mirror of https://github.com/hpcaitech/ColossalAI
aibig-modeldata-parallelismdeep-learningdistributed-computingfoundation-modelsheterogeneous-traininghpcinferencelarge-scalemodel-parallelismpipeline-parallelism
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
50 lines
1.4 KiB
50 lines
1.4 KiB
3 years ago
|
#!/usr/bin/env python
|
||
|
|
||
|
import torch
|
||
|
import torch.nn as nn
|
||
2 years ago
|
|
||
3 years ago
|
from colossalai.nn import CheckpointModule
|
||
2 years ago
|
|
||
3 years ago
|
from .registry import non_distributed_component_funcs
|
||
2 years ago
|
from .utils.dummy_data_generator import DummyDataGenerator
|
||
3 years ago
|
|
||
|
|
||
|
class NetWithRepeatedlyComputedLayers(CheckpointModule):
|
||
|
"""
|
||
|
This model is to test with layers which go through forward pass multiple times.
|
||
|
In this model, the fc1 and fc2 call forward twice
|
||
|
"""
|
||
|
|
||
|
def __init__(self, checkpoint=False) -> None:
|
||
|
super().__init__(checkpoint=checkpoint)
|
||
|
self.fc1 = nn.Linear(5, 5)
|
||
|
self.fc2 = nn.Linear(5, 5)
|
||
|
self.fc3 = nn.Linear(5, 2)
|
||
|
self.layers = [self.fc1, self.fc2, self.fc1, self.fc2, self.fc3]
|
||
|
|
||
|
def forward(self, x):
|
||
|
for layer in self.layers:
|
||
|
x = layer(x)
|
||
|
return x
|
||
|
|
||
|
|
||
|
class DummyDataLoader(DummyDataGenerator):
|
||
|
|
||
|
def generate(self):
|
||
|
data = torch.rand(16, 5)
|
||
|
label = torch.randint(low=0, high=2, size=(16,))
|
||
|
return data, label
|
||
|
|
||
|
|
||
|
@non_distributed_component_funcs.register(name='repeated_computed_layers')
|
||
|
def get_training_components():
|
||
3 years ago
|
|
||
2 years ago
|
def model_builder(checkpoint=False):
|
||
3 years ago
|
return NetWithRepeatedlyComputedLayers(checkpoint)
|
||
|
|
||
3 years ago
|
trainloader = DummyDataLoader()
|
||
|
testloader = DummyDataLoader()
|
||
3 years ago
|
|
||
3 years ago
|
criterion = torch.nn.CrossEntropyLoss()
|
||
3 years ago
|
return model_builder, trainloader, testloader, torch.optim.Adam, criterion
|