rnn.py

# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you 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
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# 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.

from __future__ import print_function
from six.moves import range

import argparse
import subprocess
from itertools import product
from time import time

import mxnet as mx
import numpy as onp
from mxnet import gluon, np, npx


_parser = argparse.ArgumentParser(description='Benchmark foreach and while_loop on RNN tasks.')
_parser.add_argument('--benchmark', choices=["foreach", "while_loop"], required=True)
_parser.add_argument('--warmup_rounds', type=int, default=20)
_parser.add_argument('--test_rounds', type=int, default=100)
_parser.add_argument('--gpu', type=bool, default=False)
args = _parser.parse_args()


class ForeachRNN(gluon.HybridBlock):
    def __init__(self, cell, length):
        super(ForeachRNN, self).__init__()
        self.length = length
        self.cell = cell

    def forward(self, inputs, states):
        out, states = npx.foreach(self.cell, inputs, states)
        return out


class WhileRNN(gluon.HybridBlock):
    def __init__(self, cell, length):
        super(WhileRNN, self).__init__()
        self.length = length
        self.cell = cell

    def forward(self, inputs, states):
        def _func(*states):
            i = states[0]
            s = states[1: ]
            data = np.squeeze(np.take(inputs, i), axis=0)
            out, new_s = self.cell(data, s)
            new_s = [i + 1] + new_s
            return out, new_s
        out, states = npx.while_loop(
            cond=lambda i, *_: i < self.length,
            func=_func,
            loop_vars=states,
            max_iterations=self.length,
        )
        return out


def _zeros(shape, ctx):
    return mx.np.zeros(shape=shape, ctx=ctx)


def _array(shape, ctx):
    return mx.np.random.normal(loc=0.0, scale=1.0, size=shape, ctx=ctx)


def _get_gpus():
    return range(mx.util.get_gpu_count())

def run_benchmark(cell_type, ctx, seq_len, batch_size, hidden_dim):
    obj = {"foreach": ForeachRNN, "while_loop": WhileRNN}[args.benchmark]
    inputs = _array((seq_len, batch_size, hidden_dim), ctx)
    states = [_array((batch_size, hidden_dim), ctx) for _ in cell_type(0).state_info()]
    if args.benchmark == "while_loop":
        states.insert(0, _zeros((1, ), ctx))

    for is_train, is_hyb_cell, is_hyb_layer in product([True, False], [False, True], [False, True]):
        cell = cell_type(hidden_dim)
        cell.infer_shape(0, inputs, False)
        if is_hyb_cell:
            cell.hybridize(static_alloc=True)
        layer = obj(cell, seq_len)
        layer.initialize(ctx=ctx)
        if is_hyb_layer:
            layer.hybridize(static_alloc=True)
        print(
            f"is_train = {repr(is_train)}, hybridize_cell = {repr(is_hyb_cell)}, hybridize_layer = {repr(is_hyb_layer)}")
        times = []
        for _ in range(args.warmup_rounds + args.test_rounds):
            tick = time()
            if not is_train:
                res = layer(inputs, states)
            else:
                with mx.autograd.record():
                    res = layer(inputs, states)
            if is_train:
                res.backward()
            mx.npx.waitall()
            tock = time()
            times.append((tock - tick) * 1000.0)
        times = times[args.warmup_rounds: ]
        print(f"Time used: mean = {onp.mean(times):.3f} ms, std = {onp.std(times):.3f} ms")


def main():
    # testing configurations
    cell_types = [gluon.rnn.RNNCell,
                  gluon.rnn.GRUCell,
                  gluon.rnn.LSTMCell]
    ctxs = [mx.cpu(0)]
    if args.gpu:
        ctxs = ctxs + [mx.gpu(i) for i in _get_gpus()]
    seq_lens = [100]
    batch_sizes = [1, 32]
    hidden_dims = [512]
    print("--------------------------------------")
    print("Benchmarking", args.benchmark)
    for cell_type, ctx, seq_len, batch_size, hidden_dim in product(  \
        cell_types, ctxs, seq_lens, batch_sizes, hidden_dims):
        print("--------------------------------------")
        print(f"cell: {cell_type.__name__}  ctx: {str(ctx)}  length: {seq_len}  batch size: {batch_size} dim: {hidden_dim}")
        run_benchmark(cell_type, ctx, seq_len, batch_size, hidden_dim)


if __name__ == "__main__":
    main()