ai-forever commited on
Commit
abf53d3
·
verified ·
1 Parent(s): ea6f3b0

Init results file.

Browse files
Files changed (1) hide show
  1. results.json +6 -263
results.json CHANGED
@@ -1,264 +1,7 @@
1
  {
2
- "items": {
3
- "0.1.1": {
4
- "4faf0f1f39744428a86b7571754e270a": {
5
- "model_name": "Gemma-3 4B",
6
- "timestamp": "2025-03-28T09:40:33",
7
- "config": {
8
- "embedding_model": "sentence-transformers/distiluse-base-multilingual-cased",
9
- "retriever_type": "mmr",
10
- "retrieval_config": {
11
- "top_k": 5,
12
- "chunk_size": 1000,
13
- "chunk_overlap": 100
14
- }
15
- },
16
- "metrics": {
17
- "cond": {
18
- "retrieval": {
19
- "hit_rate": 0.74,
20
- "mrr": 0.5920000000000001,
21
- "precision": 0.16799999999999998
22
- },
23
- "generation": {
24
- "rouge1": 0.14584126984126983,
25
- "rougeL": 0.14584126984126983
26
- }
27
- },
28
- "comp": {
29
- "retrieval": {
30
- "hit_rate": 0.72,
31
- "mrr": 0.4933333333333333,
32
- "precision": 0.2
33
- },
34
- "generation": {
35
- "rouge1": 0.17622510822510826,
36
- "rougeL": 0.17622510822510826
37
- }
38
- },
39
- "set": {
40
- "retrieval": {
41
- "hit_rate": 0.52,
42
- "mrr": 0.38066666666666665,
43
- "precision": 0.132
44
- },
45
- "generation": {
46
- "rouge1": 0.13877635775154523,
47
- "rougeL": 0.11396197213715961
48
- }
49
- },
50
- "mh": {
51
- "retrieval": {
52
- "hit_rate": 0.6,
53
- "mrr": 0.4073333333333333,
54
- "precision": 0.14800000000000002
55
- },
56
- "generation": {
57
- "rouge1": 0.047134453781512604,
58
- "rougeL": 0.047134453781512604
59
- }
60
- },
61
- "simple": {
62
- "retrieval": {
63
- "hit_rate": 0.5,
64
- "mrr": 0.3496666666666666,
65
- "precision": 0.12
66
- },
67
- "generation": {
68
- "rouge1": 0.08229166315355574,
69
- "rougeL": 0.08229166315355574
70
- }
71
- },
72
- "overall": {
73
- "retrieval": {
74
- "hit_rate": 0.616,
75
- "mrr": 0.4446,
76
- "precision": 0.15360000000000001
77
- },
78
- "generation": {
79
- "rouge1": 0.11805377055059833,
80
- "rougeL": 0.11309089342772122
81
- }
82
- }
83
- },
84
- "metadata": {
85
- "n_questions": 250,
86
- "submit_timestamp": ""
87
- }
88
- },
89
- "0ce57f9249744554a73e2c0bc1b54878": {
90
- "model_name": "Llama 3.3 70B",
91
- "timestamp": "2025-03-28T09:42:36",
92
- "config": {
93
- "embedding_model": "sentence-transformers/distiluse-base-multilingual-cased",
94
- "retriever_type": "mmr",
95
- "retrieval_config": {
96
- "top_k": 5,
97
- "chunk_size": 1000,
98
- "chunk_overlap": 100
99
- }
100
- },
101
- "metrics": {
102
- "cond": {
103
- "retrieval": {
104
- "hit_rate": 0.64,
105
- "mrr": 0.514,
106
- "precision": 0.14400000000000002
107
- },
108
- "generation": {
109
- "rouge1": 0.2493333333333333,
110
- "rougeL": 0.2493333333333333
111
- }
112
- },
113
- "comp": {
114
- "retrieval": {
115
- "hit_rate": 0.72,
116
- "mrr": 0.4933333333333333,
117
- "precision": 0.2
118
- },
119
- "generation": {
120
- "rouge1": 0.23564204559660382,
121
- "rougeL": 0.23564204559660382
122
- }
123
- },
124
- "set": {
125
- "retrieval": {
126
- "hit_rate": 0.5,
127
- "mrr": 0.3683333333333333,
128
- "precision": 0.12400000000000003
129
- },
130
- "generation": {
131
- "rouge1": 0.21594023633499185,
132
- "rougeL": 0.17318924819269935
133
- }
134
- },
135
- "mh": {
136
- "retrieval": {
137
- "hit_rate": 0.58,
138
- "mrr": 0.3846666666666666,
139
- "precision": 0.14800000000000002
140
- },
141
- "generation": {
142
- "rouge1": 0.19785714285714284,
143
- "rougeL": 0.19785714285714284
144
- }
145
- },
146
- "simple": {
147
- "retrieval": {
148
- "hit_rate": 0.48,
149
- "mrr": 0.33399999999999996,
150
- "precision": 0.12
151
- },
152
- "generation": {
153
- "rouge1": 0.16162049062049064,
154
- "rougeL": 0.16162049062049064
155
- }
156
- },
157
- "overall": {
158
- "retrieval": {
159
- "hit_rate": 0.584,
160
- "mrr": 0.41886666666666666,
161
- "precision": 0.1472
162
- },
163
- "generation": {
164
- "rouge1": 0.21207864974851248,
165
- "rougeL": 0.20352845212005397
166
- }
167
- }
168
- },
169
- "metadata": {
170
- "n_questions": 250,
171
- "submit_timestamp": ""
172
- }
173
- },
174
- "6edbacf9226e463e9e240ca8219d72b8": {
175
- "model_name": "Gemma-3 12B",
176
- "timestamp": "2025-03-28T09:43:27",
177
- "config": {
178
- "embedding_model": "sentence-transformers/distiluse-base-multilingual-cased",
179
- "retriever_type": "mmr",
180
- "retrieval_config": {
181
- "top_k": 5,
182
- "chunk_size": 1000,
183
- "chunk_overlap": 100
184
- }
185
- },
186
- "metrics": {
187
- "cond": {
188
- "retrieval": {
189
- "hit_rate": 0.74,
190
- "mrr": 0.5920000000000001,
191
- "precision": 0.16799999999999998
192
- },
193
- "generation": {
194
- "rouge1": 0.11885237343301858,
195
- "rougeL": 0.11885237343301858
196
- }
197
- },
198
- "comp": {
199
- "retrieval": {
200
- "hit_rate": 0.72,
201
- "mrr": 0.4933333333333333,
202
- "precision": 0.2
203
- },
204
- "generation": {
205
- "rouge1": 0.08331701939804853,
206
- "rougeL": 0.08331701939804853
207
- }
208
- },
209
- "set": {
210
- "retrieval": {
211
- "hit_rate": 0.52,
212
- "mrr": 0.38066666666666665,
213
- "precision": 0.132
214
- },
215
- "generation": {
216
- "rouge1": 0.09937513859436432,
217
- "rougeL": 0.0802916087780759
218
- }
219
- },
220
- "mh": {
221
- "retrieval": {
222
- "hit_rate": 0.6,
223
- "mrr": 0.4073333333333333,
224
- "precision": 0.14800000000000002
225
- },
226
- "generation": {
227
- "rouge1": 0.0643998533286459,
228
- "rougeL": 0.0643998533286459
229
- }
230
- },
231
- "simple": {
232
- "retrieval": {
233
- "hit_rate": 0.5,
234
- "mrr": 0.3496666666666666,
235
- "precision": 0.12
236
- },
237
- "generation": {
238
- "rouge1": 0.05821997465191799,
239
- "rougeL": 0.05821997465191799
240
- }
241
- },
242
- "overall": {
243
- "retrieval": {
244
- "hit_rate": 0.616,
245
- "mrr": 0.4446,
246
- "precision": 0.15360000000000001
247
- },
248
- "generation": {
249
- "rouge1": 0.08483287188119908,
250
- "rougeL": 0.08101616591794138
251
- }
252
- }
253
- },
254
- "metadata": {
255
- "n_questions": 250,
256
- "submit_timestamp": ""
257
- }
258
- }
259
- }
260
- },
261
- "last_version": "0.1.1",
262
- "n_questions": 250,
263
- "date_title": "28 \u043c\u0430\u0440\u0442\u0430 2025"
264
- }
 
1
  {
2
+ "items": {
3
+ },
4
+ "last_version": "1.0",
5
+ "n_questions": "11",
6
+ "date_title": "22 марта 2025"
7
+ }