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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
| import json
from datetime import datetime
from collections import Counter, defaultdict
import re
class GrokDataAnalyzer:
def __init__(self, json_file_path):
"""初始化分析器"""
with open(json_file_path, 'r', encoding='utf-8') as f:
self.data = json.load(f)
self.conversations = self.data.get('conversations', [])
def parse_timestamp(self, timestamp_data):
"""解析时间戳"""
if isinstance(timestamp_data, dict) and '$date' in timestamp_data:
date_data = timestamp_data['$date']
if isinstance(date_data, dict) and '$numberLong' in date_data:
timestamp_ms = int(date_data['$numberLong'])
return datetime.fromtimestamp(timestamp_ms / 1000)
elif isinstance(date_data, str):
return datetime.fromisoformat(date_data.replace('Z', '+00:00'))
elif isinstance(timestamp_data, str):
return datetime.fromisoformat(timestamp_data.replace('Z', '+00:00'))
return None
def analyze_basic_stats(self):
"""基础统计分析"""
total_conversations = len(self.conversations)
total_messages = 0
human_messages = 0
assistant_messages = 0
image_generations = 0
model_usage = Counter()
mode_usage = Counter()
date_stats = defaultdict(lambda: {'messages': 0, 'images': 0})
hour_stats = defaultdict(int)
for conv_data in self.conversations:
conversation = conv_data.get('conversation', {})
responses = conv_data.get('responses', [])
create_time_str = conversation.get('create_time')
if create_time_str:
conv_date = self.parse_timestamp(create_time_str)
if conv_date:
date_key = conv_date.strftime('%Y-%m-%d')
for resp in responses:
response_data = resp.get('response', {})
total_messages += 1
sender = response_data.get('sender', '').lower()
if 'human' in sender:
human_messages += 1
elif 'assistant' in sender:
assistant_messages += 1
model = response_data.get('model', 'unknown')
model_usage[model] += 1
metadata = response_data.get('metadata', {})
request_meta = metadata.get('request_metadata', {})
mode = request_meta.get('mode', 'default')
mode_usage[mode] += 1
if response_data.get('query_type') == 'imagine':
image_generations += 1
create_time = self.parse_timestamp(response_data.get('create_time'))
if create_time:
date_key = create_time.strftime('%Y-%m-%d')
date_stats[date_key]['messages'] += 1
if response_data.get('query_type') == 'imagine':
date_stats[date_key]['images'] += 1
hour_key = create_time.hour
hour_stats[hour_key] += 1
return {
'total_conversations': total_conversations,
'total_messages': total_messages,
'human_messages': human_messages,
'assistant_messages': assistant_messages,
'image_generations': image_generations,
'image_generation_rate': (image_generations / total_messages * 100) if total_messages > 0 else 0,
'model_usage': dict(model_usage),
'mode_usage': dict(mode_usage),
'date_stats': dict(date_stats),
'hour_stats': dict(hour_stats)
}
def analyze_image_generations(self):
"""分析图像生成情况"""
image_prompts = []
image_count_by_conv = defaultdict(int)
for conv_data in self.conversations:
conv_id = conv_data.get('conversation', {}).get('id', 'unknown')
responses = conv_data.get('responses', [])
for resp in responses:
response_data = resp.get('response', {})
if response_data.get('query_type') == 'imagine':
query = response_data.get('query', '')
image_prompts.append(query)
image_count_by_conv[conv_id] += 1
prompt_features = {
'total_prompts': len(image_prompts),
'avg_prompt_length': sum(len(p) for p in image_prompts) / len(image_prompts) if image_prompts else 0,
'common_keywords': self.extract_keywords(image_prompts),
'conversations_with_images': len(image_count_by_conv),
'images_per_conversation': dict(Counter(image_count_by_conv.values()))
}
return prompt_features
def extract_keywords(self, prompts, top_n=20):
"""从提示词中提取关键词"""
all_text = ' '.join(prompts).lower()
words = re.findall(r'\b[a-z]{3,}\b', all_text)
stop_words = {'the', 'and', 'with', 'for', 'this', 'that', 'are', 'was', 'were', 'has', 'have', 'had'}
filtered_words = [w for w in words if w not in stop_words]
word_counts = Counter(filtered_words)
return dict(word_counts.most_common(top_n))
def analyze_time_patterns(self):
"""分析时间模式"""
weekday_names = ['周一', '周二', '周三', '周四', '周五', '周六', '周日']
weekday_stats = {i: 0 for i in range(7)}
month_stats = {i: 0 for i in range(1, 13)}
time_slot_stats = {
'深夜 (0-6点)': 0,
'早晨 (6-9点)': 0,
'上午 (9-12点)': 0,
'中午 (12-14点)': 0,
'下午 (14-18点)': 0,
'晚上 (18-22点)': 0,
'深夜 (22-24点)': 0
}
for conv_data in self.conversations:
responses = conv_data.get('responses', [])
for resp in responses:
response_data = resp.get('response', {})
create_time = self.parse_timestamp(response_data.get('create_time'))
if create_time:
weekday = create_time.weekday()
weekday_stats[weekday] += 1
month = create_time.month
month_stats[month] += 1
hour = create_time.hour
if 0 <= hour < 6:
time_slot_stats['深夜 (0-6点)'] += 1
elif 6 <= hour < 9:
time_slot_stats['早晨 (6-9点)'] += 1
elif 9 <= hour < 12:
time_slot_stats['上午 (9-12点)'] += 1
elif 12 <= hour < 14:
time_slot_stats['中午 (12-14点)'] += 1
elif 14 <= hour < 18:
time_slot_stats['下午 (14-18点)'] += 1
elif 18 <= hour < 22:
time_slot_stats['晚上 (18-22点)'] += 1
else:
time_slot_stats['深夜 (22-24点)'] += 1
weekday_stats_named = {weekday_names[i]: weekday_stats[i] for i in range(7)}
return {
'weekday_stats': weekday_stats_named,
'month_stats': month_stats,
'time_slot_stats': time_slot_stats
}
def analyze_conversation_content(self):
"""分析对话内容特征"""
conversation_types = Counter()
title_keywords = []
for conv_data in self.conversations:
conversation = conv_data.get('conversation', {})
responses = conv_data.get('responses', [])
title = conversation.get('title', '').lower()
title_keywords.append(title)
has_images = any(
resp.get('response', {}).get('query_type') == 'imagine'
for resp in responses
)
if has_images:
conversation_types['图像生成'] += 1
else:
all_text = ' '.join([
resp.get('response', {}).get('message', '').lower()
for resp in responses
])
if any(keyword in all_text for keyword in ['代码', '编程', 'python']):
conversation_types['技术讨论'] += 1
elif any(keyword in all_text for keyword in ['解释', '什么', '为什么', '如何']):
conversation_types['知识问答'] += 1
elif any(keyword in all_text for keyword in ['创作', '故事', '诗歌']):
conversation_types['创意写作'] += 1
else:
conversation_types['普通对话'] += 1
all_titles = ' '.join(title_keywords)
title_words = re.findall(r'\b[a-z]{3,}\b', all_titles.lower())
title_word_counts = Counter(title_words)
return {
'conversation_types': dict(conversation_types),
'common_title_words': dict(title_word_counts.most_common(10))
}
def generate_report(self, output_file='grok_年度分析报告.md'):
"""生成分析报告"""
basic_stats = self.analyze_basic_stats()
image_analysis = self.analyze_image_generations()
time_patterns = self.analyze_time_patterns()
content_analysis = self.analyze_conversation_content()
with open(output_file, 'w', encoding='utf-8') as f:
f.write("# Grok 年度使用分析报告\n\n")
f.write(f"*报告生成时间: {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}*\n\n")
f.write("## 📊 总体概况\n\n")
f.write(f"- **总对话数**: {basic_stats['total_conversations']} 次\n")
f.write(f"- **总消息数**: {basic_stats['total_messages']} 条\n")
f.write(f"- **你的消息**: {basic_stats['human_messages']} 条\n")
f.write(f"- **AI回复**: {basic_stats['assistant_messages']} 条\n")
f.write(f"- **消息比例**: 你:{basic_stats['human_messages']} : {basic_stats['assistant_messages']} AI\n\n")
f.write("## 🎨 图像生成统计\n\n")
f.write(f"- **图像生成次数**: {basic_stats['image_generations']} 次\n")
f.write(f"- **图像生成率**: {basic_stats['image_generation_rate']:.1f}% (占所有消息比例)\n")
f.write(f"- **含图像的对话**: {image_analysis['conversations_with_images']} 个\n")
f.write(f"- **平均提示词长度**: {image_analysis['avg_prompt_length']:.0f} 字符\n\n")
f.write("## 🤖 模型使用情况\n\n")
for model, count in basic_stats['model_usage'].items():
f.write(f"- **{model}**: {count} 次 ({count/basic_stats['total_messages']*100:.1f}%)\n")
if basic_stats['mode_usage']:
f.write("\n### 模式使用\n")
for mode, count in basic_stats['mode_usage'].items():
if mode != 'default':
f.write(f"- **{mode}模式**: {count} 次\n")
f.write("\n## ⏰ 使用时间模式\n\n")
f.write("### 时段分布\n")
total_msgs = sum(time_patterns['time_slot_stats'].values())
for slot, count in time_patterns['time_slot_stats'].items():
if count > 0:
percentage = count / total_msgs * 100
bar = "█" * int(percentage / 2)
f.write(f"- {slot}: {bar} {count} ({percentage:.1f}%)\n")
f.write("\n### 星期分布\n")
for day, count in time_patterns['weekday_stats'].items():
f.write(f"- {day}: {count} 条\n")
f.write("\n## 💬 对话类型分布\n\n")
total_conv = sum(content_analysis['conversation_types'].values())
for conv_type, count in content_analysis['conversation_types'].items():
percentage = count / total_conv * 100 if total_conv > 0 else 0
f.write(f"- **{conv_type}**: {count} 次 ({percentage:.1f}%)\n")
if image_analysis['common_keywords']:
f.write("\n## 🔑 图像提示热门关键词\n\n")
for i, (keyword, count) in enumerate(image_analysis['common_keywords'].items(), 1):
f.write(f"{i}. **{keyword}**: {count} 次\n")
if content_analysis['common_title_words']:
f.write("\n## 🏷️ 对话标题关键词\n\n")
for word, count in content_analysis['common_title_words'].items():
f.write(f"- {word}: {count} 次\n")
f.write("\n## 💡 使用总结\n\n")
image_ratio = basic_stats['image_generation_rate']
if image_ratio > 30:
f.write("### 🎨 视觉创作者型\n")
f.write("- 主要使用Grok进行图像生成\n")
f.write("- 提示词详细，追求艺术效果\n")
f.write("- 视觉创意表达丰富\n")
elif '图像生成' in content_analysis['conversation_types']:
f.write("### 🎭 混合使用型\n")
f.write("- 兼顾图像生成与文本对话\n")
f.write("- 既有创意表达，也有实用查询\n")
f.write("- 使用场景多样化\n")
else:
f.write("### 💬 文本对话型\n")
f.write("- 主要进行文本交流\n")
f.write("- 关注知识获取与问题解决\n")
f.write("- 对话内容丰富\n")
top_time_slot = max(time_patterns['time_slot_stats'].items(), key=lambda x: x[1])[0]
f.write(f"\n### ⏰ 时间偏好: {top_time_slot}\n")
f.write("\n### 📝 基于样本数据的观察\n")
f.write("从提供的样本看，你的使用特点包括：\n")
f.write("- 详细的场景描述能力\n")
f.write("- 对视觉细节的高度关注\n")
f.write("- 喜欢设定特定情境（战时、浪漫等）\n")
f.write("- 使用专家模式追求高质量输出\n")
print(f"报告已生成: {output_file}")
return output_file
if __name__ == "__main__":
try:
print("正在分析Grok使用数据...")
analyzer = GrokDataAnalyzer('prod-grok-backend.json')
report_file = analyzer.generate_report()
print(f"\n✅ 分析完成！")
print(f"📄 报告文件: {report_file}")
stats = analyzer.analyze_basic_stats()
print(f"\n📊 简要统计:")
print(f" 对话总数: {stats['total_conversations']}")
print(f" 消息总数: {stats['total_messages']}")
print(f" 图像生成: {stats['image_generations']} 次")
print(f" 最常用模型: {max(stats['model_usage'].items(), key=lambda x: x[1])[0]}")
except FileNotFoundError:
print("❌ 错误: 找不到数据文件")
print("请确保 'grok_data.json' 文件在当前目录下")
except json.JSONDecodeError:
print("❌ 错误: 数据文件格式不正确")
except Exception as e:
print(f"❌ 分析过程中出现错误: {e}")
import traceback
traceback.print_exc()
|