Sarah Yakum commited on
Commit
11aec48
1 Parent(s): b7bc15f
Files changed (2) hide show
  1. .bashrc +2 -0
  2. app.py +80 -13
.bashrc ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ export PATH="$PATH:/c/Users/sarah/anaconda3:/c/Users/sarah/anaconda3/Scripts"
2
+ alias python="winpty python.exe"
app.py CHANGED
@@ -1,22 +1,89 @@
1
  import streamlit as st
2
 
3
- import streamlit as st
4
- from transformers import pipeline
5
- from PIL import Image
 
 
 
 
6
 
7
- pipeline = pipeline(task="image-classification", model="julien-c/hotdog-not-hotdog")
8
 
9
- st.title("Hot Dog? Or Not?")
 
 
 
10
 
11
- file_name = st.file_uploader("Upload a hot dog candidate image")
12
 
13
  if file_name is not None:
14
- col1, col2 = st.columns(2)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15
 
16
- image = Image.open(file_name)
17
- col1.image(image, use_column_width=True)
18
- predictions = pipeline(image)
19
 
20
- col2.header("Probabilities")
21
- for p in predictions:
22
- col2.subheader(f"{ p['label'] }: { round(p['score'] * 100, 1)}%")
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  import streamlit as st
2
 
3
+ st.title("Segmentation of Beauty Products")
4
+
5
+ file_name = st.file_uploader("Upload a a beauty product")
6
+
7
+
8
+ from transformers import SegformerImageProcessor, SegformerForSemanticSegmentation
9
+ import torch
10
 
11
+ device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
12
 
13
+ model_name = "nvidia/segformer-b5-finetuned-ade-640-640"
14
+ feature_extractor = SegformerImageProcessor.from_pretrained(model_name)
15
+ model = SegformerForSemanticSegmentation.from_pretrained(model_name)
16
+ model.to(device)
17
 
 
18
 
19
  if file_name is not None:
20
+ image = Image.open(img_name)
21
+ image.show()
22
+
23
+ pixel_values = feature_extractor(image, return_tensors="pt").pixel_values.to(device)
24
+
25
+ outputs = model(pixel_values)
26
+ logits = outputs.logits
27
+
28
+ def ade_palette():
29
+ return [[120, 120, 120], [180, 120, 120], [6, 230, 230], [80, 50, 50],
30
+ [4, 200, 3], [120, 120, 80], [140, 140, 140], [204, 5, 255],
31
+ [230, 230, 230], [4, 250, 7], [224, 5, 255], [235, 255, 7],
32
+ [150, 5, 61], [120, 120, 70], [8, 255, 51], [255, 6, 82],
33
+ [143, 255, 140], [204, 255, 4], [255, 51, 7], [204, 70, 3],
34
+ [0, 102, 200], [61, 230, 250], [255, 6, 51], [11, 102, 255],
35
+ [255, 7, 71], [255, 9, 224], [9, 7, 230], [220, 220, 220],
36
+ [255, 9, 92], [112, 9, 255], [8, 255, 214], [7, 255, 224],
37
+ [255, 184, 6], [10, 255, 71], [255, 41, 10], [7, 255, 255],
38
+ [224, 255, 8], [102, 8, 255], [255, 61, 6], [255, 194, 7],
39
+ [255, 122, 8], [0, 255, 20], [255, 8, 41], [255, 5, 153],
40
+ [6, 51, 255], [235, 12, 255], [160, 150, 20], [0, 163, 255],
41
+ [140, 140, 140], [250, 10, 15], [20, 255, 0], [31, 255, 0],
42
+ [255, 31, 0], [255, 224, 0], [153, 255, 0], [0, 0, 255],
43
+ [255, 71, 0], [0, 235, 255], [0, 173, 255], [31, 0, 255],
44
+ [11, 200, 200], [255, 82, 0], [0, 255, 245], [0, 61, 255],
45
+ [0, 255, 112], [0, 255, 133], [255, 0, 0], [255, 163, 0],
46
+ [255, 102, 0], [194, 255, 0], [0, 143, 255], [51, 255, 0],
47
+ [0, 82, 255], [0, 255, 41], [0, 255, 173], [10, 0, 255],
48
+ [173, 255, 0], [0, 255, 153], [255, 92, 0], [255, 0, 255],
49
+ [255, 0, 245], [255, 0, 102], [255, 173, 0], [255, 0, 20],
50
+ [255, 184, 184], [0, 31, 255], [0, 255, 61], [0, 71, 255],
51
+ [255, 0, 204], [0, 255, 194], [0, 255, 82], [0, 10, 255],
52
+ [0, 112, 255], [51, 0, 255], [0, 194, 255], [0, 122, 255],
53
+ [0, 255, 163], [255, 153, 0], [0, 255, 10], [255, 112, 0],
54
+ [143, 255, 0], [82, 0, 255], [163, 255, 0], [255, 235, 0],
55
+ [8, 184, 170], [133, 0, 255], [0, 255, 92], [184, 0, 255],
56
+ [255, 0, 31], [0, 184, 255], [0, 214, 255], [255, 0, 112],
57
+ [92, 255, 0], [0, 224, 255], [112, 224, 255], [70, 184, 160],
58
+ [163, 0, 255], [153, 0, 255], [71, 255, 0], [255, 0, 163],
59
+ [255, 204, 0], [255, 0, 143], [0, 255, 235], [133, 255, 0],
60
+ [255, 0, 235], [245, 0, 255], [255, 0, 122], [255, 245, 0],
61
+ [10, 190, 212], [214, 255, 0], [0, 204, 255], [20, 0, 255],
62
+ [255, 255, 0], [0, 153, 255], [0, 41, 255], [0, 255, 204],
63
+ [41, 0, 255], [41, 255, 0], [173, 0, 255], [0, 245, 255],
64
+ [71, 0, 255], [122, 0, 255], [0, 255, 184], [0, 92, 255],
65
+ [184, 255, 0], [0, 133, 255], [255, 214, 0], [25, 194, 194],
66
+ [102, 255, 0], [92, 0, 255]]
67
 
68
+ from torch import nn
69
+ import numpy as np
70
+ import matplotlib.pyplot as plt
71
 
72
+ logits = nn.functional.interpolate(outputs.logits.detach().cpu(),
73
+ size=image.size[::-1],
74
+ mode='bilinear',
75
+ align_corners=False)
76
+ seg = logits.argmax(dim=1)[0]
77
+ color_seg = np.zeros((seg.shape[0], seg.shape[1], 3), dtype=np.uint8)
78
+ palette = np.array(ade_palette())
79
+ for label, color in enumerate(palette):
80
+ color_seg[seg == label, :] = color
81
+
82
+ color_seg = color_seg[..., ::-1]
83
+
84
+ img = np.array(image) * 0.5 + color_seg * 0.5
85
+ img = img.astype(np.uint8)
86
+
87
+ plt.figure(figsize=(15, 10))
88
+ plt.imshow(img)
89
+ plt.show()