File size: 3,465 Bytes
1101a21 |
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 |
# coding=utf-8
# Copyright (c) 2020, NVIDIA CORPORATION. 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.
"""Sample Generate GPT"""
import os
import sys
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__),
os.path.pardir)))
import socket
from megatron import get_args
from megatron import print_rank_0
from megatron import mpu
from megatron.checkpointing import load_checkpoint
from megatron.initialize import initialize_megatron
from megatron.model import GPTModel
from megatron.training import get_model
from megatron.text_generation_server import MegatronServer
from megatron.text_generation import generate_and_post_process
from megatron.text_generation import beam_search_and_post_process
import torch
def model_provider(pre_process=True, post_process=True):
"""Build the model."""
print_rank_0('building GPT model ...')
model = GPTModel(num_tokentypes=0, parallel_output=False, pre_process=pre_process, post_process=post_process)
return model
def add_text_generate_args(parser):
group = parser.add_argument_group(title='text generation')
group.add_argument("--temperature", type=float, default=1.0,
help='Sampling temperature.')
group.add_argument("--top_p", type=float, default=0.0,
help='Top p sampling.')
group.add_argument("--top_k", type=int, default=0,
help='Top k sampling.')
group.add_argument("--out-seq-length", type=int, default=1024,
help='Size of the output generated text.')
return parser
if __name__ == "__main__":
initialize_megatron(extra_args_provider=add_text_generate_args,
args_defaults={'tokenizer_type': 'GPT2BPETokenizer',
'no_load_rng': True,
'no_load_optim': True})
args = get_args()
if args.num_layers_per_virtual_pipeline_stage is not None:
print("Interleaved pipeline schedule is not yet supported for text generation.")
exit()
# Set up model and load checkpoint
model = get_model(model_provider, wrap_with_ddp=False)
if args.load is not None:
_ = load_checkpoint(model, None, None)
assert len(model) == 1, "Above condition should have caught this"
model = model[0]
if mpu.is_pipeline_first_stage() and mpu.get_tensor_model_parallel_rank() == 0:
server = MegatronServer(model)
server.run("0.0.0.0")
while True:
choice = torch.cuda.LongTensor(1)
torch.distributed.broadcast(choice, 0)
if choice[0].item() == 0:
try:
generate_and_post_process(model)
except ValueError as ve:
pass
elif choice[0].item() == 1:
try:
beam_search_and_post_process(model)
except ValueError as ve:
pass
|