wehub-resource-sync
119 lines
3.9 KiB
Python
119 lines
3.9 KiB
Python
# Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
|
|
#
|
|
# 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
|
|
#
|
|
# 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.
|
|
|
|
import os
|
|
import sys
|
|
|
|
import paddle
|
|
|
|
from paddlenlp.metrics import Perplexity
|
|
from paddlenlp.utils import profiler
|
|
|
|
from .model_base import BenchmarkBase
|
|
|
|
sys.path.append(
|
|
os.path.abspath(
|
|
os.path.join(
|
|
os.path.dirname(__file__), os.pardir, os.pardir, os.pardir, os.pardir, "examples", "language_model"
|
|
)
|
|
)
|
|
)
|
|
from rnnlm.model import CrossEntropyLossForLm, RnnLm, UpdateModel # noqa: E402
|
|
from rnnlm.reader import create_data_loader # noqa: E402
|
|
|
|
|
|
class AddProfiler(paddle.callbacks.Callback):
|
|
def on_batch_end(self, mode, step=None, logs=None):
|
|
if mode == "train":
|
|
profiler.add_profiler_step(self.profiler_options)
|
|
|
|
|
|
class RNNLMBenchmark(BenchmarkBase):
|
|
def __init__(self):
|
|
super().__init__()
|
|
|
|
@staticmethod
|
|
def add_args(args, parser):
|
|
parser.add_argument("--hidden_size", type=int, default=650, help="hidden_size")
|
|
parser.add_argument("--num_steps", type=int, default=35, help="num steps")
|
|
parser.add_argument("--num_layers", type=int, default=2, help="num_layers")
|
|
parser.add_argument("--dropout", type=float, default=0.5, help="dropout")
|
|
parser.add_argument("--init_scale", type=float, default=0.05, help="init_scale")
|
|
parser.add_argument("--use_hapi", action="store_false", help="Whether to use hapi to run. ")
|
|
|
|
def create_data_loader(self, args, **kwargs):
|
|
train_loader, valid_loader, test_loader, self.vocab_size = create_data_loader(
|
|
batch_size=args.batch_size, num_steps=args.num_steps
|
|
)
|
|
|
|
self.num_batch = len(train_loader)
|
|
|
|
return train_loader, valid_loader
|
|
|
|
def build_model(self, args, **kwargs):
|
|
network = RnnLm(
|
|
vocab_size=self.vocab_size,
|
|
hidden_size=args.hidden_size,
|
|
batch_size=args.batch_size,
|
|
num_layers=args.num_layers,
|
|
init_scale=args.init_scale,
|
|
dropout=args.dropout,
|
|
)
|
|
|
|
self.cross_entropy = CrossEntropyLossForLm()
|
|
|
|
model = paddle.Model(network)
|
|
|
|
return model
|
|
|
|
def forward(self, model, args, input_data=None, **kwargs):
|
|
ppl_metric = Perplexity()
|
|
callback = UpdateModel()
|
|
|
|
scheduler = paddle.callbacks.LRScheduler(by_step=False, by_epoch=True)
|
|
|
|
model.prepare(optimizer=kwargs.get("optimizer"), loss=self.cross_entropy, metrics=ppl_metric)
|
|
|
|
benchmark_logger = self.logger(args)
|
|
|
|
if args.profiler_options is not None:
|
|
profiler_callback = AddProfiler()
|
|
profiler_callback.profiler_options = args.profiler_options
|
|
callbacks_lists = [callback, scheduler, benchmark_logger, profiler_callback]
|
|
else:
|
|
callbacks_lists = [callback, scheduler, benchmark_logger]
|
|
|
|
model.fit(
|
|
train_data=kwargs.get("train_loader"),
|
|
eval_data=kwargs.get("eval_loader"),
|
|
epochs=args.epoch,
|
|
shuffle=False,
|
|
callbacks=callbacks_lists,
|
|
)
|
|
|
|
def logger(
|
|
self,
|
|
args,
|
|
step_id=None,
|
|
pass_id=None,
|
|
batch_id=None,
|
|
loss=None,
|
|
batch_cost=None,
|
|
reader_cost=None,
|
|
num_samples=None,
|
|
ips=None,
|
|
**kwargs
|
|
):
|
|
return paddle.callbacks.ProgBarLogger(log_freq=(self.num_batch // 10), verbose=3)
|