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
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186# Image Processing locally - `for` loop vs `mapping`
## Mapping
```python
"""
This will process 1,000 images concurrently and will generate thumnails for them.
It took me on my local machine running with local Orion instance 66 seconds, same for mapped/for loop
Best to run it on local Orion to avoid issues with Cloud rate limits.
unzip cats.zip
ls cats | wc -l
python flows/10_image_processing/thumbnails.py
ls cats/thumbnails | wc -l
"""
from pathlib import Path, PosixPath
from PIL import Image
from prefect import task, flow, unmapped
from typing import Tuple
@task
def get_images(img_dir: PosixPath, extension: str = "png"):
return [i for i in img_dir.glob(f"*.{extension}")]
@task
def process_image(
infile: PosixPath,
out_dir: PosixPath,
size: Tuple[int, int] = (128, 128),
extension: str = "png",
):
with Image.open(infile) as im:
im.thumbnail(size)
im.save(Path(out_dir, infile.stem + f"-thumbnail.{extension}"))
@flow
def generate_thumbnails(
in_dir: str = "small", extension: str = "png", size: Tuple[int, int] = (128, 128)
):
img_dir = Path(".", in_dir)
out_dir = Path(".", in_dir, "thumbnails_mapped")
images = get_images.submit(img_dir, extension)
Path(out_dir).mkdir(parents=True, exist_ok=True)
process_image.map(
images.result(), unmapped(out_dir), unmapped(size), unmapped(extension)
)
if __name__ == "__main__":
generate_thumbnails(in_dir="cats", extension="jpg")
# generate_thumbnails(in_dir="small", extension="png")
```
## For loop
```python
"""
This will process 10,000 images concurrently and will generate thumnails for them.
It took me on my local machine running with local Orion instance
Best to run it on local Orion to avoid issues with Cloud rate limits.
ls train | wc -l
ls train/thumbnails | wc -l
"""
from pathlib import Path, PosixPath
from PIL import Image
from prefect import task, flow
from typing import Tuple
@task
def get_images(img_dir: PosixPath, extension: str = "png"):
return [i for i in img_dir.glob(f"*.{extension}")]
@task
def process_image(
infile: PosixPath,
out_dir: PosixPath,
size: Tuple[int, int] = (128, 128),
extension: str = "png",
):
with Image.open(infile) as im:
im.thumbnail(size)
im.save(Path(out_dir, infile.stem + f"-thumbnail.{extension}"))
@flow
def generate_thumbnails(
in_dir: str = "small", extension: str = "png", size: Tuple[int, int] = (128, 128)
):
img_dir = Path(".", in_dir)
out_dir = Path(".", in_dir, "thumbnails")
images = get_images.submit(img_dir, extension)
Path(out_dir).mkdir(parents=True, exist_ok=True)
for img in images.result():
img_ = str(img).replace(f"{in_dir}/", "")
process_image.with_options(name=img_).submit(img, out_dir, size, extension)
if __name__ == "__main__":
generate_thumbnails(in_dir="cats", extension="jpg")
```
## For loop with S3 images
```python
from dotenv import load_dotenv
import os
from prefect.filesystems import S3
from utilities.deploy_utils import DEFAULT_BLOCK, save_block
load_dotenv()
s3 = S3(
bucket_path=os.environ.get("AWS_S3_BUCKET_NAME", DEFAULT_BLOCK),
aws_access_key_id=os.environ.get("AWS_ACCESS_KEY_ID", DEFAULT_BLOCK),
aws_secret_access_key=os.environ.get("AWS_SECRET_ACCESS_KEY", DEFAULT_BLOCK),
)
```
```python
"""
pip install s3fs
pip install prefect-aws
pip install aiobotocore==2.3.4
aws s3 ls s3://prefect-orion/demo/
unzip cats.zip
"""
from prefect.filesystems import S3
s3 = S3.load("default")
s3.put_directory("cats", "cats")
# in a flow:
# s3.get_directory(from_path="cats", local_path="cats_local")
```
slow upload process file by file - easier to upload zip file with 1k images and unzip before processing & before putting to mapping
```python
"""
This will process 10,000 images concurrently and will generate thumnails for them.
It took me on my local machine running with local Orion instance
Best to run it on local Orion to avoid issues with Cloud rate limits.
unzip cats.zip
ls cats | wc -l
python flows/10_image_processing/thumbnails_for_loop_s3_images.py
ls cats/thumbnails | wc -l
"""
from pathlib import Path, PosixPath
from PIL import Image
from prefect import task, flow
from typing import Tuple
from prefect.filesystems import S3
@task
def get_images(img_dir: PosixPath, extension: str = "png"):
return [i for i in img_dir.glob(f"*.{extension}")]
@task
def process_image(
infile: PosixPath,
out_dir: PosixPath,
size: Tuple[int, int] = (128, 128),
extension: str = "png",
):
with Image.open(infile) as im:
im.thumbnail(size)
im.save(Path(out_dir, infile.stem + f"-thumbnail.{extension}"))
@flow
def generate_thumbnails(
in_dir: str = "small", extension: str = "png", size: Tuple[int, int] = (128, 128)
):
img_dir = Path(".", in_dir)
out_dir = Path(".", in_dir, "thumbnails")
s3 = S3.load("default")
s3.get_directory(from_path=in_dir, local_path=in_dir)
images = get_images.submit(img_dir, extension)
Path(out_dir).mkdir(parents=True, exist_ok=True)
for img in images.result():
img_ = str(img).replace(f"{in_dir}/", "")
process_image.with_options(name=img_).submit(img, out_dir, size, extension)
if __name__ == "__main__":
generate_thumbnails(in_dir="cats", extension="jpg")
```