mirror of
https://github.com/hpcaitech/ColossalAI.git
synced 2026-05-04 01:48:43 +00:00
32efe8e740
* [fx] modify the calculation of node_size in MetaInfoProp for activation checkpointing usages * [fx] modify the calculation of node_size in MetaInfoProp for activation checkpointing usages * [fx] modify the calculation of node_size in MetaInfoProp for activation checkpointing usages * [fx] merge development into main (#1) * [fx] activation checkpointing using Chen strategies. * [fx] add test for ckpt_solver_chen * [fx] add vanilla activation checkpoint search with test on resnet and densenet * [fx] add a namespace code for solver_chen. * [fx] fix the false interpretation of algorithm 3 in https://arxiv.org/abs/1604.06174. * [fx] fix lowercase naming conventions. * [fx] simplify test for ckpt. * [fx] add rules to linearize computation graphs for searching. (#2) * [fx] modify the calculation of node_size in MetaInfoProp for activation checkpointing usages * [fx] modify the calculation of node_size in MetaInfoProp for activation checkpointing usages * [fx] modify the calculation of node_size in MetaInfoProp for activation checkpointing usages * [fx] merge development into main (#1) * [fx] activation checkpointing using Chen strategies. * [fx] add test for ckpt_solver_chen * [fx] add vanilla activation checkpoint search with test on resnet and densenet * [fx] add a namespace code for solver_chen. * [fx] fix the false interpretation of algorithm 3 in https://arxiv.org/abs/1604.06174. * [fx] fix lowercase naming conventions. * [fx] simplify test for ckpt. * [fx] fix test and algorithm bugs in activation checkpointing. * [fx] polish ckpt_test. * [fx] add rules to linearize computation graphs for searching. * [fx] remove chen_sqrt for sake of simplicity * [fx] remove chen_sqrt for sake of simplicity * [fx] remove chen_sqrt for sake of simplicity * [fx] remove chen_sqrt for sake of simplicity * [fx] fix inconsistencies. * [fx] fix MetaInfoProp. * [fx] fix MetaInfoProp. * [fx] consider MetaInfoProp for inplace operands. * [fx] consider MetaInfoProp for inplace operands. * [fx] consider MetaInfoProp for inplace operands. * [fx] consider MetaInfoProp for inplace operands. * [fx] consider MetaInfoProp for inplace operands. * [fx] add profiler for fx nodes. * [fx] add profiler for fx nodes. * [fx] add profiler for fx nodes. * [fx] add profiler for fx nodes. * [fx] add profiler for fx nodes. * [fx] add profiler for fx nodes. * [fx] add profiler for fx nodes. * [fx] fix error in tests. * [fx] unfix bug. * [fx] unfix bug.
48 lines
1.4 KiB
Python
48 lines
1.4 KiB
Python
import torch
|
|
import torch.nn as nn
|
|
import colossalai
|
|
import colossalai.nn as col_nn
|
|
from torch.fx import symbolic_trace
|
|
from colossalai.fx.passes.meta_info_prop import MetaInfoProp
|
|
from colossalai.fx.passes.adding_split_node_pass import split_with_split_nodes_pass, uniform_split_pass
|
|
from colossalai.fx.passes.utils import get_comm_size
|
|
import pytest
|
|
|
|
MODEL_DIM = 16
|
|
BATCH_SIZE = 8
|
|
PIPELINE_SIZE = 2
|
|
|
|
|
|
class MLP(torch.nn.Module):
|
|
|
|
def __init__(self, dim: int):
|
|
super().__init__()
|
|
self.linear1 = torch.nn.Linear(dim, dim)
|
|
self.linear2 = torch.nn.Linear(dim, dim)
|
|
self.linear3 = torch.nn.Linear(dim, dim)
|
|
self.linear4 = torch.nn.Linear(dim, dim)
|
|
|
|
def forward(self, x):
|
|
x = self.linear1(x)
|
|
x = self.linear2(x)
|
|
x = self.linear3(x)
|
|
x = self.linear4(x)
|
|
return x
|
|
|
|
|
|
def test_comm_size_compute():
|
|
model = MLP(MODEL_DIM)
|
|
input_sample = torch.rand(BATCH_SIZE, MODEL_DIM, device='meta')
|
|
gm = symbolic_trace(model)
|
|
MetaInfoProp(gm).run(input_sample)
|
|
annotated_model = uniform_split_pass(gm, PIPELINE_SIZE)
|
|
split_model, split_submodules = split_with_split_nodes_pass(annotated_model)
|
|
submodule_list = list(split_model.children())
|
|
comm_size = get_comm_size(submodule_list[0], submodule_list[1])
|
|
# the shape of tensor send from partition 0 to partition 1 is (8, 16)
|
|
assert comm_size == 128
|
|
|
|
|
|
if __name__ == '__main__':
|
|
test_comm_size_compute()
|