Datasets:

Languages:
English
ArXiv:
Tags:
code
License:
melisa commited on
Commit
4263fe9
1 Parent(s): 66f0bfe

Upload 7 files

Browse files
Files changed (7) hide show
  1. detact_color.zip +3 -0
  2. detact_icon.zip +3 -0
  3. ocr.zip +3 -0
  4. split.py +75 -0
  5. test.json +0 -0
  6. train.json +0 -0
  7. val.json +0 -0
detact_color.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94ca7c177549c93d9d858748c85c360b029e19cb2c51143ee80c7575fe3632c0
3
+ size 65100
detact_icon.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23bc186a65c405d0ac8d3a84749ac5aa770980920300f9391b98fd2e774a5df3
3
+ size 67139
ocr.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74cbf1180c2263f34a723de1da93bd4d07dae8d3e0804a9c3cf00e608f3cad0a
3
+ size 200305
split.py ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from glob import glob
2
+ import os
3
+ import json
4
+ import random
5
+
6
+ train_cnt = 0
7
+ test_cnt = 0
8
+ val_cnt = 0
9
+ combined_cnt = 0
10
+
11
+ train_final = {}
12
+ test_final = {}
13
+ val_final = {}
14
+ combined_final = {}
15
+ for fn in glob("data/tasks/*/*"):
16
+ print(fn)
17
+ app_name = fn.split('/')[3]
18
+ app_type = fn.split('/')[2]
19
+ for i in range(1,10):
20
+ image = f"data/data/{app_type}/{app_name}/screen_{i}.png"
21
+ if os.path.exists(image) == False:
22
+ if i == 1:
23
+ print(image)
24
+ continue
25
+ ocr = f"ocr/{app_type}/{app_name}/screen_{i}.json"
26
+ assert os.path.exists(ocr), ocr
27
+ color = f"detact_color/{app_type}/{app_name}/screen_{i}.json"
28
+ assert os.path.exists(color), color
29
+ icon = f"detact_icon/{app_type}/{app_name}/screen_{i}.json"
30
+ assert os.path.exists(icon), icon
31
+ box = f"data/metadata/{app_type}/boxes/{app_name}/screen_{i}.json"
32
+ assert os.path.exists(box), box
33
+ script_to_task_fn = {}
34
+ cnt = 0
35
+ for task_fn in glob(f"{fn}/task_{i}.*.txt"):
36
+ txt = '\n'.join(open(task_fn).read().strip().split('\n')[1:])
37
+ if txt not in script_to_task_fn:
38
+ script_to_task_fn[txt] = []
39
+ script_to_task_fn[txt].append(task_fn)
40
+ cnt += 1
41
+
42
+
43
+ for k in script_to_task_fn:
44
+ c_tasks = script_to_task_fn[k]
45
+ rand = random.randint(1,10)
46
+ if rand <= 7:
47
+ for c in c_tasks:
48
+ train_final[train_cnt] = {"task": c, "image": image, "ocr": ocr, "color": color, "icon": icon, "box": box}
49
+ train_cnt += 1
50
+ combined_final[combined_cnt] = {"task": c, "image": image, "ocr": ocr, "color": color, "icon": icon, "box": box}
51
+ combined_cnt += 1
52
+ elif rand == 8:
53
+ for c in c_tasks:
54
+ val_final[val_cnt] = {"task": c, "image": image, "ocr": ocr, "color": color, "icon": icon, "box": box}
55
+ val_cnt += 1
56
+ combined_final[combined_cnt] = {"task": c, "image": image, "ocr": ocr, "color": color, "icon": icon, "box": box}
57
+ combined_cnt += 1
58
+ elif rand > 8:
59
+ for c in c_tasks:
60
+ test_final[test_cnt] = {"task": c, "image": image, "ocr": ocr, "color": color, "icon": icon, "box": box}
61
+ test_cnt += 1
62
+ combined_final[combined_cnt] = {"task": c, "image": image, "ocr": ocr, "color": color, "icon": icon, "box": box}
63
+ combined_cnt += 1
64
+
65
+
66
+
67
+ print(train_cnt)
68
+ print(val_cnt)
69
+ print(test_cnt)
70
+ print(combined_cnt)
71
+
72
+ json.dump(train_final, open('train.json', "w"), indent=4)
73
+ json.dump(val_final, open("val.json","w"), indent=4)
74
+ json.dump(test_final, open("test.json","w"), indent=4)
75
+ json.dump(combined_final, open('combined.json','w'), indent=4)
test.json ADDED
The diff for this file is too large to render. See raw diff
 
train.json ADDED
The diff for this file is too large to render. See raw diff
 
val.json ADDED
The diff for this file is too large to render. See raw diff