LLDDSS commited on
Commit
ba00758
·
verified ·
1 Parent(s): 36408cd

Upload folder using huggingface_hub

Browse files
.DS_Store ADDED
Binary file (8.2 kB). View file
 
.gitignore ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ # Ignore raw scene folders
2
+ Real/
3
+ Hypothetical/
4
+ Multi_View/
.vscode/settings.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "python.analysis.autoImportCompletions": true
3
+ }
Causal3D.py ADDED
@@ -0,0 +1,163 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import datasets
2
+ import pandas as pd
3
+ import os
4
+ from pathlib import Path
5
+ from tqdm import tqdm
6
+
7
+ _CITATION = """\
8
+ @article{liu2025causal3d,
9
+ title={CAUSAL3D: A Comprehensive Benchmark for Causal Learning from Visual Data},
10
+ author={Liu, Disheng and Qiao, Yiran and Liu, Wuche and Lu, Yiren and Zhou, Yunlai and Liang, Tuo and Yin, Yu and Ma, Jing},
11
+ journal={arXiv preprint arXiv:2503.04852},
12
+ year={2025}
13
+ }
14
+ """
15
+
16
+ _DESCRIPTION = """\
17
+ Causal3D is a benchmark for evaluating causal reasoning in physical and hypothetical visual scenes.
18
+ It includes both real-world recordings and rendered synthetic scenes demonstrating causal interactions.
19
+ """
20
+
21
+ _HOMEPAGE = "https://huggingface.co/datasets/LLDDSS/Causal3D"
22
+ _LICENSE = "CC-BY-4.0"
23
+
24
+ class Causal3D(datasets.GeneratorBasedBuilder):
25
+ DEFAULT_CONFIG_NAME = "Real_Water_flow"
26
+ BUILDER_CONFIGS = [
27
+ # hypothetical_scenes
28
+ datasets.BuilderConfig(name="Hypothetical_V2_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V2_linear scene"),
29
+ datasets.BuilderConfig(name="Hypothetical_V2_nonlinear", version=datasets.Version("1.0.0"), description="Hypothetical_V2_nonlinear scene"),
30
+ datasets.BuilderConfig(name="Hypothetical_V3_fully_connected_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V3_fully_connected_linear scene"),
31
+ datasets.BuilderConfig(name="Hypothetical_V3_v_structure_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V3_v_structure_linear scene"),
32
+ datasets.BuilderConfig(name="Hypothetical_V3_v_structure_nonlinear", version=datasets.Version("1.0.0"), description="Hypothetical_V3_v_structure_nonlinear scene"),
33
+ datasets.BuilderConfig(name="Hypothetical_V4_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V4_linear scene"),
34
+ datasets.BuilderConfig(name="Hypothetical_V4_v_structure_nonlinear", version=datasets.Version("1.0.0"), description="Hypothetical_V4_v_structure_nonlinear scene"),
35
+ datasets.BuilderConfig(name="Hypothetical_V4_v_structure_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V4_v_structure_linear scene"),
36
+ datasets.BuilderConfig(name="Hypothetical_V5_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V5_linear scene"),
37
+ datasets.BuilderConfig(name="Hypothetical_V5_v_structure_linear", version=datasets.Version("1.0.0"), description="Hypothetical_V5_v_structure_linear scene"),
38
+ datasets.BuilderConfig(name="Hypothetical_V5_v_structure_nonlinear", version=datasets.Version("1.0.0"), description="Hypothetical_V5_v_structure_nonlinear scene"),
39
+
40
+ # real_scenes
41
+ datasets.BuilderConfig(name="Real_Parabola", version=datasets.Version("1.0.0"), description="Real_Parabola scene"),
42
+ datasets.BuilderConfig(name="Real_Magnet", version=datasets.Version("1.0.0"), description="Real_Magnet scene"),
43
+ datasets.BuilderConfig(name="Real_Spring", version=datasets.Version("1.0.0"), description="Real_Spring scene"),
44
+ datasets.BuilderConfig(name="Real_Water_flow", version=datasets.Version("1.0.0"), description="Real_Water_flow scene"),
45
+ datasets.BuilderConfig(name="Real_Seesaw", version=datasets.Version("1.0.0"), description="Real_Seesaw scene"),
46
+ datasets.BuilderConfig(name="Real_Reflection", version=datasets.Version("1.0.0"), description="Real_Reflection scene"),
47
+ datasets.BuilderConfig(name="Real_Pendulum", version=datasets.Version("1.0.0"), description="Real_Pendulum scene"),
48
+ datasets.BuilderConfig(name="Real_Convex_len", version=datasets.Version("1.0.0"), description="Real_Convex_len scene"),
49
+ ]
50
+
51
+ def _info(self):
52
+ return datasets.DatasetInfo(
53
+ description=_DESCRIPTION,
54
+ features=datasets.Features({
55
+ "image": datasets.Image(),
56
+ "file_name": datasets.Value("string"),
57
+ "metadata": datasets.Value("string"), # optionally replace with structured fields
58
+ }),
59
+ homepage=_HOMEPAGE,
60
+ license=_LICENSE,
61
+ citation=_CITATION,
62
+ )
63
+
64
+ def _split_generators(self, dl_manager):
65
+ parts = self.config.name.split("_", 2)
66
+ category = parts[0] + "_" + parts[1]
67
+ if category not in ["real_scenes", "hypothetical_scenes"]:
68
+ raise ValueError(f"Invalid category '{category}'.")
69
+
70
+ scene = parts[2]
71
+ scene_path = os.path.join(category, scene)
72
+
73
+ if os.path.exists(scene_path):
74
+ data_dir = scene_path
75
+ else:
76
+ archive_path = dl_manager.download_and_extract(f"{scene_path}.zip")
77
+ data_dir = os.path.join(archive_path, scene)
78
+
79
+ return [
80
+ datasets.SplitGenerator(
81
+ name=datasets.Split.TRAIN,
82
+ gen_kwargs={"data_dir": data_dir},
83
+ )
84
+ ]
85
+ def _generate_examples(self, data_dir):
86
+ def color(text, code):
87
+ return f"\033[{code}m{text}\033[0m"
88
+ print("load data from {}".format(data_dir))
89
+ try:
90
+ image_files = {}
91
+ for ext in ("*.png", "*.jpg", "*.jpeg"):
92
+ for img_path in Path(data_dir).rglob(ext):
93
+ relative_path = str(img_path.relative_to(data_dir))
94
+ image_files[relative_path] = str(img_path)
95
+ parts = [i.split('/')[0] for i in list(image_files.keys())]
96
+ parts = set(parts)
97
+ if "part_000" not in parts:
98
+ parts= ['']
99
+
100
+
101
+ except Exception as e:
102
+ print(color(f"Error loading images: {e}", "31")) # Red
103
+ return
104
+
105
+ # Find the .csv file
106
+ csv_files = list(Path(data_dir).rglob("*.csv"))
107
+ csv_files = [f for f in Path(data_dir).rglob("*.csv") if not f.name.startswith("._")]
108
+ if not csv_files:
109
+ # print(f"\033[33m[SKIP] No CSV found in {data_dir}, skipping this config.\033[0m")
110
+ pass
111
+ # print(f"\033[33m[INFO] Found CSV: {csv_files}\033[0m")
112
+ csv_path = csv_files[0] if csv_files else None
113
+ df = pd.read_csv(csv_path) if csv_path else None
114
+ image_col_exists = True
115
+ if df is not None and "image" not in df.columns:
116
+ image_col_exists = False
117
+
118
+ images = df["image"].tolist() if image_col_exists and df is not None else []
119
+ images = [i.split('/')[-1].split('.')[0] for i in images if i.endswith(('.png', '.jpg', '.jpeg'))]
120
+
121
+ try:
122
+ # Match CSV rows with image paths
123
+ if df is None:
124
+ for i, j in tqdm(image_files.items(), desc="Processing images", unit="image"):
125
+ yield i, {
126
+ "image": j,
127
+ "file_name": i,
128
+ "metadata": None,
129
+ }
130
+
131
+ else:
132
+ for idx, row in tqdm(df.iterrows(), total=len(df), desc="Processing rows", unit="row"):
133
+ fname = row["ID"]
134
+ raw_record_img_path = images[idx] if images else "" #row["image"]
135
+ record_img_name = raw_record_img_path.split('/')[-1]
136
+ for part in parts:
137
+ if part == '':
138
+ record_img_path = record_img_name
139
+ else:
140
+ record_img_path = "/".join([part, record_img_name.strip()])
141
+ if "Water_flow_scene_render" in data_dir:
142
+ record_img_path = "/".join([part, str(int(record_img_name.strip().split('.')[0]))+".png"])
143
+ if record_img_path in image_files:
144
+ # print(color(f"record_img_path: { image_files[record_img_path]}", "34")) # Blue
145
+ yield idx, {
146
+ "image": image_files[record_img_path],
147
+ "file_name": fname,
148
+ "metadata": row.to_json(),
149
+ }
150
+ break
151
+
152
+ else:
153
+ yield idx, {
154
+ # "image": "",
155
+ "file_name": fname,
156
+ "metadata": row.to_json(),
157
+ }
158
+ break
159
+
160
+
161
+ except Exception as e:
162
+ print(color(f"Error processing CSV rows: {e}", "31"))
163
+
Code/explore.ipynb ADDED
@@ -0,0 +1,857 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cells": [
3
+ {
4
+ "cell_type": "code",
5
+ "execution_count": 1,
6
+ "id": "cc0b451f",
7
+ "metadata": {},
8
+ "outputs": [],
9
+ "source": [
10
+ "import os\n",
11
+ "import pandas as pd\n",
12
+ "\n",
13
+ "base = os.getcwd()"
14
+ ]
15
+ },
16
+ {
17
+ "cell_type": "markdown",
18
+ "id": "e3b39b13",
19
+ "metadata": {},
20
+ "source": [
21
+ "# Real"
22
+ ]
23
+ },
24
+ {
25
+ "cell_type": "code",
26
+ "execution_count": 2,
27
+ "id": "85e4d8c2",
28
+ "metadata": {},
29
+ "outputs": [
30
+ {
31
+ "data": {
32
+ "text/plain": [
33
+ "'base: /Users/dsl/Desktop/Causal3D_Dataset/Code'"
34
+ ]
35
+ },
36
+ "metadata": {},
37
+ "output_type": "display_data"
38
+ },
39
+ {
40
+ "data": {
41
+ "text/plain": [
42
+ "'data_path: /Users/dsl/Desktop/Causal3D_Dataset/Code/../Real'"
43
+ ]
44
+ },
45
+ "metadata": {},
46
+ "output_type": "display_data"
47
+ },
48
+ {
49
+ "data": {
50
+ "text/plain": [
51
+ "\"Scene: Convex_len, Number of files: 10001, File types: {'.csv', '.png'}\""
52
+ ]
53
+ },
54
+ "metadata": {},
55
+ "output_type": "display_data"
56
+ },
57
+ {
58
+ "data": {
59
+ "text/plain": [
60
+ "\"Scene: Seesaw, Number of files: 10001, File types: {'.csv', '.png'}\""
61
+ ]
62
+ },
63
+ "metadata": {},
64
+ "output_type": "display_data"
65
+ },
66
+ {
67
+ "data": {
68
+ "text/plain": [
69
+ "\"Scene: Pendulum, Number of files: 10001, File types: {'.csv', '.png'}\""
70
+ ]
71
+ },
72
+ "metadata": {},
73
+ "output_type": "display_data"
74
+ },
75
+ {
76
+ "data": {
77
+ "text/plain": [
78
+ "\"Scene: Water_flow, Number of files: 10001, File types: {'.csv', '.png'}\""
79
+ ]
80
+ },
81
+ "metadata": {},
82
+ "output_type": "display_data"
83
+ },
84
+ {
85
+ "data": {
86
+ "text/plain": [
87
+ "\"Scene: Parabola, Number of files: 10001, File types: {'.csv', '.png'}\""
88
+ ]
89
+ },
90
+ "metadata": {},
91
+ "output_type": "display_data"
92
+ },
93
+ {
94
+ "data": {
95
+ "text/plain": [
96
+ "\"Scene: Magnet, Number of files: 10001, File types: {'.csv', '.png'}\""
97
+ ]
98
+ },
99
+ "metadata": {},
100
+ "output_type": "display_data"
101
+ },
102
+ {
103
+ "data": {
104
+ "text/plain": [
105
+ "\"Scene: Spring, Number of files: 10001, File types: {'.csv', '.png'}\""
106
+ ]
107
+ },
108
+ "metadata": {},
109
+ "output_type": "display_data"
110
+ },
111
+ {
112
+ "data": {
113
+ "text/plain": [
114
+ "\"Scene: Reflection, Number of files: 10001, File types: {'.csv', '.png'}\""
115
+ ]
116
+ },
117
+ "metadata": {},
118
+ "output_type": "display_data"
119
+ }
120
+ ],
121
+ "source": [
122
+ "from IPython.display import display\n",
123
+ "\n",
124
+ "base = os.getcwd()\n",
125
+ "data_path = os.path.join(base, \"../Real\")\n",
126
+ "scenes = os.listdir(data_path)\n",
127
+ "\n",
128
+ "display(f\"base: {base}\")\n",
129
+ "display(f\"data_path: {data_path}\")\n",
130
+ "all_scenes = [scene for scene in scenes if os.path.isdir(os.path.join(data_path, scene))]\n",
131
+ "# get the number of files and the corresponding type under the each scene\n",
132
+ "for scene in all_scenes:\n",
133
+ " scene_path = os.path.join(data_path, scene)\n",
134
+ " files = os.listdir(scene_path)\n",
135
+ " file_types = set([os.path.splitext(file)[1] for file in files])\n",
136
+ " display(f\"Scene: {scene}, Number of files: {len(files)}, File types: {file_types}\") \n",
137
+ " # You see an empty string '' in file_types because some files in the directory do not have an extension.\n",
138
+ " # For example, hidden files like '.DS_Store' or files without a dot will result in an empty extension from os.path.splitext.\n",
139
+ " # To check which files have no extension:\n",
140
+ " "
141
+ ]
142
+ },
143
+ {
144
+ "cell_type": "code",
145
+ "execution_count": 35,
146
+ "id": "6979c835",
147
+ "metadata": {},
148
+ "outputs": [
149
+ {
150
+ "data": {
151
+ "text/plain": [
152
+ "{'Convex_len': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Convex_len/tabular.csv'],\n",
153
+ " 'Seesaw': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Seesaw/tabular.csv'],\n",
154
+ " 'Pendulum': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Pendulum/tabular.csv'],\n",
155
+ " 'Water_flow': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Water_flow/tabular.csv'],\n",
156
+ " 'Parabola': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Parabola/tabular.csv'],\n",
157
+ " 'Magnet': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Magnet/tabular.csv'],\n",
158
+ " 'Spring': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Spring/tabular.csv'],\n",
159
+ " 'Reflection': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Reflection/tabular.csv']}"
160
+ ]
161
+ },
162
+ "execution_count": 35,
163
+ "metadata": {},
164
+ "output_type": "execute_result"
165
+ }
166
+ ],
167
+ "source": [
168
+ "scene_csv_paths = {}\n",
169
+ "for scene in all_scenes:\n",
170
+ " folder = os.path.join(data_path, scene)\n",
171
+ " if not os.path.isdir(folder):\n",
172
+ " continue\n",
173
+ " csv_files = [f for f in os.listdir(folder) if f.endswith('.csv')]\n",
174
+ " scene_csv_paths[scene] = [os.path.join(folder, f) for f in csv_files]\n",
175
+ "scene_csv_paths\n"
176
+ ]
177
+ },
178
+ {
179
+ "cell_type": "code",
180
+ "execution_count": 66,
181
+ "id": "c58bbbd3",
182
+ "metadata": {},
183
+ "outputs": [
184
+ {
185
+ "data": {
186
+ "text/plain": [
187
+ "{'Convex_len': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Convex_len/tabular.csv'],\n",
188
+ " 'Seesaw': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Seesaw/tabular.csv'],\n",
189
+ " 'Pendulum': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Pendulum/tabular.csv'],\n",
190
+ " 'Water_flow': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Water_flow/tabular.csv'],\n",
191
+ " 'Parabola': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Parabola/tabular.csv'],\n",
192
+ " 'Magnet': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Magnet/tabular.csv'],\n",
193
+ " 'Spring': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Spring/tabular.csv'],\n",
194
+ " 'Reflection': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Real/Reflection/tabular.csv']}"
195
+ ]
196
+ },
197
+ "execution_count": 66,
198
+ "metadata": {},
199
+ "output_type": "execute_result"
200
+ }
201
+ ],
202
+ "source": [
203
+ "scene_csv_paths"
204
+ ]
205
+ },
206
+ {
207
+ "cell_type": "code",
208
+ "execution_count": null,
209
+ "id": "54888282",
210
+ "metadata": {},
211
+ "outputs": [],
212
+ "source": [
213
+ "import pandas as pd\n",
214
+ "\n",
215
+ "# === Function to convert filenames ===\n",
216
+ "def convert_filename(filename):\n",
217
+ " if isinstance(filename, str) and filename.endswith('.png'):\n",
218
+ " # Remove leading zeros from the number part\n",
219
+ " number_part = filename.split('.')[0]\n",
220
+ " new_number = str(int(number_part))\n",
221
+ " return f\"{new_number}.png\"\n",
222
+ " return filename # if not a string or doesn't match, return as is\n",
223
+ "\n"
224
+ ]
225
+ },
226
+ {
227
+ "cell_type": "code",
228
+ "execution_count": 70,
229
+ "id": "9b3afa52",
230
+ "metadata": {},
231
+ "outputs": [],
232
+ "source": [
233
+ "# check each csv file, are there all of png files exists\n",
234
+ "for i in scene_csv_paths:\n",
235
+ " df = pd.read_csv(scene_csv_paths[i][0])\n",
236
+ " df['imgs'] = df['imgs'].apply(convert_filename)\n",
237
+ " df.to_csv(scene_csv_paths[i][0], index=False)\n"
238
+ ]
239
+ },
240
+ {
241
+ "cell_type": "code",
242
+ "execution_count": 71,
243
+ "id": "50723819",
244
+ "metadata": {},
245
+ "outputs": [],
246
+ "source": [
247
+ "# check each csv file, are there all of png files exists\n",
248
+ "for i in scene_csv_paths:\n",
249
+ " df = pd.read_csv(scene_csv_paths[i][0])\n",
250
+ " imgs = df['imgs'].tolist()\n",
251
+ " for img in imgs:\n",
252
+ " img_path = os.path.join(data_path, i, img)\n",
253
+ " if not os.path.exists(img_path):\n",
254
+ " print(f\"Image {img} does not exist in scene {i}.\")\n",
255
+ "\n"
256
+ ]
257
+ },
258
+ {
259
+ "cell_type": "markdown",
260
+ "id": "8d7f2bad",
261
+ "metadata": {},
262
+ "source": [
263
+ "# Hypothetic\n"
264
+ ]
265
+ },
266
+ {
267
+ "cell_type": "code",
268
+ "execution_count": 40,
269
+ "id": "d9c4b659",
270
+ "metadata": {},
271
+ "outputs": [
272
+ {
273
+ "name": "stdout",
274
+ "output_type": "stream",
275
+ "text": [
276
+ "Scene: V4_linear, Number of files: 10003, File types: {'.png': 10001, '': 1, '.csv': 1}\n",
277
+ "Scene: V4_v_structure_linear, Number of files: 10001, File types: {'.png': 10000, '.csv': 1}\n",
278
+ "Scene: V3_fully_connected_linear, Number of files: 10002, File types: {'.png': 10000, '.md': 1, '.csv': 1}\n",
279
+ "Scene: V2_linear, Number of files: 10002, File types: {'.png': 10000, '.md': 1, '.csv': 1}\n",
280
+ "Scene: V2_nonlinear, Number of files: 10002, File types: {'.png': 10000, '.md': 1, '.csv': 1}\n",
281
+ "Scene: V5_linear, Number of files: 10001, File types: {'.png': 10000, '.csv': 1}\n",
282
+ "Scene: V3_v_structure_nonlinear, Number of files: 10002, File types: {'.png': 10000, '': 1, '.csv': 1}\n",
283
+ "Scene: V3_v_structure_linear, Number of files: 10002, File types: {'.png': 10000, '': 1, '.csv': 1}\n",
284
+ "Scene: V5_v_structure_nonlinear, Number of files: 10001, File types: {'.png': 10000, '.csv': 1}\n",
285
+ "Scene: V5_v_structure_linear, Number of files: 10001, File types: {'.png': 10000, '.csv': 1}\n",
286
+ "Scene: V4_v_strcuture_nonlinear, Number of files: 10001, File types: {'.png': 10000, '.csv': 1}\n"
287
+ ]
288
+ }
289
+ ],
290
+ "source": [
291
+ "hy_base_path = os.path.join(base, \"../Hypothetic\")\n",
292
+ "hy_scenes = os.listdir(hy_base_path)\n",
293
+ "for scene in hy_scenes:\n",
294
+ " scene_path = os.path.join(hy_base_path, scene)\n",
295
+ " if not os.path.isdir(scene_path) or scene.startswith('.'):\n",
296
+ " continue\n",
297
+ " files = os.listdir(scene_path)\n",
298
+ " file_types = {}\n",
299
+ " for file in files:\n",
300
+ " ext = os.path.splitext(file)[1]\n",
301
+ " file_types[ext] = file_types.get(ext, 0) + 1\n",
302
+ " print(f\"Scene: {scene}, Number of files: {len(files)}, File types: {file_types}\")"
303
+ ]
304
+ },
305
+ {
306
+ "cell_type": "code",
307
+ "execution_count": 7,
308
+ "id": "6444b695",
309
+ "metadata": {},
310
+ "outputs": [
311
+ {
312
+ "data": {
313
+ "text/plain": [
314
+ "{'V4_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V4_linear/tabular.csv'],\n",
315
+ " 'V4_v_structure_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V4_v_structure_linear/tabular.csv'],\n",
316
+ " 'V3_fully_connected_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V3_fully_connected_linear/tabular.csv'],\n",
317
+ " 'V2_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V2_linear/tabular.csv'],\n",
318
+ " 'V2_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V2_nonlinear/tabular.csv'],\n",
319
+ " 'V5_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V5_linear/tabular.csv'],\n",
320
+ " 'V3_v_structure_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V3_v_structure_nonlinear/tabular.csv'],\n",
321
+ " 'V3_v_structure_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V3_v_structure_linear/tabular.csv'],\n",
322
+ " 'V5_v_structure_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V5_v_structure_nonlinear/tabular.csv'],\n",
323
+ " 'V5_v_structure_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V5_v_structure_linear/tabular.csv'],\n",
324
+ " 'V4_v_strcuture_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V4_v_strcuture_nonlinear/tabular.csv']}"
325
+ ]
326
+ },
327
+ "execution_count": 7,
328
+ "metadata": {},
329
+ "output_type": "execute_result"
330
+ }
331
+ ],
332
+ "source": [
333
+ "hy_scene_csv_paths = {}\n",
334
+ "for scene in hy_scenes:\n",
335
+ " scene_folder = os.path.join(hy_base_path, scene)\n",
336
+ " if not os.path.isdir(scene_folder) or scene.startswith('.'):\n",
337
+ " continue\n",
338
+ " csvs = [f for f in os.listdir(scene_folder) if f.endswith('.csv')]\n",
339
+ " hy_scene_csv_paths[scene] = [os.path.join(scene_folder, f) for f in csvs]\n",
340
+ "hy_scene_csv_paths"
341
+ ]
342
+ },
343
+ {
344
+ "cell_type": "code",
345
+ "execution_count": 8,
346
+ "id": "4bc1be73",
347
+ "metadata": {},
348
+ "outputs": [
349
+ {
350
+ "name": "stdout",
351
+ "output_type": "stream",
352
+ "text": [
353
+ "Hypothetic Scene: V4_linear, CSV: tabular.csv\n",
354
+ "Columns: ['volume_ball', 'height_cuboid', 'base_area_cuboid', 'base_area_cone', 'imgs']\n",
355
+ "\n",
356
+ "Hypothetic Scene: V4_v_structure_linear, CSV: tabular.csv\n",
357
+ "Columns: ['volumn_ball', 'height_cuboid', 'base_area_cuboid', 'base_area_cone', 'imgs']\n",
358
+ "\n",
359
+ "Hypothetic Scene: V3_fully_connected_linear, CSV: tabular.csv\n",
360
+ "Columns: ['iter', 'volume_ball', 'height_of_cuboid', 'base_area_cone', 'img_path']\n",
361
+ "\n",
362
+ "Hypothetic Scene: V2_linear, CSV: tabular.csv\n",
363
+ "Columns: ['iter', 'volume_ball', 'r_ball', 'volume_cube', 'edge_cube', 'img_path']\n",
364
+ "\n",
365
+ "Hypothetic Scene: V2_nonlinear, CSV: tabular.csv\n",
366
+ "Columns: ['iter', 'volume_ball', 'r_ball', 'scaled_volume_ball', 'volume_cube', 'edge_cube', 'img_path']\n",
367
+ "\n",
368
+ "Hypothetic Scene: V5_linear, CSV: tabular.csv\n",
369
+ "Columns: ['volumn_ball', 'height_cuboid', 'base_area_cuboid', 'base_area_cone', 'height_cone', 'imgs']\n",
370
+ "\n",
371
+ "Hypothetic Scene: V3_v_structure_nonlinear, CSV: tabular.csv\n",
372
+ "Columns: ['volume_ball', 'scaled_volume_ball', 'radius of ball', 'height_cylinder', 'radius of cone', 'basal_area_cone', 'imgs']\n",
373
+ "\n",
374
+ "Hypothetic Scene: V3_v_structure_linear, CSV: tabular.csv\n",
375
+ "Columns: ['volume_ball', 'height_cylinder', 'basal_area_cone', 'imgs', 'Unnamed: 4']\n",
376
+ "\n",
377
+ "Hypothetic Scene: V5_v_structure_nonlinear, CSV: tabular.csv\n",
378
+ "Columns: ['volumn_ball', 'scaled_volumn_ball', 'height_cuboid', 'base_area_cuboid', 'scaled_base_area_cuboid', 'base_area_cone', 'scaled_base_area_cone', 'height_cone', 'imgs']\n",
379
+ "\n",
380
+ "Hypothetic Scene: V5_v_structure_linear, CSV: tabular.csv\n",
381
+ "Columns: ['volumn_ball', 'height_cuboid', 'base_area_cuboid', 'base_area_cone', 'height_cone', 'imgs']\n",
382
+ "\n",
383
+ "Hypothetic Scene: V4_v_strcuture_nonlinear, CSV: tabular.csv\n",
384
+ "Columns: ['volumn_ball', 'scaled_volumn_ball', 'height_cuboid', 'base_area_cuboid', 'scaled_base_area_cuboid', 'base_area_cone', 'imgs']\n",
385
+ "\n"
386
+ ]
387
+ }
388
+ ],
389
+ "source": [
390
+ "for scene, csv_paths in hy_scene_csv_paths.items():\n",
391
+ " for csv_file in csv_paths:\n",
392
+ " df = pd.read_csv(csv_file, nrows=0)\n",
393
+ " print(f\"Hypothetic Scene: {scene}, CSV: {os.path.basename(csv_file)}\")\n",
394
+ " print(\"Columns:\", list(df.columns))\n",
395
+ " print()"
396
+ ]
397
+ },
398
+ {
399
+ "cell_type": "code",
400
+ "execution_count": 43,
401
+ "id": "fefc0d47",
402
+ "metadata": {},
403
+ "outputs": [
404
+ {
405
+ "data": {
406
+ "text/plain": [
407
+ "dict_keys(['/Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V3_fully_connected_linear', '/Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V2_nonlinear', '/Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V2_linear'])"
408
+ ]
409
+ },
410
+ "execution_count": 43,
411
+ "metadata": {},
412
+ "output_type": "execute_result"
413
+ }
414
+ ],
415
+ "source": [
416
+ "process_path = [\"/Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V3_fully_connected_linear\",\n",
417
+ " \"/Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V2_nonlinear\",\n",
418
+ " \"/Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V2_linear\"]\n",
419
+ "dfs_dict = {}\n",
420
+ "for path in process_path:\n",
421
+ " csv_file = os.path.join(path, \"tabular.csv\")\n",
422
+ " if os.path.exists(csv_file):\n",
423
+ " df = pd.read_csv(csv_file)\n",
424
+ " dfs_dict[path] = df\n",
425
+ "dfs_dict.keys()\n"
426
+ ]
427
+ },
428
+ {
429
+ "cell_type": "code",
430
+ "execution_count": 46,
431
+ "id": "e6c53182",
432
+ "metadata": {},
433
+ "outputs": [
434
+ {
435
+ "name": "stdout",
436
+ "output_type": "stream",
437
+ "text": [
438
+ "DataFrame for: /Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V3_fully_connected_linear\n"
439
+ ]
440
+ },
441
+ {
442
+ "data": {
443
+ "text/html": [
444
+ "<div>\n",
445
+ "<style scoped>\n",
446
+ " .dataframe tbody tr th:only-of-type {\n",
447
+ " vertical-align: middle;\n",
448
+ " }\n",
449
+ "\n",
450
+ " .dataframe tbody tr th {\n",
451
+ " vertical-align: top;\n",
452
+ " }\n",
453
+ "\n",
454
+ " .dataframe thead th {\n",
455
+ " text-align: right;\n",
456
+ " }\n",
457
+ "</style>\n",
458
+ "<table border=\"1\" class=\"dataframe\">\n",
459
+ " <thead>\n",
460
+ " <tr style=\"text-align: right;\">\n",
461
+ " <th></th>\n",
462
+ " <th>iter</th>\n",
463
+ " <th>volume_ball</th>\n",
464
+ " <th>height_of_cuboid</th>\n",
465
+ " <th>base_area_cone</th>\n",
466
+ " <th>imgs</th>\n",
467
+ " </tr>\n",
468
+ " </thead>\n",
469
+ " <tbody>\n",
470
+ " <tr>\n",
471
+ " <th>0</th>\n",
472
+ " <td>1</td>\n",
473
+ " <td>4.461709</td>\n",
474
+ " <td>2.230855</td>\n",
475
+ " <td>4.461709</td>\n",
476
+ " <td>00001.png</td>\n",
477
+ " </tr>\n",
478
+ " <tr>\n",
479
+ " <th>1</th>\n",
480
+ " <td>2</td>\n",
481
+ " <td>4.641952</td>\n",
482
+ " <td>2.320976</td>\n",
483
+ " <td>4.641952</td>\n",
484
+ " <td>00002.png</td>\n",
485
+ " </tr>\n",
486
+ " <tr>\n",
487
+ " <th>2</th>\n",
488
+ " <td>3</td>\n",
489
+ " <td>5.732580</td>\n",
490
+ " <td>2.866290</td>\n",
491
+ " <td>5.732580</td>\n",
492
+ " <td>00003.png</td>\n",
493
+ " </tr>\n",
494
+ " <tr>\n",
495
+ " <th>3</th>\n",
496
+ " <td>4</td>\n",
497
+ " <td>9.686783</td>\n",
498
+ " <td>4.843392</td>\n",
499
+ " <td>9.686783</td>\n",
500
+ " <td>00004.png</td>\n",
501
+ " </tr>\n",
502
+ " <tr>\n",
503
+ " <th>4</th>\n",
504
+ " <td>5</td>\n",
505
+ " <td>2.608935</td>\n",
506
+ " <td>1.304468</td>\n",
507
+ " <td>2.608935</td>\n",
508
+ " <td>00005.png</td>\n",
509
+ " </tr>\n",
510
+ " </tbody>\n",
511
+ "</table>\n",
512
+ "</div>"
513
+ ],
514
+ "text/plain": [
515
+ " iter volume_ball height_of_cuboid base_area_cone imgs\n",
516
+ "0 1 4.461709 2.230855 4.461709 00001.png\n",
517
+ "1 2 4.641952 2.320976 4.641952 00002.png\n",
518
+ "2 3 5.732580 2.866290 5.732580 00003.png\n",
519
+ "3 4 9.686783 4.843392 9.686783 00004.png\n",
520
+ "4 5 2.608935 1.304468 2.608935 00005.png"
521
+ ]
522
+ },
523
+ "metadata": {},
524
+ "output_type": "display_data"
525
+ },
526
+ {
527
+ "name": "stdout",
528
+ "output_type": "stream",
529
+ "text": [
530
+ "Shape: (10000, 5)\n",
531
+ "------------------------------------------------------------\n",
532
+ "DataFrame for: /Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V2_nonlinear\n"
533
+ ]
534
+ },
535
+ {
536
+ "data": {
537
+ "text/html": [
538
+ "<div>\n",
539
+ "<style scoped>\n",
540
+ " .dataframe tbody tr th:only-of-type {\n",
541
+ " vertical-align: middle;\n",
542
+ " }\n",
543
+ "\n",
544
+ " .dataframe tbody tr th {\n",
545
+ " vertical-align: top;\n",
546
+ " }\n",
547
+ "\n",
548
+ " .dataframe thead th {\n",
549
+ " text-align: right;\n",
550
+ " }\n",
551
+ "</style>\n",
552
+ "<table border=\"1\" class=\"dataframe\">\n",
553
+ " <thead>\n",
554
+ " <tr style=\"text-align: right;\">\n",
555
+ " <th></th>\n",
556
+ " <th>iter</th>\n",
557
+ " <th>volume_ball</th>\n",
558
+ " <th>r_ball</th>\n",
559
+ " <th>scaled_volume_ball</th>\n",
560
+ " <th>volume_cube</th>\n",
561
+ " <th>edge_cube</th>\n",
562
+ " <th>imgs</th>\n",
563
+ " </tr>\n",
564
+ " </thead>\n",
565
+ " <tbody>\n",
566
+ " <tr>\n",
567
+ " <th>0</th>\n",
568
+ " <td>1</td>\n",
569
+ " <td>8.369589</td>\n",
570
+ " <td>1.259520</td>\n",
571
+ " <td>-0.260683</td>\n",
572
+ " <td>5.797285</td>\n",
573
+ " <td>1.796421</td>\n",
574
+ " <td>00001.png</td>\n",
575
+ " </tr>\n",
576
+ " <tr>\n",
577
+ " <th>1</th>\n",
578
+ " <td>2</td>\n",
579
+ " <td>8.748098</td>\n",
580
+ " <td>1.278228</td>\n",
581
+ " <td>-0.201078</td>\n",
582
+ " <td>5.879111</td>\n",
583
+ " <td>1.804834</td>\n",
584
+ " <td>00002.png</td>\n",
585
+ " </tr>\n",
586
+ " <tr>\n",
587
+ " <th>2</th>\n",
588
+ " <td>3</td>\n",
589
+ " <td>11.038418</td>\n",
590
+ " <td>1.381251</td>\n",
591
+ " <td>0.159586</td>\n",
592
+ " <td>5.923759</td>\n",
593
+ " <td>1.809391</td>\n",
594
+ " <td>00003.png</td>\n",
595
+ " </tr>\n",
596
+ " <tr>\n",
597
+ " <th>3</th>\n",
598
+ " <td>4</td>\n",
599
+ " <td>19.342245</td>\n",
600
+ " <td>1.665224</td>\n",
601
+ " <td>1.467218</td>\n",
602
+ " <td>0.620362</td>\n",
603
+ " <td>0.852868</td>\n",
604
+ " <td>00004.png</td>\n",
605
+ " </tr>\n",
606
+ " <tr>\n",
607
+ " <th>4</th>\n",
608
+ " <td>5</td>\n",
609
+ " <td>4.478764</td>\n",
610
+ " <td>1.022562</td>\n",
611
+ " <td>-0.873384</td>\n",
612
+ " <td>3.853417</td>\n",
613
+ " <td>1.567769</td>\n",
614
+ " <td>00005.png</td>\n",
615
+ " </tr>\n",
616
+ " </tbody>\n",
617
+ "</table>\n",
618
+ "</div>"
619
+ ],
620
+ "text/plain": [
621
+ " iter volume_ball r_ball scaled_volume_ball volume_cube edge_cube \\\n",
622
+ "0 1 8.369589 1.259520 -0.260683 5.797285 1.796421 \n",
623
+ "1 2 8.748098 1.278228 -0.201078 5.879111 1.804834 \n",
624
+ "2 3 11.038418 1.381251 0.159586 5.923759 1.809391 \n",
625
+ "3 4 19.342245 1.665224 1.467218 0.620362 0.852868 \n",
626
+ "4 5 4.478764 1.022562 -0.873384 3.853417 1.567769 \n",
627
+ "\n",
628
+ " imgs \n",
629
+ "0 00001.png \n",
630
+ "1 00002.png \n",
631
+ "2 00003.png \n",
632
+ "3 00004.png \n",
633
+ "4 00005.png "
634
+ ]
635
+ },
636
+ "metadata": {},
637
+ "output_type": "display_data"
638
+ },
639
+ {
640
+ "name": "stdout",
641
+ "output_type": "stream",
642
+ "text": [
643
+ "Shape: (10000, 7)\n",
644
+ "------------------------------------------------------------\n",
645
+ "DataFrame for: /Users/dsl/Desktop/Causal3D_Dataset/Hypothetic/V2_linear\n"
646
+ ]
647
+ },
648
+ {
649
+ "data": {
650
+ "text/html": [
651
+ "<div>\n",
652
+ "<style scoped>\n",
653
+ " .dataframe tbody tr th:only-of-type {\n",
654
+ " vertical-align: middle;\n",
655
+ " }\n",
656
+ "\n",
657
+ " .dataframe tbody tr th {\n",
658
+ " vertical-align: top;\n",
659
+ " }\n",
660
+ "\n",
661
+ " .dataframe thead th {\n",
662
+ " text-align: right;\n",
663
+ " }\n",
664
+ "</style>\n",
665
+ "<table border=\"1\" class=\"dataframe\">\n",
666
+ " <thead>\n",
667
+ " <tr style=\"text-align: right;\">\n",
668
+ " <th></th>\n",
669
+ " <th>iter</th>\n",
670
+ " <th>volume_ball</th>\n",
671
+ " <th>r_ball</th>\n",
672
+ " <th>volume_cube</th>\n",
673
+ " <th>edge_cube</th>\n",
674
+ " <th>imgs</th>\n",
675
+ " </tr>\n",
676
+ " </thead>\n",
677
+ " <tbody>\n",
678
+ " <tr>\n",
679
+ " <th>0</th>\n",
680
+ " <td>1</td>\n",
681
+ " <td>7.118523</td>\n",
682
+ " <td>1.193348</td>\n",
683
+ " <td>10.677784</td>\n",
684
+ " <td>2.202049</td>\n",
685
+ " <td>00001.png</td>\n",
686
+ " </tr>\n",
687
+ " <tr>\n",
688
+ " <th>1</th>\n",
689
+ " <td>2</td>\n",
690
+ " <td>7.440114</td>\n",
691
+ " <td>1.211054</td>\n",
692
+ " <td>11.160170</td>\n",
693
+ " <td>2.234723</td>\n",
694
+ " <td>00002.png</td>\n",
695
+ " </tr>\n",
696
+ " <tr>\n",
697
+ " <th>2</th>\n",
698
+ " <td>3</td>\n",
699
+ " <td>9.386024</td>\n",
700
+ " <td>1.308572</td>\n",
701
+ " <td>14.079037</td>\n",
702
+ " <td>2.414669</td>\n",
703
+ " <td>00003.png</td>\n",
704
+ " </tr>\n",
705
+ " <tr>\n",
706
+ " <th>3</th>\n",
707
+ " <td>4</td>\n",
708
+ " <td>16.441156</td>\n",
709
+ " <td>1.577422</td>\n",
710
+ " <td>24.661734</td>\n",
711
+ " <td>2.910770</td>\n",
712
+ " <td>00004.png</td>\n",
713
+ " </tr>\n",
714
+ " <tr>\n",
715
+ " <th>4</th>\n",
716
+ " <td>5</td>\n",
717
+ " <td>3.812784</td>\n",
718
+ " <td>0.969136</td>\n",
719
+ " <td>5.719176</td>\n",
720
+ " <td>1.788317</td>\n",
721
+ " <td>00005.png</td>\n",
722
+ " </tr>\n",
723
+ " </tbody>\n",
724
+ "</table>\n",
725
+ "</div>"
726
+ ],
727
+ "text/plain": [
728
+ " iter volume_ball r_ball volume_cube edge_cube imgs\n",
729
+ "0 1 7.118523 1.193348 10.677784 2.202049 00001.png\n",
730
+ "1 2 7.440114 1.211054 11.160170 2.234723 00002.png\n",
731
+ "2 3 9.386024 1.308572 14.079037 2.414669 00003.png\n",
732
+ "3 4 16.441156 1.577422 24.661734 2.910770 00004.png\n",
733
+ "4 5 3.812784 0.969136 5.719176 1.788317 00005.png"
734
+ ]
735
+ },
736
+ "metadata": {},
737
+ "output_type": "display_data"
738
+ },
739
+ {
740
+ "name": "stdout",
741
+ "output_type": "stream",
742
+ "text": [
743
+ "Shape: (10000, 6)\n",
744
+ "------------------------------------------------------------\n"
745
+ ]
746
+ }
747
+ ],
748
+ "source": [
749
+ "for path, df in dfs_dict.items():\n",
750
+ " print(f\"DataFrame for: {path}\")\n",
751
+ " display(df.head())\n",
752
+ " print(f\"Shape: {df.shape}\")\n",
753
+ " print(\"-\" * 60)\n",
754
+ " # if 'img_path' in df.columns:\n",
755
+ " # df = df.rename(columns={'img_path': 'imgs'})\n",
756
+ " # # Update the original CSV file\n",
757
+ " # csv_file = os.path.join(path, \"tabular.csv\")\n",
758
+ " # df.to_csv(csv_file, index=False)\n",
759
+ " # print(f\"Updated column name and saved: {csv_file}\")"
760
+ ]
761
+ },
762
+ {
763
+ "cell_type": "code",
764
+ "execution_count": 41,
765
+ "id": "835862a9",
766
+ "metadata": {},
767
+ "outputs": [
768
+ {
769
+ "data": {
770
+ "text/plain": [
771
+ "{'V4_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V4_linear/tabular.csv'],\n",
772
+ " 'V4_v_structure_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V4_v_structure_linear/tabular.csv'],\n",
773
+ " 'V3_fully_connected_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V3_fully_connected_linear/tabular.csv'],\n",
774
+ " 'V2_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V2_linear/tabular.csv'],\n",
775
+ " 'V2_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V2_nonlinear/tabular.csv'],\n",
776
+ " 'V5_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V5_linear/tabular.csv'],\n",
777
+ " 'V3_v_structure_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V3_v_structure_nonlinear/tabular.csv'],\n",
778
+ " 'V3_v_structure_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V3_v_structure_linear/tabular.csv'],\n",
779
+ " 'V5_v_structure_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V5_v_structure_nonlinear/tabular.csv'],\n",
780
+ " 'V5_v_structure_linear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V5_v_structure_linear/tabular.csv'],\n",
781
+ " 'V4_v_strcuture_nonlinear': ['/Users/dsl/Desktop/Causal3D_Dataset/Code/../Hypothetic/V4_v_strcuture_nonlinear/tabular.csv']}"
782
+ ]
783
+ },
784
+ "execution_count": 41,
785
+ "metadata": {},
786
+ "output_type": "execute_result"
787
+ }
788
+ ],
789
+ "source": [
790
+ "hy_scene_csv_paths"
791
+ ]
792
+ },
793
+ {
794
+ "cell_type": "code",
795
+ "execution_count": 42,
796
+ "id": "f5d085eb",
797
+ "metadata": {},
798
+ "outputs": [
799
+ {
800
+ "name": "stdout",
801
+ "output_type": "stream",
802
+ "text": [
803
+ "Scene: V4_linear, all images exist.\n",
804
+ "Scene: V4_v_structure_linear, all images exist.\n",
805
+ "Scene: V3_fully_connected_linear, all images exist.\n",
806
+ "Scene: V2_linear, all images exist.\n",
807
+ "Scene: V2_nonlinear, all images exist.\n",
808
+ "Scene: V5_linear, all images exist.\n",
809
+ "Scene: V3_v_structure_nonlinear, all images exist.\n",
810
+ "Scene: V3_v_structure_linear, all images exist.\n",
811
+ "Scene: V5_v_structure_nonlinear, all images exist.\n",
812
+ "Scene: V5_v_structure_linear, all images exist.\n",
813
+ "Scene: V4_v_strcuture_nonlinear, all images exist.\n"
814
+ ]
815
+ }
816
+ ],
817
+ "source": [
818
+ "for scene, csv_paths in hy_scene_csv_paths.items():\n",
819
+ " for csv_file in csv_paths:\n",
820
+ " df = pd.read_csv(csv_file)\n",
821
+ " img_col = 'imgs' if 'imgs' in df.columns else 'img_path'\n",
822
+ " img_dir = os.path.dirname(csv_file)\n",
823
+ " missing_imgs = []\n",
824
+ " for img_file in df[img_col]:\n",
825
+ " img_path = os.path.join(img_dir, img_file)\n",
826
+ " if not os.path.exists(img_path):\n",
827
+ " missing_imgs.append(img_file)\n",
828
+ " if missing_imgs:\n",
829
+ " print(f\"Scene: {scene}, Missing images: {len(missing_imgs)}\")\n",
830
+ " print(missing_imgs[:10]) # show up to 10 missing images\n",
831
+ " else:\n",
832
+ " print(f\"Scene: {scene}, all images exist.\")"
833
+ ]
834
+ }
835
+ ],
836
+ "metadata": {
837
+ "kernelspec": {
838
+ "display_name": "base",
839
+ "language": "python",
840
+ "name": "python3"
841
+ },
842
+ "language_info": {
843
+ "codemirror_mode": {
844
+ "name": "ipython",
845
+ "version": 3
846
+ },
847
+ "file_extension": ".py",
848
+ "mimetype": "text/x-python",
849
+ "name": "python",
850
+ "nbconvert_exporter": "python",
851
+ "pygments_lexer": "ipython3",
852
+ "version": "3.13.2"
853
+ }
854
+ },
855
+ "nbformat": 4,
856
+ "nbformat_minor": 5
857
+ }
Hypothetical_V2_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:821ea1b7005bf602b429a62abf7d7501a6345b24050c0c68413ac2af5f55d805
3
+ size 797636768
Hypothetical_V2_nonlinear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b860d2e64a140642e8de87a2e295ca16963539a75474110975f64f1d0e4df38
3
+ size 800066916
Hypothetical_V3_fully_connected_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8719d474608f9f15642cf7ddda795d0bacbc701fd49ed140d68923d4e4e1eb59
3
+ size 822351161
Hypothetical_V3_v_structure_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3a9ca82dd5760fb7729aa022f28836ce12c5e4cfb69c037fdf2540a7a9ca2fe
3
+ size 779911922
Hypothetical_V3_v_structure_nonlinear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a788325be484894ac056ac0e6ae83c79f58ceb51aa0e49d330c2aa3f990ea23c
3
+ size 782501114
Hypothetical_V4_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5948e9a9a4331113e81893c0b194af1174f28968f270238ef1192f425bd4997a
3
+ size 825828410
Hypothetical_V4_v_strcuture_nonlinear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a6372cc5adcc3bf4d122e6ae05fee13fa4fa8fcb177d93af36f787b1ee55067
3
+ size 811083458
Hypothetical_V4_v_structure_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2041f4fa92c7b57326babf9549a7bda94e9a0934f48bf33db645f1a1598f894f
3
+ size 807961243
Hypothetical_V5_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87c7ccb4a6492eae84b609193984805538adbd61f1b904d0eb08290f3f2f011f
3
+ size 809089702
Hypothetical_V5_v_structure_linear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:417818cf5602914ef06dae749761a72dc9468b1e512d948067e4b0a3c8eadf1a
3
+ size 810077103
Hypothetical_V5_v_structure_nonlinear.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c9f66d78b5e801976cf419986ffe5f62c57150501f71b2116903fd86f0266d5
3
+ size 810083426
Real_Convex_len.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e39b46a68b4ba5a4252af37a97abfc658dd637efddffc632d70cd11a2cb53f9a
3
+ size 1266758341
Real_Magnet.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ba9b5da9cdab0bf6712e78c452f9d68f149a0c6d7e9dee7275cfe9d5718a49a
3
+ size 133242122
Real_Parabola.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:044dea8695e172d1a76b4e0023584aef358c3315ba9f9a598671dbf33ca08c0d
3
+ size 1383325161
Real_Pendulum.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66ce6473481ee775e01a9015bfd9e9a7d6c872cd7fbcb7ae5cef64f7da108e80
3
+ size 712987799
Real_Reflection.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df752e9f8e50b9e6232cd2b37cf6b649021dedd6856d636ac610780b5367dabc
3
+ size 790883783
Real_Seesaw.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a212e1e66843e607687ee175289bb8c274eca08b246f227f1389f2937e3d7ced
3
+ size 787423731
Real_Spring.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c8c55d52c242e38f593135a03571e1d37b40396cb82a4ba12dd514e30713864
3
+ size 733517324
Real_Water_flow.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dac569760cf2b057620e7888d39dae504fcdb745f4802fbefe1eb4c2685c1b9a
3
+ size 733356836
zip_scene.py ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+ import zipfile
3
+ from pathlib import Path
4
+
5
+ def zip_folder(folder_path, output_path):
6
+ """
7
+ 压缩单个文件夹
8
+ """
9
+ folder = Path(folder_path)
10
+ with zipfile.ZipFile(output_path, 'w', zipfile.ZIP_DEFLATED) as zipf:
11
+ for file in folder.rglob('*'):
12
+ zipf.write(file, file.relative_to(folder.parent))
13
+ print(f"✅ Zipped {folder} -> {output_path}")
14
+
15
+ def zip_all_scenes(root_dir):
16
+ """
17
+ 扫描 Real, Hypothetical, Multi_View 文件夹,自动打包每个 scene
18
+ """
19
+ root = Path(root_dir)
20
+ categories = ["Real", "Hypothetical", "Multi_View"]
21
+
22
+ for category in categories:
23
+ category_path = root / category
24
+ if not category_path.exists():
25
+ print(f"⚠️ Skip {category}: not found")
26
+ continue
27
+
28
+ for scene in category_path.iterdir():
29
+ if scene.is_dir():
30
+ output_zip = root / f"{category}_{scene.name}.zip"
31
+ zip_folder(scene, output_zip)
32
+
33
+ if __name__ == "__main__":
34
+ # 修改为你的根目录
35
+ zip_all_scenes(".")
36
+
37
+ print("✅ All scenes zipped successfully.")