File size: 5,763 Bytes
ed7a497 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 |
# Copyright 2020 The HuggingFace Team. 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 sys
from typing import Dict
from transformers import EvalPrediction, HfArgumentParser, TrainingArguments, is_torch_available
from transformers.testing_utils import (
TestCasePlus,
execute_subprocess_async,
get_torch_dist_unique_port,
require_torch_multi_gpu,
require_torch_neuroncore,
)
from transformers.utils import logging
logger = logging.get_logger(__name__)
if is_torch_available():
import torch
from torch import nn
from torch.utils.data import Dataset
from transformers import Trainer
class DummyDataset(Dataset):
def __init__(self, length: int = 101):
self.length = length
def __len__(self):
return self.length
def __getitem__(self, i) -> int:
return i
class DummyDataCollator:
def __call__(self, features):
return {"input_ids": torch.tensor(features), "labels": torch.tensor(features)}
class DummyModel(nn.Module):
def __init__(self):
super().__init__()
# Add some (unused) params otherwise DDP will complain.
self.fc = nn.Linear(120, 80)
def forward(self, input_ids, labels=None):
if labels is not None:
return torch.tensor(0.0, device=input_ids.device), input_ids
else:
return input_ids
class TestTrainerDistributedNeuronCore(TestCasePlus):
@require_torch_neuroncore
def test_trainer(self):
distributed_args = f"""
-m torch.distributed.launch
--nproc_per_node=2
--master_port={get_torch_dist_unique_port()}
{self.test_file_dir}/test_trainer_distributed.py
""".split()
output_dir = self.get_auto_remove_tmp_dir()
args = f"--output_dir {output_dir}".split()
cmd = [sys.executable] + distributed_args + args
execute_subprocess_async(cmd, env=self.get_env())
# successful return here == success - any errors would have caused an error in the sub-call
class TestTrainerDistributed(TestCasePlus):
@require_torch_multi_gpu
def test_trainer(self):
distributed_args = f"""
-m torch.distributed.launch
--nproc_per_node={torch.cuda.device_count()}
--master_port={get_torch_dist_unique_port()}
{self.test_file_dir}/test_trainer_distributed.py
""".split()
output_dir = self.get_auto_remove_tmp_dir()
args = f"--output_dir {output_dir}".split()
cmd = [sys.executable] + distributed_args + args
execute_subprocess_async(cmd, env=self.get_env())
# successful return here == success - any errors would have caused an error in the sub-call
if __name__ == "__main__":
# The script below is meant to be run under torch.distributed, on a machine with multiple GPUs:
#
# PYTHONPATH="src" python -m torch.distributed.launch --nproc_per_node 2 --output_dir output_dir ./tests/test_trainer_distributed.py
parser = HfArgumentParser((TrainingArguments,))
training_args = parser.parse_args_into_dataclasses()[0]
logger.warning(
f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}, "
f"distributed training: {training_args.local_rank != -1}"
)
# Essentially, what we want to verify in the distributed case is that we get all samples back,
# in the right order. (this is crucial for prediction for instance)
for dataset_length in [101, 40, 7]:
dataset = DummyDataset(dataset_length)
def compute_metrics(p: EvalPrediction) -> Dict:
sequential = list(range(len(dataset)))
success = p.predictions.tolist() == sequential and p.label_ids.tolist() == sequential
if not success and training_args.local_rank == 0:
logger.warning(
"Predictions and/or labels do not match expected results:\n - predictions: "
f"{p.predictions.tolist()}\n - labels: {p.label_ids.tolist()}\n - expected: {sequential}"
)
return {"success": success}
trainer = Trainer(
model=DummyModel(),
args=training_args,
data_collator=DummyDataCollator(),
eval_dataset=dataset,
compute_metrics=compute_metrics,
)
metrics = trainer.evaluate()
logger.info(metrics)
if metrics["eval_success"] is not True:
logger.error(metrics)
exit(1)
p = trainer.predict(dataset)
logger.info(p.metrics)
if p.metrics["test_success"] is not True:
logger.error(p.metrics)
exit(1)
trainer.args.eval_accumulation_steps = 2
metrics = trainer.evaluate()
logger.info(metrics)
if metrics["eval_success"] is not True:
logger.error(metrics)
exit(1)
p = trainer.predict(dataset)
logger.info(p.metrics)
if p.metrics["test_success"] is not True:
logger.error(p.metrics)
exit(1)
trainer.args.eval_accumulation_steps = None
|