# Copyright 2022 Cerebras Systems.
#
# 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 torchvision
from torch.utils.data.dataloader import default_collate
import cerebras.pytorch.distributed as dist
from cerebras.modelzoo.common.registry import registry
from cerebras.modelzoo.data.vision.diffusion.DiffusionBaseProcessor import (
DiffusionBaseProcessor,
)
from cerebras.modelzoo.data.vision.utils import create_worker_cache
# NOTE: Inorder to use this dataloader,
# model side changes as follows are needed
# 1. initializing VAEModel.
# 2. Using `vae_noise` to create latent from forward pass of VAEEncoder
[docs]@registry.register_datasetprocessor("DiffusionImageNet1KProcessor")
class DiffusionImageNet1KProcessor(DiffusionBaseProcessor):
[docs] def __init__(self, params):
super().__init__(params)
self.num_classes = 1000
def create_dataset(self, use_training_transforms=True, split="train"):
if self.use_worker_cache and dist.is_streamer():
self.data_dir = create_worker_cache(self.data_dir)
self.check_split_valid(split)
transform, target_transform = self.process_transform()
if not os.path.isfile(os.path.join(self.data_dir, "meta.bin")):
raise RuntimeError(
"The meta file meta.bin is not present in the root directory. "
"Check data/vision/classification/data/README.md for "
"more details on downloading the dataset."
)
if not os.path.isdir(os.path.join(self.data_dir, split)):
raise RuntimeError(
f"No directory {split} under root dir. Refer to "
"data/vision/classification/data/README.md on how to "
"prepare the dataset."
)
dataset = torchvision.datasets.ImageNet(
root=self.data_dir,
split=split,
transform=transform,
target_transform=target_transform,
)
return dataset
def _custom_collate_fn(self, batch):
batch = default_collate(batch)
input, label = batch
data = self.noise_generator(*self.label_dropout(input, label))
return data
[docs] def create_dataloader(self, dataset, is_training=False):
dataloader = super().create_dataloader(dataset, is_training)
self.latent_dist_fn = self._passthrough
dataloader.collate_fn = self._custom_collate_fn
return dataloader
if __name__ == "__main__":
import os
import numpy as np
import torch
import yaml
from cerebras.modelzoo.models.vision.dit.utils import set_defaults
fpath = os.path.abspath(
os.path.join(
os.path.dirname(__file__),
"../configs/params_dit_small_patchsize_2x2.yaml",
)
)
with open(fpath, "r") as fid:
params = yaml.safe_load(fid)
params = set_defaults(params)
data_obj = DiffusionImageNet1KProcessor(params['train_input'])
dataset = data_obj.create_dataset(
use_training_transforms=True, split=params["train_input"]["split"]
)
print("Dataset features: \n")
sample = dataset[0]
print(
f"Dataset 0th sample: {sample}, {sample[0].shape}, unique vals: img: {torch.unique(sample[0])}, label:{np.unique(sample[1])}"
)
dataloader = data_obj.create_dataloader(dataset, is_training=True)
for ii, data in enumerate(dataloader):
if ii == 1:
break
for k, v in data.items():
print(f"{k} -- {v.shape}, {v.dtype}")
print("----")