File size: 6,630 Bytes
12e023b |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 |
import re
from typing import List, Dict
import os
def parse_generation_history(file_path: str) -> Dict[int, List[str]]:
"""Improved parser that handles math symbols and spaces correctly"""
history = {}
token_pattern = re.compile(r"\*([^&]*)&?")
with open(file_path, 'r', encoding='utf-8') as f:
for line in f:
line = line.strip()
if not line:
continue
try:
step_part, content_part = line.split(',', 1)
step = int(step_part.strip())
except ValueError:
continue
tokens = []
for match in token_pattern.finditer(content_part):
raw_token = match.group(1).strip()
if raw_token == "":
tokens.append(" ")
elif raw_token == "*":
tokens.append("*")
else:
tokens.append(raw_token)
while len(tokens) < 64:
tokens.append(" ")
if len(tokens) > 64:
print(f"Truncating extra tokens: Step {step} ({len(tokens)} tokens)")
tokens = tokens[:64]
elif len(tokens) < 64:
print(f"Padding missing tokens: Step {step} ({len(tokens)} tokens)")
tokens += [" "] * (64 - len(tokens))
tokens = tokens[:62]
history[step] = tokens
return history
def track_token_positions(history: Dict[int, List[str]]) -> List[int]:
"""Track the first generation step for each token"""
num_positions = 64
steps_to_unmask = [-1] * num_positions
for step in sorted(history.keys()):
tokens = history[step]
for idx in range(num_positions):
if idx >= len(tokens):
continue
token = tokens[idx]
if steps_to_unmask[idx] == -1 and token != '<|mdm_mask|>':
steps_to_unmask[idx] = step
return steps_to_unmask
def generate_background_color(step: int, max_step: int) -> str:
"""Generate gradient color for text (darker version)"""
color_stops = [
(176, 202, 224),
(143, 179, 207),
(110, 156, 191),
(80, 130, 240),
(40, 90, 200),
(20, 70, 180),
(0, 50, 160),
]
color_index = min(int(step ** 0.7 / max_step ** 0.7 * 6), 6)
ratio = (step % 2) / 2
start = color_stops[color_index]
end = color_stops[min(color_index + 1, 6)]
r = int(start[0] + (end[0] - start[0]) * ratio)
g = int(start[1] + (end[1] - start[1]) * ratio)
b = int(start[2] + (end[2] - start[2]) * ratio)
return f"#{r:02x}{g:02x}{b:02x}"
def generate_step_visualization(current_step: int, current_tokens: List[str],
token_steps: List[int], max_step: int) -> str:
"""Final visualization version (completely borderless)"""
html = []
for idx, token in enumerate(current_tokens):
style = [
"padding: 6px 8px",
"margin: 2px",
"border-radius: 6px",
"display: inline-block",
"font-weight: 600",
"font-size: 16px",
"font-family: 'Segoe UI', sans-serif",
"transition: all 0.2s ease",
"width: 120px",
"min-width: 120px",
"text-align: center",
"white-space: nowrap",
"overflow: hidden",
"text-overflow: ellipsis",
"box-sizing: border-box",
"vertical-align: middle",
"border: 0 !important",
"outline: 0 !important",
"box-shadow: none !important",
"position: relative",
"z-index: 1"
]
if token == '<|mdm_mask|>':
style.extend([
"color: transparent",
"background: #f8fafc",
"text-shadow: none"
])
display_text = "​"
else:
text_color = generate_background_color(token_steps[idx], max_step)
style.append(f"color: {text_color}")
display_text = token if token != " " else "␣"
html.append(f'''
<div style="display: inline-block; border: none !important; margin: 0 !important; padding: 0 !important;">
<span style="{"; ".join(style)}">{display_text}</span>
</div>
''')
return '\n'.join(html)
def main(target_step: int = 64):
"""Main function supporting target step specification"""
file_path = "sample_process.txt"
final_step = 64
history = parse_generation_history(file_path)
if target_step not in history:
raise ValueError(f"Invalid target step: {target_step}")
token_steps = track_token_positions(history)
current_tokens = history[target_step]
html_content = generate_step_visualization(
target_step, current_tokens, token_steps, final_step
)
example_steps = [0, 16, 32, 48, 64]
example_colors = [generate_background_color(s, final_step) for s in example_steps]
legend_html = ''.join(
f'<div style="background-color: {color}; color: black;">Step {s}</div>'
for s, color in zip(example_steps, example_colors)
)
target_dir = "html/sample_process_zhihu"
if not os.path.exists(target_dir):
os.makedirs(target_dir)
with open(f"{target_dir}/visualization_step_{target_step}.html", "w", encoding="utf-8") as f:
f.write(f"""<html>
<head>
<title>Step {target_step} Visualization</title>
<style>
body {{
padding: 40px;
background: #f8fafc;
font-family: 'Segoe UI', sans-serif;
}}
.legend {{
display: flex;
gap: 15px;
margin: 20px 0;
}}
.legend div {{
padding: 10px;
border-radius: 5px;
color: white;
min-width: 80px;
text-align: center;
}}
.token:hover {{
transform: translateY(-2px);
}}
</style>
</head>
<body>
<div style="max-width: 1000px; margin: auto;">
<div>{html_content}</div>
</div>
</body>
</html>""")
if __name__ == "__main__":
for step in range(1, 65):
main(target_step=step) |