File size: 3,790 Bytes
d72ec76
 
 
9acfc48
d72ec76
 
 
 
0344d09
d72ec76
9acfc48
d72ec76
9acfc48
d72ec76
 
 
 
 
 
 
 
aa91494
 
d72ec76
 
 
 
aa91494
 
d72ec76
 
 
 
aa91494
d72ec76
 
aa91494
0344d09
 
d72ec76
 
9acfc48
aa91494
0344d09
d72ec76
0344d09
d72ec76
 
0344d09
 
d72ec76
 
 
0344d09
d72ec76
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9acfc48
d72ec76
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
aa91494
d72ec76
 
 
 
 
 
 
9acfc48
aa91494
d72ec76
 
 
 
 
 
 
 
 
 
0344d09
 
 
d72ec76
 
 
 
 
f23b303
9acfc48
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
import streamlit as st
import kornia
import torch
from torch import nn
from torchvision.transforms import functional as F
from torchvision.utils import make_grid
from streamlit_ace import st_ace
from PIL import Image
import numpy as np

IS_LOCAL = False  # Change this

@st.cache_data
def set_transform(content):
    try:
        transform = eval(content, {"kornia": kornia, "nn": nn}, None)
    except Exception as e:
        st.write(f"There was an error: {e}")
        transform = nn.Sequential()
    return transform

st.set_page_config(page_title="Kornia Augmentations Demo", layout="wide")

st.markdown("# Kornia Augmentations Demo")
st.sidebar.markdown(
    "[Kornia](https://github.com/kornia/kornia) is a *differentiable* computer vision library for PyTorch."
)

uploaded_file = st.sidebar.file_uploader("Choose a file", type=['png', 'jpg', 'jpeg'])
if uploaded_file is not None:
    im = Image.open(uploaded_file)
else:
    im = Image.open("./images/pretty_bird.jpg")

scaler = int(im.height / 2)
st.sidebar.image(im, caption="Input Image", width=256)

# Convert PIL Image to torch tensor
image = torch.from_numpy(np.array(im).transpose((2, 0, 1))).float() / 255.0

# batch size is just for show
batch_size = st.sidebar.slider("batch_size", min_value=4, max_value=16, value=8)

gpu = st.sidebar.checkbox("Use GPU!", value=False)
if not gpu:
    st.sidebar.markdown("Using CPU for operations.")
    device = torch.device("cpu")
else:
    if not IS_LOCAL or not torch.cuda.is_available():
        st.sidebar.markdown("GPU not available, using CPU.")
        device = torch.device("cpu")
    else:
        st.sidebar.markdown("Running on GPU~")
        device = torch.device("cuda:0")

predefined_transforms = [
    """
nn.Sequential(
   kornia.augmentation.RandomAffine(degrees=360,p=0.5),
   kornia.augmentation.ColorJitter(brightness=0.2, contrast=0.3, saturation=0.2, hue=0.3, p=1)
)
# p=0.5 is the probability of applying the transformation
""",
    """
nn.Sequential(
   kornia.augmentation.RandomErasing(scale=(.4, .8), ratio=(.3, 1/.3), p=0.5),
)
""",
    """
nn.Sequential(
   kornia.augmentation.RandomErasing(scale=(.4, .8), ratio=(.3, 1/.3), p=1, same_on_batch=True),
)
#By setting same_on_batch=True you can apply the same transform across the batch
""",
    f"""
nn.Sequential(
    kornia.augmentation.RandomResizedCrop(size=({scaler}, {scaler}), scale=(3., 3.), ratio=(2., 2.), p=1.),
    kornia.augmentation.RandomHorizontalFlip(p=0.7),
    kornia.augmentation.RandomGrayscale(p=0.5),
)
"""
]

selected_transform = st.selectbox(
    "Pick an augmentation pipeline example:", predefined_transforms
)

st.write("Transform to apply:")
readonly = False
content = st_ace(
    value=selected_transform,
    height=150,
    language="python",
    keybinding="vscode",
    show_gutter=True,
    show_print_margin=True,
    wrap=False,
    auto_update=False,
    readonly=readonly,
)

if content:
    transform = set_transform(content)

process = st.button("Next Batch")

# Fake dataloader
image_batch = torch.stack(batch_size * [image])
image_batch = image_batch.to(device)

transformeds = None
try:
    transformeds = transform(image_batch)
except Exception as e:
    st.write(f"There was an error: {e}")

cols = st.columns(4)
if transformeds is not None:
    for i, x in enumerate(transformeds):
        i = i % 4
        img_np = x.cpu().numpy().transpose((1, 2, 0))
        img_np = (img_np * 255).astype(np.uint8)
        cols[i].image(img_np, use_column_width=True)

st.markdown(
    "There are a lot more transformations available: [Documentation](https://kornia.readthedocs.io/en/latest/augmentation.module.html)"
)
st.markdown(
    "Kornia can do a lot more than augmentations~ [Check it out](https://kornia.readthedocs.io/en/latest/get-started/introduction.html)"
)