzero 已提交
165 166

        parent_block.append_op(
167
            type='listen_and_serv',
Y
Yancey1989 已提交
168
            inputs={"X": self.inputs},
T
typhoonzero 已提交
169 170 171 172
            outputs={},
            attrs={
                'endpoint': self.endpoint,
                'Fanin': self.fan_in,
T
typhoonzero 已提交
173
                'OptimizeBlock': current_block,
174 175
                'PrefetchBlock': empty_block,
                'sync_mode': True,  # did not support async now in layers
Q
qiaolongfei 已提交
176
                'grad_to_block_id': [""]
T
typhoonzero 已提交
177 178 179
            })


T
typhoonzero 已提交
180
def Send(endpoints, send_vars, get_vars=None):
T
typhoonzero 已提交
181 182 183 184 185 186 187 188 189 190 191 192 193 194 195
    """
    Send layer

    Args:
        endpoints: comma seperated IP:PORT pairs in the order
                   of send_vars to send
        send_vars: vars to send
        get_vars: vars to get from server after send completes.

    Send variables to the server side, and get vars from server
    side when server have finished running server side program.
    """
    assert (type(send_vars) == list)

    epmap = endpoints.split(",")
T
typhoonzero 已提交
196
    endpoints = list(set(epmap))
T
typhoonzero 已提交
197 198

    helper = LayerHelper("Send", **locals())
T
typhoonzero 已提交
199 200 201 202 203
    if not get_vars:
        get_vars = []
        for s in send_vars:
            v = helper.create_tmp_variable(dtype=s.dtype, stop_gradient=True)
            get_vars.append(v)
Y
Yancey1989 已提交
204
    rpc_op_role_name = core.op_proto_and_checker_maker.kOpRoleAttrName()
Y
Yancey1989 已提交
205

T
typhoonzero 已提交
206 207 208
    helper.append_op(
        type="send",
        inputs={"X": send_vars},
Y
Yancey1989 已提交
209 210 211 212 213 214 215
        outputs={"Out": get_vars},
        attrs={
            "endpoints": endpoints,
            "epmap": epmap,
            rpc_op_role_name: core.op_proto_and_checker_maker.OpRole.RPC
        })

T
typhoonzero 已提交
216
    return get_vars
217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244


def Recv(endpoints, get_vars):
    """
    Recv layer

    Args:
        endpoints: comma seperated IP:PORT pairs in the order
                   of send_vars to send
        send_vars: vars to send
        get_vars: vars to get from server after send completes.

    Send variables to the server side, and get vars from server
    side when server have finished running server side program.
    """
    assert (type(send_vars) == list)
    assert (type(get_vars) == list)

    epmap = endpoints.split(",")
    endpoints = list(set(epmap))

    helper = LayerHelper("Recv", **locals())
    helper.append_op(
        type="recv",
        inputs={"X": get_vars},
        outputs={"Out": get_vars},
        attrs={"endpoints": endpoints,
               "epmap": epmap})
Y
Yu Yang 已提交
245 246


Y
Refine  
Yu Yang 已提交
247 248 249 250 251 252 253 254 255 256
def monkey_patch_reader_methods(reader):
    def __get_reader__():
        scope = global_scope()
        var = scope.find_var(reader.name)
        return var.get_reader()

    def reset():
        return __get_reader__().reset()

    reader.reset = reset
Y
Yu Yang 已提交
257 258
    reader.stop_gradient = True
    reader.persistable = True
Y
Refine  
Yu Yang 已提交
259 260 261
    return reader


Y
Yu Yang 已提交
262 263 264 265 266
def _copy_reader_var_(block, var):
    new_var = block.create_var(name=var.name, type=core.VarDesc.VarType.READER)
    new_var.desc.set_shapes(var.desc.shapes())
    new_var.desc.set_dtypes(var.desc.dtypes())
    new_var.persistable = True
F
fengjiayi 已提交
267 268 269 270
    return new_var


def _copy_reader_create_op_(block, op):
F
fengjiayi 已提交
271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
    input_param_names = op.input_names
    new_input_map = {}
    for param_name in input_param_names:
        new_input_map[param_name] = []
        arg_names = op.input(param_name)
        for arg_name in arg_names:
            new_input_map[param_name].append(block.var(arg_name))

    output_param_names = op.output_names
    new_output_map = {}
    for param_name in output_param_names:
        new_output_map[param_name] = []
        arg_names = op.output(param_name)
        for arg_name in arg_names:
            new_output_map[param_name].append(block.var(arg_name))

F
fengjiayi 已提交
287
    new_op = block.append_op(
F
fengjiayi 已提交
288 289 290
        type=op.type,
        inputs=new_input_map,
        outputs=new_output_map,
J
JiayiFeng 已提交
291
        attrs=op.all_attrs())
F
fengjiayi 已提交
292
    return new_op
Y
Yu Yang 已提交
293 294


F
fengjiayi 已提交
295 296 297 298 299
def open_recordio_file(filename,
                       shapes,
                       lod_levels,
                       dtypes,
                       pass_num=1,
F
fengjiayi 已提交
300
                       for_parallel=True):
F
fengjiayi 已提交
301 302 303 304 305 306 307 308 309 310 311
    """
    Open a RecordIO file

    This layer takes a RecordIO file to read from and returns a Reader Variable.
    Via the Reader Variable, we can get data from the given RecordIO file.

    Args:
       filename(str): The RecordIO file's name.
       shapes(list): List of tuples which declaring data shapes.
       lod_levels(list): List of ints which declaring data lod_level.
       dtypes(list): List of strs which declaring data type.
F
fengjiayi 已提交
312
       pass_num(int): Number of passes to run.
F
fengjiayi 已提交
313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
       for_parallel(Bool): Set it as True if you are going to run
            subsequent operators in parallel.

    Returns:
       Variable: A Reader Variable via which we can get RecordIO file data.

    Examples:
       .. code-block:: python

         reader = fluid.layers.io.open_recordio_file(
                                          filename='./data.recordio',
                                          shapes=[(3,224,224), (1)],
                                          lod_levels=[0, 0],
                                          dtypes=['float32', 'int64'])

         # Via the reader, we can use 'read_file' layer to get data:
F
fengjiayi 已提交
329
         image, label = fluid.layers.io.read_file(reader)
F
fengjiayi 已提交
330
    """
Y
Yu Yang 已提交
331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354
    dtypes = [convert_np_dtype_to_dtype_(dt) for dt in dtypes]
    shape_concat = []
    ranks = []

    for shape in shapes:
        shape_concat.extend(shape)
        ranks.append(len(shape))

    var_name = unique_name('open_recordio_file')

    startup_blk = default_startup_program().current_block()
    startup_var = startup_blk.create_var(name=var_name)
    startup_blk.append_op(
        type='create_recordio_file_reader',
        outputs={'Out': [startup_var]},
        attrs={
            'shape_concat': shape_concat,
            'lod_levels': lod_levels,
            'filename': filename,
            'ranks': ranks
        })

    startup_var.desc.set_dtypes(dtypes)
    startup_var.persistable = True
F
fengjiayi 已提交
355 356
    main_prog_var = _copy_reader_var_(default_main_program().current_block(),
                                      startup_var)
F
fengjiayi 已提交
357 358 359 360 361

    if pass_num > 1:
        main_prog_var = multi_pass(reader=main_prog_var, pass_num=pass_num)

    if for_parallel:
J
JiayiFeng 已提交
362
        main_prog_var = parallel(reader=main_prog_var)
F
fengjiayi 已提交
363

F
fengjiayi 已提交
364
    return monkey_patch_reader_methods(main_prog_var)
Y
Yu Yang 已提交
365 366


F
fengjiayi 已提交
367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433
def random_data_generator(low, high, shapes, lod_levels, for_parallel=True):
    """
    Create a uniform random data generator

    This layer returns a Reader Variable.
    Instead of opening a file and reading data from it, this 
    Reader Variable generates float uniform random data by itself. 
    It can be used as a dummy reader to test a network without 
    opening a real file.

    Args:
       low(float): The lower bound of data's uniform distribution.
       high(float): The upper bound of data's uniform distribution.
       shapes(list): List of tuples which declaring data shapes.
       lod_levels(list): List of ints which declaring data lod_level.
       for_parallel(Bool): Set it as True if you are going to run
            subsequent operators in parallel.

    Returns:
       Variable: A Reader Variable from which we can get random data.

    Examples:
       .. code-block:: python

         reader = fluid.layers.io.random_data_generator(
                                          low=0.0,
                                          high=1.0,
                                          shapes=[(3,224,224), (1)],
                                          lod_levels=[0, 0])

         # Via the reader, we can use 'read_file' layer to get data:
         image, label = fluid.layers.io.read_file(reader)
    """
    dtypes = [core.VarDesc.VarType.FP32] * len(shapes)
    shape_concat = []
    ranks = []

    for shape in shapes:
        shape_concat.extend(shape)
        ranks.append(len(shape))

    var_name = unique_name('random_data_generator')

    startup_blk = default_startup_program().current_block()
    startup_var = startup_blk.create_var(name=var_name)
    startup_blk.append_op(
        type='create_random_data_generator',
        outputs={'Out': [startup_var]},
        attrs={
            'low': low,
            'high': high,
            'shape_concat': shape_concat,
            'lod_levels': lod_levels,
            'ranks': ranks
        })

    startup_var.desc.set_dtypes(dtypes)
    startup_var.persistable = True
    main_prog_var = _copy_reader_var_(default_main_program().current_block(),
                                      startup_var)

    if for_parallel:
        main_prog_var = parallel(reader=main_prog_var)

    return monkey_patch_reader_methods(main_prog_var)


434 435 436 437
def open_files(filenames,
               shapes,
               lod_levels,
               dtypes,
Y
yi.wu 已提交
438
               thread_num=1,
F
fengjiayi 已提交
439 440
               buffer_size=None,
               pass_num=1,
F
fengjiayi 已提交
441
               for_parallel=True):
F
fengjiayi 已提交
442 443 444
    """
    Open files

F
fengjiayi 已提交
445 446 447
    This layer takes a list of files to read from and returns a Reader Variable. 
    Via the Reader Variable, we can get data from given files. All files must 
    have name suffixs to indicate their formats, e.g., '*.recordio'. 
F
fengjiayi 已提交
448 449 450 451 452 453 454 455

    Args:
       filenames(list): The list of file names.
       shapes(list): List of tuples which declaring data shapes.
       lod_levels(list): List of ints which declaring data lod_level.
       dtypes(list): List of strs which declaring data type.
       thread_num(int): The maximal concurrent prefetch thread number.
       buffer_size(int): The size of prefetch buffer.
F
fengjiayi 已提交
456
       pass_num(int): Number of passes to run.
F
fengjiayi 已提交
457 458
       for_parallel(Bool): Set it as True if you are going to run 
            subsequent operators in parallel.
F
fengjiayi 已提交
459 460 461 462 463 464 465

    Returns:
       Variable: A Reader Variable via which we can get file data.

    Examples:
       .. code-block:: python

F
fengjiayi 已提交
466
         reader = fluid.layers.io.open_files(filenames=['./data1.recordio',
F
fengjiayi 已提交
467
                                                     './data2.recordio'],
F
fengjiayi 已提交
468 469 470 471 472
                                             shapes=[(3,224,224), (1)],
                                             lod_levels=[0, 0],
                                             dtypes=['float32', 'int64'],
                                             thread_num=2,
                                             buffer_size=2)
F
fengjiayi 已提交
473 474

         # Via the reader, we can use 'read_file' layer to get data:
F
fengjiayi 已提交
475
         image, label = fluid.layers.io.read_file(reader)
F
fengjiayi 已提交
476
    """
477 478
    if buffer_size is None:
        buffer_size = thread_num
F
fengjiayi 已提交
479 480
    if isinstance(filenames, basestring):
        filenames = [filenames]
F
fengjiayi 已提交
481 482 483 484 485 486 487 488
    dtypes = [convert_np_dtype_to_dtype_(dt) for dt in dtypes]
    shape_concat = []
    ranks = []

    for shape in shapes:
        shape_concat.extend(shape)
        ranks.append(len(shape))

F
fengjiayi 已提交
489
    multi_file_reader_name = unique_name('multi_file_reader')
F
fengjiayi 已提交
490
    startup_blk = default_startup_program().current_block()
F
fengjiayi 已提交
491
    startup_reader = startup_blk.create_var(name=multi_file_reader_name)
F
fengjiayi 已提交
492 493
    startup_blk.append_op(
        type='open_files',
F
fengjiayi 已提交
494
        outputs={'Out': [startup_reader]},
F
fengjiayi 已提交
495 496 497 498
        attrs={
            'shape_concat': shape_concat,
            'lod_levels': lod_levels,
            'ranks': ranks,
F
fengjiayi 已提交
499
            'file_names': filenames,
500 501
            'thread_num': thread_num,
            'buffer_size': buffer_size
F
fengjiayi 已提交
502 503
        })

F
fengjiayi 已提交
504 505 506 507 508 509 510
    startup_reader.desc.set_dtypes(dtypes)
    startup_reader.persistable = True
    main_prog_reader = _copy_reader_var_(default_main_program().current_block(),
                                         startup_reader)
    if pass_num > 1:
        main_prog_reader = multi_pass(
            reader=main_prog_reader, pass_num=pass_num)
F
fengjiayi 已提交
511

F
fengjiayi 已提交
512
    if for_parallel:
J
JiayiFeng 已提交
513
        main_prog_reader = parallel(reader=main_prog_reader)
F
fengjiayi 已提交
514

F
fengjiayi 已提交
515 516 517
    return monkey_patch_reader_methods(main_prog_reader)


J
JiayiFeng 已提交
518
def __create_shared_decorated_reader__(op_type, reader, attrs):
Y
Yu Yang 已提交
519 520 521
    var_name = unique_name(op_type)
    startup_blk = default_startup_program().current_block()
    startup_var = startup_blk.create_var(name=var_name)
F
fengjiayi 已提交
522
    startop_op = startup_blk.append_op(
Y
Yu Yang 已提交
523 524 525 526 527
        type=op_type,
        inputs={'UnderlyingReader': reader},
        outputs={'Out': [startup_var]},
        attrs=attrs)
    startup_var.persistable = True
F
fengjiayi 已提交
528 529 530 531
    main_prog_block = default_main_program().current_block()
    main_prog_var = _copy_reader_var_(main_prog_block, startup_var)
    _copy_reader_create_op_(main_prog_block, startop_op)
    return monkey_patch_reader_methods(main_prog_var)
Y
Yu Yang 已提交
532 533


534 535
def __create_unshared_decorated_reader__(op_type, reader, attrs, name=None):
    new_reader_name = name if name is not None else unique_name(op_type)
536 537 538 539 540 541 542 543 544 545
    main_blk = default_main_program().current_block()
    new_reader = main_blk.create_var(name=new_reader_name)
    main_blk.append_op(
        type=op_type,
        inputs={'UnderlyingReader': reader},
        outputs={'Out': [new_reader]},
        attrs=attrs)
    return monkey_patch_reader_methods(new_reader)


F
fengjiayi 已提交
546
def shuffle(reader, buffer_size):
547 548
    return __create_unshared_decorated_reader__(
        'create_shuffle_reader', reader, {'buffer_size': int(buffer_size)})
Y
Yu Yang 已提交
549 550


J
JiayiFeng 已提交
551 552 553 554 555
def batch(reader, batch_size):
    return __create_unshared_decorated_reader__(
        'create_batch_reader', reader, {'batch_size': int(batch_size)})


556
def double_buffer(reader, place=None, name=None):
Y
Yu Yang 已提交
557 558 559
    attrs = dict()
    if place is not None:
        attrs['place'] = str(place).upper()
560 561
    return __create_unshared_decorated_reader__(
        'create_double_buffer_reader', reader, attrs, name=name)
Y
Yu Yang 已提交
562 563


F
fengjiayi 已提交
564
def multi_pass(reader, pass_num):
565 566
    return __create_shared_decorated_reader__(
        'create_multi_pass_reader', reader, {'pass_num': int(pass_num)})
F
fengjiayi 已提交
567 568


J
JiayiFeng 已提交
569
def parallel(reader):
J
JiayiFeng 已提交
570 571
    return __create_shared_decorated_reader__('create_threaded_reader', reader,
                                              {})
F
fengjiayi 已提交
572 573


Y
Yu Yang 已提交
574 575 576 577 578
def read_file(file_obj):
    helper = LayerHelper('read_file')
    out = [
        helper.create_tmp_variable(
            stop_gradient=True, dtype='float32')
Y
Yu Yang 已提交
579
        for _ in range(len(file_obj.desc.shapes()))
Y
Yu Yang 已提交
580 581 582 583 584 585 586
    ]
    helper.append_op(
        type='read', inputs={'Reader': [file_obj]}, outputs={'Out': out})
    if len(out) == 1:
        return out[0]
    else:
        return out
F
fengjiayi 已提交
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631


class Preprocessor(object):
    BEFORE_SUB_BLOCK = 0
    IN_SUB_BLOCK = 1
    AFTER_SUB_BLOCK = 2

    def __init__(self, reader, name=None):
        self.underlying_reader = reader
        new_reader_name = name if name is not None else unique_name(
            "create_custom_reader")
        self.main_prog = default_main_program()
        self.reader = self.main_prog.current_block().create_var(
            name=new_reader_name)
        self.sub_block = None
        self.source_var_names = None
        self.sink_var_names = None
        self.status = Preprocessor.BEFORE_SUB_BLOCK

    def is_completed(self):
        return self.sub_block and self.source_var_names and self.sink_var_names

    @contextlib.contextmanager
    def block(self):
        self.status = Preprocessor.IN_SUB_BLOCK
        self.sub_block = self.main_prog.create_block()
        yield
        self.main_prog.rollback()
        self.status = Preprocessor.AFTER_SUB_BLOCK
        if not self.is_completed():
            raise RuntimeError(
                "The definition of preprocessor is incompleted! "
                "Please make sure that you have set input and output "
                "variables by invoking 'inputs' and 'outputs' in "
                "Preprocessor's sub-block.")

    def inputs(self):
        if self.status != Preprocessor.IN_SUB_BLOCK:
            raise RuntimeError(
                "Preprocessor.inputs() can only be invoked inside the sub-block."
            )

        source_shapes = self.underlying_reader.desc.shapes()
        source_dtypes = self.underlying_reader.desc.dtypes()
        source_lod_levels = self.underlying_reader.desc.lod_levels()
F
fengjiayi 已提交
632 633 634 635
        self.source_var_names = [
            unique_name("preprocessor_source")
            for _ in xrange(len(source_shapes))
        ]
F
fengjiayi 已提交
636
        source_vars = []
F
fengjiayi 已提交
637 638 639
        for var_name, shape, dtype, lod_level in zip(
                self.source_var_names, source_shapes, source_dtypes,
                source_lod_levels):
F
fengjiayi 已提交
640
            source_vars.append(self.main_prog.current_block().create_var(
F
fengjiayi 已提交
641
                name=var_name, shape=shape, dtype=dtype, lod_level=lod_level))
F
fengjiayi 已提交
642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665
        return source_vars

    def outputs(self, *outs):
        if self.status != Preprocessor.IN_SUB_BLOCK:
            raise RuntimeError(
                "Preprocessor.outputs() can only be invoked inside the sub-block."
            )
        self.sink_var_names = [var.name for var in outs]

    def __call__(self, *args, **kwargs):
        if self.status != Preprocessor.AFTER_SUB_BLOCK:
            raise RuntimeError(
                "Preprocessor output can only be retrieved after rnn block.")

        self.main_prog.current_block().append_op(
            type="create_custom_reader",
            inputs={'UnderlyingReader': self.underlying_reader},
            outputs={'Out': [self.reader]},
            attrs={
                "sub_block": self.sub_block,
                "source_var_names": self.source_var_names,
                "sink_var_names": self.sink_var_names
            })
        return monkey_patch_reader_methods(self.reader)
Y
yuyang18 已提交
666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691


@templatedoc()
def load(out, file_path, load_as_fp16=None):
    """
    ${comment}

    >>> import paddle.fluid as fluid
    >>> tmp_tensor = fluid.layers.create_tensor(dtype='float32')
    >>> fluid.layers.load(tmp_tensor, "./tmp_tensor.bin")

    Args:
        out(${out_type}): ${out_comment}.

        file_path(${file_path_type}): ${file_path_comment}.

        load_as_fp16(${load_as_fp16_type}): ${load_as_fp16_comment}.

    Returns:
        None
    """
    helper = LayerHelper("load", **locals())
    attrs = {"file_path": file_path}
    if load_as_fp16 is not None:
        attrs['load_as_fp16'] = load_as_fp16
    helper.append_op(type="load", inputs={}, output={"Out": out}, args=attrs)