Hugging Face's logo

EvanDu
/
TSRD_rram_classification

Model card Files
xet
 Community
EvanDu commited on
Commit
c9d6a0b
·
verified ·
1 Parent(s): 273cbb6

Upload 3 files

Browse files
Files changed (3) hide show
  1. checkpoints_v2m_part2/2025-04-11_14-02-06_train.log +734 -0
  2. checkpoints_v2m_part2/base_training_metrics.csv +51 -0
  3. checkpoints_v2m_part2/fine_tuning_metrics.csv +51 -0
checkpoints_v2m_part2/2025-04-11_14-02-06_train.log ADDED
@@ -0,0 +1,734 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [2025-04-11 14:02:06] [INFO] 使用设备: cuda:0
2
+ [2025-04-11 14:02:06] [INFO] 训练集注释文件: /data0/work/DuYiFan/projects/traffic_classify/4_directions/TsignRecgTrainAnnotation.txt
3
+ [2025-04-11 14:02:06] [INFO] 测试集注释文件: /data0/work/DuYiFan/projects/traffic_classify/4_directions/TsignRecgTestAnnotation.txt
4
+ [2025-04-11 14:02:06] [INFO] 训练图像目录: /data0/work/DuYiFan/projects/traffic_classify/4_directions/train
5
+ [2025-04-11 14:02:06] [INFO] 测试图像目录: /data0/work/DuYiFan/projects/traffic_classify/4_directions/test
6
+ [2025-04-11 14:02:06] [INFO] 创建数据集和数据加载器
7
+ [2025-04-11 14:02:06] [INFO] 创建efficientnet-v2-m模型，类别数: 4
8
+ [2025-04-11 14:02:07] [INFO] 设置损失函数、优化器和学习率调度器，初始学习率: 0.0001
9
+ [2025-04-11 14:02:07] [INFO] 开始训练，总共 50 轮
10
+ [2025-04-11 14:02:07] [INFO] 当前学习率: 0.000100
11
+ [2025-04-11 14:02:07] [INFO] Epoch 1/50 开始训练
12
+ [2025-04-11 14:02:09] [INFO] Epoch 1/50 开始验证
13
+ [2025-04-11 14:02:09] [INFO] Epoch 1/50 - Train Loss: 1.3665, Train Acc: 0.3803, Val Loss: 1.3432, Val Acc: 0.4483
14
+ [2025-04-11 14:02:10] [INFO] 已保存最佳模型，准确率: 0.4483
15
+ [2025-04-11 14:02:11] [INFO] 当前学习率: 0.000100
16
+ [2025-04-11 14:02:11] [INFO] Epoch 2/50 开始训练
17
+ [2025-04-11 14:02:12] [INFO] Epoch 2/50 开始验证
18
+ [2025-04-11 14:02:12] [INFO] Epoch 2/50 - Train Loss: 1.2475, Train Acc: 0.6690, Val Loss: 1.3103, Val Acc: 0.4483
19
+ [2025-04-11 14:02:13] [INFO] 当前学习率: 0.000100
20
+ [2025-04-11 14:02:13] [INFO] Epoch 3/50 开始训练
21
+ [2025-04-11 14:02:14] [INFO] Epoch 3/50 开始验证
22
+ [2025-04-11 14:02:14] [INFO] Epoch 3/50 - Train Loss: 1.1308, Train Acc: 0.7042, Val Loss: 1.2931, Val Acc: 0.4483
23
+ [2025-04-11 14:02:15] [INFO] 当前学习率: 0.000099
24
+ [2025-04-11 14:02:15] [INFO] Epoch 4/50 开始训练
25
+ [2025-04-11 14:02:16] [INFO] Epoch 4/50 开始验证
26
+ [2025-04-11 14:02:17] [INFO] Epoch 4/50 - Train Loss: 1.0179, Train Acc: 0.7042, Val Loss: 1.2997, Val Acc: 0.4483
27
+ [2025-04-11 14:02:18] [INFO] 当前学习率: 0.000098
28
+ [2025-04-11 14:02:18] [INFO] Epoch 5/50 开始训练
29
+ [2025-04-11 14:02:19] [INFO] Epoch 5/50 开始验证
30
+ [2025-04-11 14:02:19] [INFO] Epoch 5/50 - Train Loss: 0.9342, Train Acc: 0.7042, Val Loss: 1.3250, Val Acc: 0.4483
31
+ [2025-04-11 14:02:20] [INFO] 当前学习率: 0.000098
32
+ [2025-04-11 14:02:20] [INFO] Epoch 6/50 开始训练
33
+ [2025-04-11 14:02:21] [INFO] Epoch 6/50 开始验证
34
+ [2025-04-11 14:02:21] [INFO] Epoch 6/50 - Train Loss: 0.8822, Train Acc: 0.7042, Val Loss: 1.3368, Val Acc: 0.4483
35
+ [2025-04-11 14:02:22] [INFO] 当前学习率: 0.000097
36
+ [2025-04-11 14:02:22] [INFO] Epoch 7/50 开始训练
37
+ [2025-04-11 14:02:23] [INFO] Epoch 7/50 开始验证
38
+ [2025-04-11 14:02:24] [INFO] Epoch 7/50 - Train Loss: 0.8348, Train Acc: 0.7042, Val Loss: 1.3392, Val Acc: 0.4483
39
+ [2025-04-11 14:02:25] [INFO] 当前学习率: 0.000095
40
+ [2025-04-11 14:02:25] [INFO] Epoch 8/50 开始训练
41
+ [2025-04-11 14:02:26] [INFO] Epoch 8/50 开始验证
42
+ [2025-04-11 14:02:26] [INFO] Epoch 8/50 - Train Loss: 0.7929, Train Acc: 0.7042, Val Loss: 1.3446, Val Acc: 0.4483
43
+ [2025-04-11 14:02:27] [INFO] 当前学习率: 0.000094
44
+ [2025-04-11 14:02:27] [INFO] Epoch 9/50 开始训练
45
+ [2025-04-11 14:02:28] [INFO] Epoch 9/50 开始验证
46
+ [2025-04-11 14:02:28] [INFO] Epoch 9/50 - Train Loss: 0.7403, Train Acc: 0.7042, Val Loss: 1.4892, Val Acc: 0.4483
47
+ [2025-04-11 14:02:29] [INFO] 当前学习率: 0.000092
48
+ [2025-04-11 14:02:29] [INFO] Epoch 10/50 开始训练
49
+ [2025-04-11 14:02:30] [INFO] Epoch 10/50 开始验证
50
+ [2025-04-11 14:02:31] [INFO] Epoch 10/50 - Train Loss: 0.6734, Train Acc: 0.7113, Val Loss: 1.7414, Val Acc: 0.4483
51
+ [2025-04-11 14:02:32] [INFO] 当前学习率: 0.000091
52
+ [2025-04-11 14:02:32] [INFO] Epoch 11/50 开始训练
53
+ [2025-04-11 14:02:33] [INFO] Epoch 11/50 开始验证
54
+ [2025-04-11 14:02:33] [INFO] Epoch 11/50 - Train Loss: 0.6278, Train Acc: 0.7183, Val Loss: 1.2002, Val Acc: 0.4828
55
+ [2025-04-11 14:02:33] [INFO] 已保存最佳模型，准确率: 0.4828
56
+ [2025-04-11 14:02:34] [INFO] 当前学习率: 0.000089
57
+ [2025-04-11 14:02:34] [INFO] Epoch 12/50 开始训练
58
+ [2025-04-11 14:02:35] [INFO] Epoch 12/50 开始验证
59
+ [2025-04-11 14:02:36] [INFO] Epoch 12/50 - Train Loss: 0.5761, Train Acc: 0.7324, Val Loss: 1.1805, Val Acc: 0.4483
60
+ [2025-04-11 14:02:37] [INFO] 当前学习率: 0.000087
61
+ [2025-04-11 14:02:37] [INFO] Epoch 13/50 开始训练
62
+ [2025-04-11 14:02:38] [INFO] Epoch 13/50 开始验证
63
+ [2025-04-11 14:02:38] [INFO] Epoch 13/50 - Train Loss: 0.5605, Train Acc: 0.7606, Val Loss: 1.2032, Val Acc: 0.5862
64
+ [2025-04-11 14:02:38] [INFO] 已保存最佳模型，准确率: 0.5862
65
+ [2025-04-11 14:02:39] [INFO] 当前学习率: 0.000084
66
+ [2025-04-11 14:02:39] [INFO] Epoch 14/50 开始训练
67
+ [2025-04-11 14:02:41] [INFO] Epoch 14/50 开始验证
68
+ [2025-04-11 14:02:41] [INFO] Epoch 14/50 - Train Loss: 0.5306, Train Acc: 0.7887, Val Loss: 1.5191, Val Acc: 0.4483
69
+ [2025-04-11 14:02:42] [INFO] 当前学习率: 0.000082
70
+ [2025-04-11 14:02:42] [INFO] Epoch 15/50 开始训练
71
+ [2025-04-11 14:02:43] [INFO] Epoch 15/50 开始验证
72
+ [2025-04-11 14:02:43] [INFO] Epoch 15/50 - Train Loss: 0.4984, Train Acc: 0.8028, Val Loss: 1.1684, Val Acc: 0.6207
73
+ [2025-04-11 14:02:44] [INFO] 已保存最佳模型，准确率: 0.6207
74
+ [2025-04-11 14:02:45] [INFO] 当前学习率: 0.000080
75
+ [2025-04-11 14:02:45] [INFO] Epoch 16/50 开始训练
76
+ [2025-04-11 14:02:46] [INFO] Epoch 16/50 开始验证
77
+ [2025-04-11 14:02:46] [INFO] Epoch 16/50 - Train Loss: 0.4516, Train Acc: 0.8380, Val Loss: 1.1488, Val Acc: 0.3793
78
+ [2025-04-11 14:02:47] [INFO] 当前学习率: 0.000077
79
+ [2025-04-11 14:02:47] [INFO] Epoch 17/50 开始训练
80
+ [2025-04-11 14:02:48] [INFO] Epoch 17/50 开始验证
81
+ [2025-04-11 14:02:48] [INFO] Epoch 17/50 - Train Loss: 0.4322, Train Acc: 0.8310, Val Loss: 1.1146, Val Acc: 0.4138
82
+ [2025-04-11 14:02:49] [INFO] 当前学习率: 0.000074
83
+ [2025-04-11 14:02:49] [INFO] Epoch 18/50 开始训练
84
+ [2025-04-11 14:02:50] [INFO] Epoch 18/50 开始验证
85
+ [2025-04-11 14:02:51] [INFO] Epoch 18/50 - Train Loss: 0.3943, Train Acc: 0.8662, Val Loss: 1.1947, Val Acc: 0.4483
86
+ [2025-04-11 14:02:52] [INFO] 当前学习率: 0.000072
87
+ [2025-04-11 14:02:52] [INFO] Epoch 19/50 开始训练
88
+ [2025-04-11 14:02:53] [INFO] Epoch 19/50 开始验证
89
+ [2025-04-11 14:02:53] [INFO] Epoch 19/50 - Train Loss: 0.3648, Train Acc: 0.8873, Val Loss: 1.1186, Val Acc: 0.4138
90
+ [2025-04-11 14:02:54] [INFO] 当前学习率: 0.000069
91
+ [2025-04-11 14:02:54] [INFO] Epoch 20/50 开始训练
92
+ [2025-04-11 14:02:55] [INFO] Epoch 20/50 开始验证
93
+ [2025-04-11 14:02:55] [INFO] Epoch 20/50 - Train Loss: 0.3696, Train Acc: 0.9085, Val Loss: 1.0720, Val Acc: 0.6552
94
+ [2025-04-11 14:02:56] [INFO] 已保存最佳模型，准确率: 0.6552
95
+ [2025-04-11 14:02:57] [INFO] 当前学习率: 0.000066
96
+ [2025-04-11 14:02:57] [INFO] Epoch 21/50 开始训练
97
+ [2025-04-11 14:02:58] [INFO] Epoch 21/50 开始验证
98
+ [2025-04-11 14:02:58] [INFO] Epoch 21/50 - Train Loss: 0.3843, Train Acc: 0.8521, Val Loss: 1.0828, Val Acc: 0.5862
99
+ [2025-04-11 14:02:59] [INFO] 当前学习率: 0.000063
100
+ [2025-04-11 14:02:59] [INFO] Epoch 22/50 开始训练
101
+ [2025-04-11 14:03:00] [INFO] Epoch 22/50 开始验证
102
+ [2025-04-11 14:03:00] [INFO] Epoch 22/50 - Train Loss: 0.3440, Train Acc: 0.8944, Val Loss: 1.0306, Val Acc: 0.4483
103
+ [2025-04-11 14:03:02] [INFO] 当前学习率: 0.000060
104
+ [2025-04-11 14:03:02] [INFO] Epoch 23/50 开始训练
105
+ [2025-04-11 14:03:03] [INFO] Epoch 23/50 开始验证
106
+ [2025-04-11 14:03:03] [INFO] Epoch 23/50 - Train Loss: 0.3167, Train Acc: 0.9085, Val Loss: 1.0836, Val Acc: 0.5172
107
+ [2025-04-11 14:03:04] [INFO] 当前学习率: 0.000057
108
+ [2025-04-11 14:03:04] [INFO] Epoch 24/50 开始训练
109
+ [2025-04-11 14:03:05] [INFO] Epoch 24/50 开始验证
110
+ [2025-04-11 14:03:05] [INFO] Epoch 24/50 - Train Loss: 0.3757, Train Acc: 0.8662, Val Loss: 1.0405, Val Acc: 0.5172
111
+ [2025-04-11 14:03:06] [INFO] 当前学习率: 0.000054
112
+ [2025-04-11 14:03:06] [INFO] Epoch 25/50 开始训练
113
+ [2025-04-11 14:03:07] [INFO] Epoch 25/50 开始验证
114
+ [2025-04-11 14:03:08] [INFO] Epoch 25/50 - Train Loss: 0.2902, Train Acc: 0.8803, Val Loss: 1.0648, Val Acc: 0.4828
115
+ [2025-04-11 14:03:09] [INFO] 当前学习率: 0.000050
116
+ [2025-04-11 14:03:09] [INFO] Epoch 26/50 开始训练
117
+ [2025-04-11 14:03:10] [INFO] Epoch 26/50 开始验证
118
+ [2025-04-11 14:03:10] [INFO] Epoch 26/50 - Train Loss: 0.3106, Train Acc: 0.8873, Val Loss: 1.0448, Val Acc: 0.5862
119
+ [2025-04-11 14:03:11] [INFO] 当前学习率: 0.000047
120
+ [2025-04-11 14:03:11] [INFO] Epoch 27/50 开始训练
121
+ [2025-04-11 14:03:12] [INFO] Epoch 27/50 开始验证
122
+ [2025-04-11 14:03:12] [INFO] Epoch 27/50 - Train Loss: 0.3001, Train Acc: 0.9014, Val Loss: 1.0292, Val Acc: 0.5172
123
+ [2025-04-11 14:03:13] [INFO] 当前学习率: 0.000044
124
+ [2025-04-11 14:03:13] [INFO] Epoch 28/50 开始训练
125
+ [2025-04-11 14:03:14] [INFO] Epoch 28/50 开始验证
126
+ [2025-04-11 14:03:15] [INFO] Epoch 28/50 - Train Loss: 0.3327, Train Acc: 0.8803, Val Loss: 0.9527, Val Acc: 0.5517
127
+ [2025-04-11 14:03:15] [INFO] 当前学习率: 0.000041
128
+ [2025-04-11 14:03:15] [INFO] Epoch 29/50 开始训练
129
+ [2025-04-11 14:03:16] [INFO] Epoch 29/50 开始验证
130
+ [2025-04-11 14:03:17] [INFO] Epoch 29/50 - Train Loss: 0.3186, Train Acc: 0.9225, Val Loss: 1.0192, Val Acc: 0.5517
131
+ [2025-04-11 14:03:18] [INFO] 当前学习率: 0.000038
132
+ [2025-04-11 14:03:18] [INFO] Epoch 30/50 开始训练
133
+ [2025-04-11 14:03:19] [INFO] Epoch 30/50 开始验证
134
+ [2025-04-11 14:03:19] [INFO] Epoch 30/50 - Train Loss: 0.2801, Train Acc: 0.9366, Val Loss: 0.9282, Val Acc: 0.5172
135
+ [2025-04-11 14:03:20] [INFO] 当前学习率: 0.000035
136
+ [2025-04-11 14:03:20] [INFO] Epoch 31/50 开始训练
137
+ [2025-04-11 14:03:21] [INFO] Epoch 31/50 开始验证
138
+ [2025-04-11 14:03:21] [INFO] Epoch 31/50 - Train Loss: 0.2773, Train Acc: 0.9155, Val Loss: 0.9423, Val Acc: 0.5862
139
+ [2025-04-11 14:03:23] [INFO] 当前学习率: 0.000032
140
+ [2025-04-11 14:03:23] [INFO] Epoch 32/50 开始训练
141
+ [2025-04-11 14:03:24] [INFO] Epoch 32/50 开始验证
142
+ [2025-04-11 14:03:24] [INFO] Epoch 32/50 - Train Loss: 0.2920, Train Acc: 0.9014, Val Loss: 1.0064, Val Acc: 0.5862
143
+ [2025-04-11 14:03:25] [INFO] 当前学习率: 0.000029
144
+ [2025-04-11 14:03:25] [INFO] Epoch 33/50 开始训练
145
+ [2025-04-11 14:03:26] [INFO] Epoch 33/50 开始验证
146
+ [2025-04-11 14:03:26] [INFO] Epoch 33/50 - Train Loss: 0.2702, Train Acc: 0.9296, Val Loss: 0.9927, Val Acc: 0.6552
147
+ [2025-04-11 14:03:27] [INFO] 当前学习率: 0.000027
148
+ [2025-04-11 14:03:27] [INFO] Epoch 34/50 开始训练
149
+ [2025-04-11 14:03:28] [INFO] Epoch 34/50 开始验证
150
+ [2025-04-11 14:03:29] [INFO] Epoch 34/50 - Train Loss: 0.2532, Train Acc: 0.9225, Val Loss: 1.0585, Val Acc: 0.6552
151
+ [2025-04-11 14:03:30] [INFO] 当前学习率: 0.000024
152
+ [2025-04-11 14:03:30] [INFO] Epoch 35/50 开始训练
153
+ [2025-04-11 14:03:31] [INFO] Epoch 35/50 开始验证
154
+ [2025-04-11 14:03:31] [INFO] Epoch 35/50 - Train Loss: 0.2324, Train Acc: 0.9577, Val Loss: 1.0661, Val Acc: 0.6207
155
+ [2025-04-11 14:03:32] [INFO] 当前学习率: 0.000021
156
+ [2025-04-11 14:03:32] [INFO] Epoch 36/50 开始训练
157
+ [2025-04-11 14:03:33] [INFO] Epoch 36/50 开始验证
158
+ [2025-04-11 14:03:33] [INFO] Epoch 36/50 - Train Loss: 0.2451, Train Acc: 0.9085, Val Loss: 0.9552, Val Acc: 0.6552
159
+ [2025-04-11 14:03:34] [INFO] 当前学习率: 0.000019
160
+ [2025-04-11 14:03:34] [INFO] Epoch 37/50 开始训练
161
+ [2025-04-11 14:03:35] [INFO] Epoch 37/50 开始验证
162
+ [2025-04-11 14:03:36] [INFO] Epoch 37/50 - Train Loss: 0.2731, Train Acc: 0.9437, Val Loss: 0.9102, Val Acc: 0.6897
163
+ [2025-04-11 14:03:36] [INFO] 已保存最佳模型，准确率: 0.6897
164
+ [2025-04-11 14:03:37] [INFO] 当前学习率: 0.000017
165
+ [2025-04-11 14:03:37] [INFO] Epoch 38/50 开始训练
166
+ [2025-04-11 14:03:38] [INFO] Epoch 38/50 开始验证
167
+ [2025-04-11 14:03:38] [INFO] Epoch 38/50 - Train Loss: 0.2497, Train Acc: 0.9437, Val Loss: 0.8856, Val Acc: 0.6552
168
+ [2025-04-11 14:03:39] [INFO] 当前学习率: 0.000014
169
+ [2025-04-11 14:03:39] [INFO] Epoch 39/50 开始训练
170
+ [2025-04-11 14:03:40] [INFO] Epoch 39/50 开始验证
171
+ [2025-04-11 14:03:41] [INFO] Epoch 39/50 - Train Loss: 0.2333, Train Acc: 0.9648, Val Loss: 0.8399, Val Acc: 0.6897
172
+ [2025-04-11 14:03:42] [INFO] 当前学习率: 0.000012
173
+ [2025-04-11 14:03:42] [INFO] Epoch 40/50 开始训练
174
+ [2025-04-11 14:03:43] [INFO] Epoch 40/50 开始验证
175
+ [2025-04-11 14:03:43] [INFO] Epoch 40/50 - Train Loss: 0.2431, Train Acc: 0.9507, Val Loss: 0.8593, Val Acc: 0.6552
176
+ [2025-04-11 14:03:44] [INFO] 当前学习率: 0.000010
177
+ [2025-04-11 14:03:44] [INFO] Epoch 41/50 开始训练
178
+ [2025-04-11 14:03:45] [INFO] Epoch 41/50 开始验证
179
+ [2025-04-11 14:03:46] [INFO] Epoch 41/50 - Train Loss: 0.2487, Train Acc: 0.9507, Val Loss: 0.8634, Val Acc: 0.6207
180
+ [2025-04-11 14:03:47] [INFO] 当前学习率: 0.000009
181
+ [2025-04-11 14:03:47] [INFO] Epoch 42/50 开始训练
182
+ [2025-04-11 14:03:48] [INFO] Epoch 42/50 开始验证
183
+ [2025-04-11 14:03:48] [INFO] Epoch 42/50 - Train Loss: 0.2074, Train Acc: 0.9718, Val Loss: 0.8873, Val Acc: 0.6897
184
+ [2025-04-11 14:03:49] [INFO] 当前学习率: 0.000007
185
+ [2025-04-11 14:03:49] [INFO] Epoch 43/50 开始训练
186
+ [2025-04-11 14:03:50] [INFO] Epoch 43/50 开始验证
187
+ [2025-04-11 14:03:50] [INFO] Epoch 43/50 - Train Loss: 0.2321, Train Acc: 0.9437, Val Loss: 0.8828, Val Acc: 0.6207
188
+ [2025-04-11 14:03:51] [INFO] 当前学习率: 0.000006
189
+ [2025-04-11 14:03:51] [INFO] Epoch 44/50 开始训练
190
+ [2025-04-11 14:03:52] [INFO] Epoch 44/50 开始验证
191
+ [2025-04-11 14:03:53] [INFO] Epoch 44/50 - Train Loss: 0.2352, Train Acc: 0.9648, Val Loss: 0.8669, Val Acc: 0.6552
192
+ [2025-04-11 14:03:54] [INFO] 当前学习率: 0.000004
193
+ [2025-04-11 14:03:54] [INFO] Epoch 45/50 开始训练
194
+ [2025-04-11 14:03:55] [INFO] Epoch 45/50 开始验证
195
+ [2025-04-11 14:03:55] [INFO] Epoch 45/50 - Train Loss: 0.2563, Train Acc: 0.9648, Val Loss: 0.8832, Val Acc: 0.6897
196
+ [2025-04-11 14:03:56] [INFO] 当前学习率: 0.000003
197
+ [2025-04-11 14:03:56] [INFO] Epoch 46/50 开始训练
198
+ [2025-04-11 14:03:57] [INFO] Epoch 46/50 开始验证
199
+ [2025-04-11 14:03:57] [INFO] Epoch 46/50 - Train Loss: 0.2093, Train Acc: 0.9859, Val Loss: 0.8795, Val Acc: 0.6552
200
+ [2025-04-11 14:03:58] [INFO] 当前学习率: 0.000003
201
+ [2025-04-11 14:03:58] [INFO] Epoch 47/50 开始训练
202
+ [2025-04-11 14:03:59] [INFO] Epoch 47/50 开始验证
203
+ [2025-04-11 14:04:00] [INFO] Epoch 47/50 - Train Loss: 0.2569, Train Acc: 0.9437, Val Loss: 0.9025, Val Acc: 0.7241
204
+ [2025-04-11 14:04:00] [INFO] 已保存最佳模型，准确率: 0.7241
205
+ [2025-04-11 14:04:01] [INFO] 当前学习率: 0.000002
206
+ [2025-04-11 14:04:01] [INFO] Epoch 48/50 开始训练
207
+ [2025-04-11 14:04:02] [INFO] Epoch 48/50 开始验证
208
+ [2025-04-11 14:04:02] [INFO] Epoch 48/50 - Train Loss: 0.2156, Train Acc: 0.9648, Val Loss: 0.9006, Val Acc: 0.7241
209
+ [2025-04-11 14:04:04] [INFO] 当前学习率: 0.000001
210
+ [2025-04-11 14:04:04] [INFO] Epoch 49/50 开始训练
211
+ [2025-04-11 14:04:05] [INFO] Epoch 49/50 开始验证
212
+ [2025-04-11 14:04:05] [INFO] Epoch 49/50 - Train Loss: 0.2208, Train Acc: 0.9648, Val Loss: 0.9028, Val Acc: 0.6552
213
+ [2025-04-11 14:04:06] [INFO] 当前学习率: 0.000001
214
+ [2025-04-11 14:04:06] [INFO] Epoch 50/50 开始训练
215
+ [2025-04-11 14:04:07] [INFO] Epoch 50/50 开始验证
216
+ [2025-04-11 14:04:07] [INFO] Epoch 50/50 - Train Loss: 0.2216, Train Acc: 0.9577, Val Loss: 0.8998, Val Acc: 0.6552
217
+ [2025-04-11 14:04:08] [INFO] 绘制训练过程图表
218
+ [2025-04-11 14:04:09] [INFO] 标准训练完成!
219
+ [2025-04-11 14:04:09] [INFO] 评估原始模型性能...
220
+ [2025-04-11 14:04:09] [INFO] 评估结果 - Loss: 0.8998, Accuracy: 0.6552
221
+ [2025-04-11 14:04:09] [INFO] 开始执行RRAM映射...
222
+ [2025-04-11 14:04:09] [INFO] 加载了 100 个RRAM电导值
223
+ [2025-04-11 14:04:09] [INFO] features.0.0.weight 的平均映射误差: 0.018842
224
+ [2025-04-11 14:04:09] [INFO] features.0.1.weight 的平均映射误差: 0.033772
225
+ [2025-04-11 14:04:09] [INFO] features.1.0.block.0.0.weight 的平均映射误差: 0.005888
226
+ [2025-04-11 14:04:09] [INFO] features.1.0.block.0.1.weight 的平均映射误差: 0.035463
227
+ [2025-04-11 14:04:09] [INFO] features.1.1.block.0.0.weight 的平均映射误差: 0.004058
228
+ [2025-04-11 14:04:09] [INFO] features.1.1.block.0.1.weight 的平均映射误差: 0.034065
229
+ [2025-04-11 14:04:09] [INFO] features.1.2.block.0.0.weight 的平均映射误差: 0.003652
230
+ [2025-04-11 14:04:09] [INFO] features.1.2.block.0.1.weight 的平均映射误差: 0.035255
231
+ [2025-04-11 14:04:09] [INFO] features.2.0.block.0.0.weight 的平均映射误差: 0.003249
232
+ [2025-04-11 14:04:09] [INFO] features.2.0.block.0.1.weight 的平均映射误差: 0.035290
233
+ [2025-04-11 14:04:09] [INFO] features.2.0.block.1.0.weight 的平均映射误差: 0.006492
234
+ [2025-04-11 14:04:09] [INFO] features.2.0.block.1.1.weight 的平均映射误差: 0.035411
235
+ [2025-04-11 14:04:09] [INFO] features.2.1.block.0.0.weight 的平均映射误差: 0.001784
236
+ [2025-04-11 14:04:09] [INFO] features.2.1.block.0.1.weight 的平均映射误差: 0.035374
237
+ [2025-04-11 14:04:09] [INFO] features.2.1.block.1.0.weight 的平均映射误差: 0.003037
238
+ [2025-04-11 14:04:09] [INFO] features.2.1.block.1.1.weight 的平均映射误差: 0.036986
239
+ [2025-04-11 14:04:09] [INFO] features.2.2.block.0.0.weight 的平均映射误差: 0.001778
240
+ [2025-04-11 14:04:09] [INFO] features.2.2.block.0.1.weight 的平均映射误差: 0.035899
241
+ [2025-04-11 14:04:09] [INFO] features.2.2.block.1.0.weight 的平均映射误差: 0.002751
242
+ [2025-04-11 14:04:09] [INFO] features.2.2.block.1.1.weight 的平均映射误差: 0.035042
243
+ [2025-04-11 14:04:09] [INFO] features.2.3.block.0.0.weight 的平均映射误差: 0.001793
244
+ [2025-04-11 14:04:09] [INFO] features.2.3.block.0.1.weight 的平均映射误差: 0.036626
245
+ [2025-04-11 14:04:09] [INFO] features.2.3.block.1.0.weight 的平均映射误差: 0.002673
246
+ [2025-04-11 14:04:09] [INFO] features.2.3.block.1.1.weight 的平均映射误差: 0.034752
247
+ [2025-04-11 14:04:09] [INFO] features.2.4.block.0.0.weight 的平均映射误差: 0.001805
248
+ [2025-04-11 14:04:09] [INFO] features.2.4.block.0.1.weight 的平均映射误差: 0.038931
249
+ [2025-04-11 14:04:09] [INFO] features.2.4.block.1.0.weight 的平均映射误差: 0.002595
250
+ [2025-04-11 14:04:09] [INFO] features.2.4.block.1.1.weight 的平均映射误差: 0.036464
251
+ [2025-04-11 14:04:09] [INFO] features.3.0.block.0.0.weight 的平均映射误差: 0.002078
252
+ [2025-04-11 14:04:09] [INFO] features.3.0.block.0.1.weight 的平均映射误差: 0.035271
253
+ [2025-04-11 14:04:09] [INFO] features.3.0.block.1.0.weight 的平均映射误差: 0.003948
254
+ [2025-04-11 14:04:09] [INFO] features.3.0.block.1.1.weight 的平均映射误差: 0.035356
255
+ [2025-04-11 14:04:09] [INFO] features.3.1.block.0.0.weight 的平均映射误差: 0.001615
256
+ [2025-04-11 14:04:09] [INFO] features.3.1.block.0.1.weight 的平均映射误差: 0.037526
257
+ [2025-04-11 14:04:09] [INFO] features.3.1.block.1.0.weight 的平均映射误差: 0.001997
258
+ [2025-04-11 14:04:09] [INFO] features.3.1.block.1.1.weight 的平均映射误差: 0.035886
259
+ [2025-04-11 14:04:09] [INFO] features.3.2.block.0.0.weight 的平均映射误差: 0.001612
260
+ [2025-04-11 14:04:09] [INFO] features.3.2.block.0.1.weight 的平均映射误差: 0.046501
261
+ [2025-04-11 14:04:09] [INFO] features.3.2.block.1.0.weight 的平均映射误差: 0.001937
262
+ [2025-04-11 14:04:09] [INFO] features.3.2.block.1.1.weight 的平均映射误差: 0.035704
263
+ [2025-04-11 14:04:09] [INFO] features.3.3.block.0.0.weight 的平均映射误差: 0.001616
264
+ [2025-04-11 14:04:09] [INFO] features.3.3.block.0.1.weight 的平均映射误差: 0.048405
265
+ [2025-04-11 14:04:09] [INFO] features.3.3.block.1.0.weight 的平均映射误差: 0.001908
266
+ [2025-04-11 14:04:09] [INFO] features.3.3.block.1.1.weight 的平均映射误差: 0.037015
267
+ [2025-04-11 14:04:09] [INFO] features.3.4.block.0.0.weight 的平均映射误差: 0.001609
268
+ [2025-04-11 14:04:09] [INFO] features.3.4.block.0.1.weight 的平均映射误差: 0.040396
269
+ [2025-04-11 14:04:09] [INFO] features.3.4.block.1.0.weight 的平均映射误差: 0.001843
270
+ [2025-04-11 14:04:09] [INFO] features.3.4.block.1.1.weight 的平均映射误差: 0.035037
271
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.0.0.weight 的平均映射误差: 0.003849
272
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.0.1.weight 的平均映射误差: 0.041293
273
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.1.0.weight 的平均映射误差: 0.004803
274
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.1.1.weight 的平均映射误差: 0.047467
275
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.2.fc1.weight 的平均映射误差: 0.001522
276
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.2.fc2.weight 的平均映射误差: 0.001604
277
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.3.0.weight 的平均映射误差: 0.002927
278
+ [2025-04-11 14:04:09] [INFO] features.4.0.block.3.1.weight 的平均映射误差: 0.036195
279
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.0.0.weight 的平均映射误差: 0.001678
280
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.0.1.weight 的平均映射误差: 0.035676
281
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.1.0.weight 的平均映射误差: 0.002729
282
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.1.1.weight 的平均映射误差: 0.036697
283
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.2.fc1.weight 的平均映射误差: 0.001411
284
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.2.fc2.weight 的平均映射误差: 0.002050
285
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.3.0.weight 的平均映射误差: 0.001683
286
+ [2025-04-11 14:04:09] [INFO] features.4.1.block.3.1.weight 的平均映射误差: 0.038947
287
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.0.0.weight 的平均映射误差: 0.001681
288
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.0.1.weight 的平均映射误差: 0.037459
289
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.1.0.weight 的平均映射误差: 0.002711
290
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.1.1.weight 的平均映射误差: 0.036801
291
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.2.fc1.weight 的平均映射误差: 0.001278
292
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.2.fc2.weight 的平均映射误差: 0.001888
293
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.3.0.weight 的平均映射误差: 0.001645
294
+ [2025-04-11 14:04:09] [INFO] features.4.2.block.3.1.weight 的平均映射误差: 0.035763
295
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.0.0.weight 的平均映射误差: 0.001662
296
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.0.1.weight 的平均映射误差: 0.036247
297
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.1.0.weight 的平均映射误差: 0.002566
298
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.1.1.weight 的平均映射误差: 0.036660
299
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.2.fc1.weight 的平均映射误差: 0.000994
300
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.2.fc2.weight 的平均映射误差: 0.001397
301
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.3.0.weight 的平均映射误差: 0.001628
302
+ [2025-04-11 14:04:09] [INFO] features.4.3.block.3.1.weight 的平均映射误差: 0.036731
303
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.0.0.weight 的平均映射误差: 0.001657
304
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.0.1.weight 的平均映射误差: 0.037271
305
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.1.0.weight 的平均映射误差: 0.002587
306
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.1.1.weight 的平均映射误差: 0.034789
307
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.2.fc1.weight 的平均映射误差: 0.000931
308
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.2.fc2.weight 的平均映射误差: 0.000870
309
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.3.0.weight 的平均映射误差: 0.001629
310
+ [2025-04-11 14:04:09] [INFO] features.4.4.block.3.1.weight 的平均映射误差: 0.039036
311
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.0.0.weight 的平均映射误差: 0.001656
312
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.0.1.weight 的平均映射误差: 0.038480
313
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.1.0.weight 的平均映射误差: 0.002324
314
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.1.1.weight 的平均映射误差: 0.035013
315
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.2.fc1.weight 的平均映射误差: 0.000697
316
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.2.fc2.weight 的平均映射误差: 0.000833
317
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.3.0.weight 的平均映射误差: 0.001629
318
+ [2025-04-11 14:04:09] [INFO] features.4.5.block.3.1.weight 的平均映射误差: 0.037597
319
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.0.0.weight 的平均映射误差: 0.001672
320
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.0.1.weight 的平均映射误差: 0.037576
321
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.1.0.weight 的平均映射误差: 0.002219
322
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.1.1.weight 的平均映射误差: 0.042608
323
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.2.fc1.weight 的平均映射误差: 0.000758
324
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.2.fc2.weight 的平均映射误差: 0.000936
325
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.3.0.weight 的平均映射误差: 0.001620
326
+ [2025-04-11 14:04:09] [INFO] features.4.6.block.3.1.weight 的平均映射误差: 0.036787
327
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.0.0.weight 的平均映射误差: 0.002142
328
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.0.1.weight 的���均映射误差: 0.036103
329
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.1.0.weight 的平均映射误差: 0.003579
330
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.1.1.weight 的平均映射误差: 0.039109
331
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.2.fc1.weight 的平均映射误差: 0.001812
332
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.2.fc2.weight 的平均映射误差: 0.002000
333
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.3.0.weight 的平均映射误差: 0.001975
334
+ [2025-04-11 14:04:09] [INFO] features.5.0.block.3.1.weight 的平均映射误差: 0.035447
335
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.0.0.weight 的平均映射误差: 0.001631
336
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.0.1.weight 的平均映射误差: 0.039400
337
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.1.0.weight 的平均映射误差: 0.002177
338
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.1.1.weight 的平均映射误差: 0.040357
339
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.2.fc1.weight 的平均映射误差: 0.001004
340
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.2.fc2.weight 的平均映射误差: 0.001839
341
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.3.0.weight 的平均映射误差: 0.001607
342
+ [2025-04-11 14:04:09] [INFO] features.5.1.block.3.1.weight 的平均映射误差: 0.043082
343
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.0.0.weight 的平均映射误差: 0.001615
344
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.0.1.weight 的平均映射误差: 0.037653
345
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.1.0.weight 的平均映射误差: 0.002106
346
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.1.1.weight 的平均映射误差: 0.039797
347
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.2.fc1.weight 的平均映射误差: 0.000975
348
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.2.fc2.weight 的平均映射误差: 0.001606
349
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.3.0.weight 的平均映射误差: 0.001593
350
+ [2025-04-11 14:04:09] [INFO] features.5.2.block.3.1.weight 的平均映射误差: 0.035553
351
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.0.0.weight 的平均映射误差: 0.001601
352
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.0.1.weight 的平均映射误差: 0.038712
353
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.1.0.weight 的平均映射误差: 0.002018
354
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.1.1.weight 的平均映射误差: 0.041242
355
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.2.fc1.weight 的平均映射误差: 0.000880
356
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.2.fc2.weight 的平均映射误差: 0.001263
357
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.3.0.weight 的平均映射误差: 0.001566
358
+ [2025-04-11 14:04:09] [INFO] features.5.3.block.3.1.weight 的平均映射误差: 0.034203
359
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.0.0.weight 的平均映射误差: 0.001612
360
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.0.1.weight 的平均映射误差: 0.039207
361
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.1.0.weight 的平均映射误差: 0.001985
362
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.1.1.weight 的平均映射误差: 0.042814
363
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.2.fc1.weight 的平均映射误差: 0.000883
364
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.2.fc2.weight 的平均映射误差: 0.001161
365
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.3.0.weight 的平均映射误差: 0.001562
366
+ [2025-04-11 14:04:09] [INFO] features.5.4.block.3.1.weight 的平均映射误差: 0.034334
367
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.0.0.weight 的平均映射误差: 0.001609
368
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.0.1.weight 的平均映射误差: 0.038458
369
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.1.0.weight 的平均映射误差: 0.001952
370
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.1.1.weight 的平均映射误差: 0.042985
371
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.2.fc1.weight 的平均映射误差: 0.000744
372
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.2.fc2.weight 的平均映射误差: 0.000928
373
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.3.0.weight 的平均映射误差: 0.001533
374
+ [2025-04-11 14:04:09] [INFO] features.5.5.block.3.1.weight 的平均映射误差: 0.032241
375
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.0.0.weight 的平均映射误差: 0.001601
376
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.0.1.weight 的平均映射误差: 0.037920
377
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.1.0.weight 的平均映射误差: 0.001837
378
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.1.1.weight 的平均映射误差: 0.044167
379
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.2.fc1.weight 的平均映射误差: 0.001000
380
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.2.fc2.weight 的平均映射误差: 0.001211
381
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.3.0.weight 的平均映射误差: 0.001525
382
+ [2025-04-11 14:04:09] [INFO] features.5.6.block.3.1.weight 的平均映射误差: 0.033185
383
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.0.0.weight 的平均映射误差: 0.001583
384
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.0.1.weight 的平均映射误差: 0.037759
385
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.1.0.weight 的平均映射误差: 0.001893
386
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.1.1.weight 的平均映射误差: 0.044461
387
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.2.fc1.weight 的平均映射误差: 0.000729
388
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.2.fc2.weight 的平均映射误差: 0.000682
389
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.3.0.weight 的平均映射误差: 0.001500
390
+ [2025-04-11 14:04:09] [INFO] features.5.7.block.3.1.weight 的平均映射误差: 0.031199
391
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.0.0.weight 的平均映射误差: 0.001576
392
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.0.1.weight 的平均映射误差: 0.037488
393
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.1.0.weight 的平均映射误差: 0.001827
394
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.1.1.weight 的平均映射误差: 0.044563
395
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.2.fc1.weight 的平均映射误差: 0.000721
396
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.2.fc2.weight 的平均映射误差: 0.000761
397
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.3.0.weight 的平均映射误差: 0.001521
398
+ [2025-04-11 14:04:09] [INFO] features.5.8.block.3.1.weight 的平均映射误差: 0.032891
399
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.0.0.weight 的平均映射误差: 0.001583
400
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.0.1.weight 的平均映射误差: 0.037175
401
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.1.0.weight 的平均映射误差: 0.001795
402
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.1.1.weight 的平均映射误差: 0.045632
403
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.2.fc1.weight 的平均映射误差: 0.000896
404
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.2.fc2.weight 的平均映射误差: 0.000932
405
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.3.0.weight 的平均映射误差: 0.001511
406
+ [2025-04-11 14:04:09] [INFO] features.5.9.block.3.1.weight 的平均映射误差: 0.031393
407
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.0.0.weight 的平均映射误差: 0.001595
408
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.0.1.weight 的平均映射误差: 0.036152
409
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.1.0.weight 的平均映射误差: 0.001876
410
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.1.1.weight 的平均映射误差: 0.044134
411
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.2.fc1.weight 的平均映射误差: 0.000720
412
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.2.fc2.weight 的平均映射误差: 0.000697
413
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.3.0.weight 的平均映射误差: 0.001556
414
+ [2025-04-11 14:04:09] [INFO] features.5.10.block.3.1.weight 的平均映射误差: 0.040700
415
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.0.0.weight 的平均映射误差: 0.001601
416
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.0.1.weight 的平均映射误差: 0.036955
417
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.1.0.weight 的平均映射误差: 0.001872
418
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.1.1.weight 的平均映射误差: 0.046341
419
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.2.fc1.weight 的平均映射误差: 0.000779
420
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.2.fc2.weight 的平均映射误差: 0.000740
421
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.3.0.weight 的平均映射误差: 0.001563
422
+ [2025-04-11 14:04:09] [INFO] features.5.11.block.3.1.weight 的平均映射误差: 0.039411
423
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.0.0.weight 的平均映射误差: 0.001593
424
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.0.1.weight 的平均映射误差: 0.036985
425
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.1.0.weight 的平均映射误差: 0.001790
426
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.1.1.weight 的平均映射误差: 0.047870
427
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.2.fc1.weight 的平均映射误差: 0.000735
428
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.2.fc2.weight 的平均映射误差: 0.000687
429
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.3.0.weight 的平均映射误差: 0.001555
430
+ [2025-04-11 14:04:09] [INFO] features.5.12.block.3.1.weight 的平均映射误差: 0.040334
431
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.0.0.weight 的平均映射误差: 0.001592
432
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.0.1.weight 的平均映射误差: 0.036830
433
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.1.0.weight 的平均映射误差: 0.001799
434
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.1.1.weight 的平均映射误差: 0.046893
435
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.2.fc1.weight 的平均映射误差: 0.000533
436
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.2.fc2.weight 的平均映射误差: 0.000704
437
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.3.0.weight 的平均映射误差: 0.001563
438
+ [2025-04-11 14:04:09] [INFO] features.5.13.block.3.1.weight 的平均映射误差: 0.043067
439
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.0.0.weight 的平均映射误差: 0.002128
440
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.0.1.weight 的平均映射误差: 0.039298
441
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.1.0.weight 的平均映射误差: 0.003184
442
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.1.1.weight 的平均映射误差: 0.039935
443
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.2.fc1.weight 的平均映射误差: 0.000593
444
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.2.fc2.weight 的平均映射误差: 0.000691
445
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.3.0.weight 的平均映射误差: 0.001895
446
+ [2025-04-11 14:04:09] [INFO] features.6.0.block.3.1.weight 的平均映射误差: 0.035337
447
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.0.0.weight 的平均映射误差: 0.001571
448
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.0.1.weight 的平均映射误差: 0.039118
449
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.1.0.weight 的平均映射误差: 0.001993
450
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.1.1.weight 的平均映射误差: 0.039696
451
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.2.fc1.weight 的平均映射误差: 0.000750
452
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.2.fc2.weight 的平均映射误差: 0.001492
453
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.3.0.weight 的平均映射误差: 0.001569
454
+ [2025-04-11 14:04:09] [INFO] features.6.1.block.3.1.weight 的平均映射误差: 0.047702
455
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.0.0.weight 的平均映射误差: 0.001567
456
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.0.1.weight 的平均映射误差: 0.039382
457
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.1.0.weight 的平均映射误差: 0.001989
458
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.1.1.weight 的平均映射误差: 0.040895
459
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.2.fc1.weight 的平均映射误差: 0.000717
460
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.2.fc2.weight 的平均映射误差: 0.001370
461
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.3.0.weight 的平均映射误差: 0.001558
462
+ [2025-04-11 14:04:09] [INFO] features.6.2.block.3.1.weight 的平均映射误差: 0.045440
463
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.0.0.weight 的平均映射误差: 0.001554
464
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.0.1.weight 的平均映射误差: 0.039295
465
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.1.0.weight 的平均映射误差: 0.001931
466
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.1.1.weight 的平均映射误差: 0.044459
467
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.2.fc1.weight 的平均映射误差: 0.000728
468
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.2.fc2.weight 的平均映射误差: 0.001147
469
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.3.0.weight 的平均映射误差: 0.001530
470
+ [2025-04-11 14:04:09] [INFO] features.6.3.block.3.1.weight 的平均映射误差: 0.045182
471
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.0.0.weight 的平均映射误差: 0.001557
472
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.0.1.weight 的平均映射误差: 0.038390
473
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.1.0.weight 的平均映射误差: 0.001897
474
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.1.1.weight 的平均映射误差: 0.045389
475
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.2.fc1.weight 的平均映射误差: 0.000786
476
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.2.fc2.weight 的平均映射误差: 0.001205
477
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.3.0.weight 的平均映射误差: 0.001528
478
+ [2025-04-11 14:04:09] [INFO] features.6.4.block.3.1.weight 的平均映射误差: 0.045710
479
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.0.0.weight 的平均映射误差: 0.001558
480
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.0.1.weight 的平均映射误差: 0.039380
481
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.1.0.weight 的平均映射误差: 0.001850
482
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.1.1.weight 的平均映射误差: 0.045480
483
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.2.fc1.weight 的平均映射误差: 0.000766
484
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.2.fc2.weight 的平均映射误差: 0.001158
485
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.3.0.weight 的平均映射误差: 0.001533
486
+ [2025-04-11 14:04:09] [INFO] features.6.5.block.3.1.weight 的平均映射误差: 0.043887
487
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.0.0.weight 的平均映射误差: 0.001550
488
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.0.1.weight 的平均映射误差: 0.037942
489
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.1.0.weight 的平均映射误差: 0.001844
490
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.1.1.weight 的平均映射误差: 0.047148
491
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.2.fc1.weight 的平均映射误差: 0.000710
492
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.2.fc2.weight 的平均映射误差: 0.000932
493
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.3.0.weight 的平均映射误差: 0.001515
494
+ [2025-04-11 14:04:09] [INFO] features.6.6.block.3.1.weight 的平均映射误差: 0.041574
495
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.0.0.weight 的平均映射误差: 0.001556
496
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.0.1.weight 的平均映射误差: 0.039379
497
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.1.0.weight 的平均映射误差: 0.001848
498
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.1.1.weight 的平均映射误差: 0.048073
499
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.2.fc1.weight 的平均映射误差: 0.000833
500
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.2.fc2.weight 的平均映射误差: 0.001102
501
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.3.0.weight 的平均映射误差: 0.001527
502
+ [2025-04-11 14:04:09] [INFO] features.6.7.block.3.1.weight 的平均映射误差: 0.045150
503
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.0.0.weight 的平均映射误差: 0.001560
504
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.0.1.weight 的平均映射误差: 0.039538
505
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.1.0.weight 的平均映射误差: 0.001824
506
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.1.1.weight 的平均映射误差: 0.047611
507
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.2.fc1.weight 的平均映射误差: 0.000725
508
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.2.fc2.weight 的平均映射误差: 0.000960
509
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.3.0.weight 的平均映射误差: 0.001530
510
+ [2025-04-11 14:04:09] [INFO] features.6.8.block.3.1.weight 的平均映射误差: 0.044146
511
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.0.0.weight 的平均映射误差: 0.001563
512
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.0.1.weight 的平均映射误差: 0.038693
513
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.1.0.weight 的平均映射误差: 0.001812
514
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.1.1.weight 的平均映射误差: 0.043046
515
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.2.fc1.weight 的平均映射误差: 0.000733
516
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.2.fc2.weight 的平均映射误差: 0.000939
517
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.3.0.weight 的平均映射误差: 0.001535
518
+ [2025-04-11 14:04:09] [INFO] features.6.9.block.3.1.weight 的平均映射误差: 0.043821
519
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.0.0.weight 的平均映射误差: 0.001568
520
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.0.1.weight 的平均映射误差: 0.039959
521
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.1.0.weight 的平均映射误差: 0.001812
522
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.1.1.weight 的平均映射误差: 0.042267
523
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.2.fc1.weight 的平均映射误差: 0.000677
524
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.2.fc2.weight 的平均映射误差: 0.000963
525
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.3.0.weight 的平均映射误差: 0.001543
526
+ [2025-04-11 14:04:09] [INFO] features.6.10.block.3.1.weight 的平均映射误差: 0.043737
527
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.0.0.weight 的平均映射误差: 0.001571
528
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.0.1.weight 的平均映射误差: 0.040201
529
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.1.0.weight 的平均映射误差: 0.001807
530
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.1.1.weight 的平均映射误差: 0.042812
531
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.2.fc1.weight 的平均映射误差: 0.000684
532
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.2.fc2.weight 的平均映射误差: 0.000984
533
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.3.0.weight 的平均映射误差: 0.001556
534
+ [2025-04-11 14:04:09] [INFO] features.6.11.block.3.1.weight 的平均映射误差: 0.042713
535
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.0.0.weight 的平均映射误差: 0.001578
536
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.0.1.weight 的平均映射误差: 0.040653
537
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.1.0.weight 的平均映射误差: 0.001772
538
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.1.1.weight 的平均映射误差: 0.036763
539
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.2.fc1.weight 的平均映射误差: 0.000536
540
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.2.fc2.weight 的平均映射误差: 0.000682
541
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.3.0.weight 的平均映射误差: 0.001564
542
+ [2025-04-11 14:04:09] [INFO] features.6.12.block.3.1.weight 的平均映射误差: 0.039051
543
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.0.0.weight 的平均映射误差: 0.001572
544
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.0.1.weight 的平均映射误差: 0.040055
545
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.1.0.weight 的平均映射误差: 0.001756
546
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.1.1.weight 的平均映射误差: 0.042804
547
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.2.fc1.weight 的平均映射误差: 0.000655
548
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.2.fc2.weight 的平均映射误差: 0.001058
549
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.3.0.weight 的平均映射误差: 0.001555
550
+ [2025-04-11 14:04:09] [INFO] features.6.13.block.3.1.weight 的平均映射误差: 0.041251
551
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.0.0.weight 的平均映射误差: 0.001583
552
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.0.1.weight 的平均映射误差: 0.042735
553
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.1.0.weight 的平均映射误差: 0.001723
554
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.1.1.weight 的平均映射误差: 0.040822
555
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.2.fc1.weight 的平均映射误差: 0.000731
556
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.2.fc2.weight 的平均映射误差: 0.000846
557
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.3.0.weight 的平均映射误差: 0.001557
558
+ [2025-04-11 14:04:09] [INFO] features.6.14.block.3.1.weight 的平均映射误差: 0.037729
559
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.0.0.weight 的平均映射误差: 0.001579
560
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.0.1.weight 的平均映射误差: 0.044750
561
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.1.0.weight 的平均映射误差: 0.001703
562
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.1.1.weight 的平均映射误差: 0.040765
563
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.2.fc1.weight 的平均映射误差: 0.000709
564
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.2.fc2.weight 的平均映射误差: 0.000733
565
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.3.0.weight 的平均映射误差: 0.001568
566
+ [2025-04-11 14:04:09] [INFO] features.6.15.block.3.1.weight 的平均映射误差: 0.037312
567
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.0.0.weight 的平均映射误差: 0.001562
568
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.0.1.weight 的平均映射误差: 0.041841
569
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.1.0.weight 的平均映射误差: 0.001684
570
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.1.1.weight 的平均映射误差: 0.048556
571
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.2.fc1.weight 的平均映射误差: 0.000644
572
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.2.fc2.weight 的平均映射误差: 0.000983
573
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.3.0.weight 的平均映射误差: 0.001531
574
+ [2025-04-11 14:04:09] [INFO] features.6.16.block.3.1.weight 的平均映射误差: 0.039829
575
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.0.0.weight 的平均映射误差: 0.001553
576
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.0.1.weight 的平均映射误差: 0.043430
577
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.1.0.weight 的平均映射误差: 0.001669
578
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.1.1.weight 的平均映射误差: 0.049339
579
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.2.fc1.weight 的平均映射误差: 0.000773
580
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.2.fc2.weight 的平均映射误差: 0.001068
581
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.3.0.weight 的平均映射误差: 0.001521
582
+ [2025-04-11 14:04:09] [INFO] features.6.17.block.3.1.weight 的平均映射误差: 0.040126
583
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.0.0.weight 的平均映射误差: 0.001854
584
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.0.1.weight 的平均映射误差: 0.032257
585
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.1.0.weight 的平均映射误差: 0.002046
586
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.1.1.weight 的平均映射误差: 0.033026
587
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.2.fc1.weight 的平均映射误差: 0.001505
588
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.2.fc2.weight 的平均映射误差: 0.001698
589
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.3.0.weight 的平均映射误差: 0.001625
590
+ [2025-04-11 14:04:09] [INFO] features.7.0.block.3.1.weight 的平均映射误差: 0.034802
591
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.0.0.weight 的平均映射误差: 0.001533
592
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.0.1.weight 的平均映射误差: 0.041285
593
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.1.0.weight 的平均映射误差: 0.001758
594
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.1.1.weight 的平均映射误差: 0.039994
595
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.2.fc1.weight 的平均映射误差: 0.001125
596
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.2.fc2.weight 的平均映射误差: 0.001551
597
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.3.0.weight 的平均映射误差: 0.001506
598
+ [2025-04-11 14:04:09] [INFO] features.7.1.block.3.1.weight 的平均映射误差: 0.048225
599
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.0.0.weight 的平均映射误差: 0.001510
600
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.0.1.weight 的平均映射误差: 0.047484
601
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.1.0.weight 的平均映射误差: 0.002199
602
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.1.1.weight 的平均映射误差: 0.044308
603
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.2.fc1.weight 的平均映射误差: 0.000882
604
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.2.fc2.weight 的平均映射误差: 0.001283
605
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.3.0.weight 的平均映射误差: 0.001465
606
+ [2025-04-11 14:04:09] [INFO] features.7.2.block.3.1.weight 的平均映射误差: 0.037651
607
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.0.0.weight 的平均映射误差: 0.001421
608
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.0.1.weight 的平均映射误差: 0.045544
609
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.1.0.weight 的平均映射误差: 0.002499
610
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.1.1.weight 的平均映射误差: 0.039849
611
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.2.fc1.weight 的平均映射误差: 0.000900
612
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.2.fc2.weight 的平均映射误差: 0.001267
613
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.3.0.weight 的平均映射误差: 0.001375
614
+ [2025-04-11 14:04:09] [INFO] features.7.3.block.3.1.weight 的平均映射误差: 0.037865
615
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.0.0.weight 的平均映射误差: 0.001373
616
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.0.1.weight 的平均映射误差: 0.035588
617
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.1.0.weight 的平均映射误差: 0.002182
618
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.1.1.weight 的平均映射误差: 0.034097
619
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.2.fc1.weight 的平均映射误差: 0.001164
620
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.2.fc2.weight 的平均映射误差: 0.001211
621
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.3.0.weight 的平均映射误差: 0.001331
622
+ [2025-04-11 14:04:09] [INFO] features.7.4.block.3.1.weight 的平均映射误差: 0.039661
623
+ [2025-04-11 14:04:09] [INFO] features.8.0.weight 的平均映射误差: 0.001614
624
+ [2025-04-11 14:04:09] [INFO] features.8.1.weight 的平均映射误差: 0.035294
625
+ [2025-04-11 14:04:09] [INFO] classifier.1.weight 的平均映射误差: 0.001793
626
+ [2025-04-11 14:04:10] [INFO] 评估结果 - Loss: 1.1450, Accuracy: 0.4483
627
+ [2025-04-11 14:04:10] [INFO] RRAM映射模型已保存到 checkpoints/rram_mapped_model.pth
628
+ [2025-04-11 14:04:10] [INFO] RRAM映射前后精度对比: 原始 0.6552 vs RRAM映射后 0.4483, 变化: -0.2069
629
+ [2025-04-11 14:04:10] [INFO] 开始微调全连接层 (epochs=50, lr=5e-05)...
630
+ [2025-04-11 14:04:10] [INFO] 微调过程中的模型将保存到: checkpoints/fine_tune_checkpoints
631
+ [2025-04-11 14:04:11] [INFO] Fine-tuning Epoch 1/50 - Train Acc: 0.9859, Val Acc: 0.6207
632
+ [2025-04-11 14:04:12] [INFO] 已保存第 1 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_1.pth
633
+ [2025-04-11 14:04:13] [INFO] Fine-tuning Epoch 2/50 - Train Acc: 0.9437, Val Acc: 0.7586
634
+ [2025-04-11 14:04:14] [INFO] 已保存第 2 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_2.pth
635
+ [2025-04-11 14:04:15] [INFO] Fine-tuning Epoch 3/50 - Train Acc: 0.9648, Val Acc: 0.7586
636
+ [2025-04-11 14:04:16] [INFO] 已保存第 3 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_3.pth
637
+ [2025-04-11 14:04:18] [INFO] Fine-tuning Epoch 4/50 - Train Acc: 0.9859, Val Acc: 0.7241
638
+ [2025-04-11 14:04:18] [INFO] 已保存第 4 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_4.pth
639
+ [2025-04-11 14:04:20] [INFO] Fine-tuning Epoch 5/50 - Train Acc: 0.9507, Val Acc: 0.7241
640
+ [2025-04-11 14:04:21] [INFO] 已保存第 5 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_5.pth
641
+ [2025-04-11 14:04:22] [INFO] Fine-tuning Epoch 6/50 - Train Acc: 0.9789, Val Acc: 0.6552
642
+ [2025-04-11 14:04:23] [INFO] 已保存第 6 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_6.pth
643
+ [2025-04-11 14:04:24] [INFO] Fine-tuning Epoch 7/50 - Train Acc: 0.9859, Val Acc: 0.6552
644
+ [2025-04-11 14:04:25] [INFO] 已保存第 7 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_7.pth
645
+ [2025-04-11 14:04:26] [INFO] Fine-tuning Epoch 8/50 - Train Acc: 0.9648, Val Acc: 0.6552
646
+ [2025-04-11 14:04:27] [INFO] 已保存第 8 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_8.pth
647
+ [2025-04-11 14:04:28] [INFO] Fine-tuning Epoch 9/50 - Train Acc: 0.9577, Val Acc: 0.6897
648
+ [2025-04-11 14:04:29] [INFO] 已保存第 9 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_9.pth
649
+ [2025-04-11 14:04:30] [INFO] Fine-tuning Epoch 10/50 - Train Acc: 1.0000, Val Acc: 0.6897
650
+ [2025-04-11 14:04:31] [INFO] 已保存第 10 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_10.pth
651
+ [2025-04-11 14:04:32] [INFO] Fine-tuning Epoch 11/50 - Train Acc: 0.9789, Val Acc: 0.7586
652
+ [2025-04-11 14:04:33] [INFO] 已保存第 11 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_11.pth
653
+ [2025-04-11 14:04:34] [INFO] Fine-tuning Epoch 12/50 - Train Acc: 0.9859, Val Acc: 0.8276
654
+ [2025-04-11 14:04:35] [INFO] 已保存第 12 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_12.pth
655
+ [2025-04-11 14:04:37] [INFO] Fine-tuning Epoch 13/50 - Train Acc: 0.9930, Val Acc: 0.7241
656
+ [2025-04-11 14:04:37] [INFO] 已保存第 13 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_13.pth
657
+ [2025-04-11 14:04:39] [INFO] Fine-tuning Epoch 14/50 - Train Acc: 0.9930, Val Acc: 0.7586
658
+ [2025-04-11 14:04:39] [INFO] 已保存第 14 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_14.pth
659
+ [2025-04-11 14:04:41] [INFO] Fine-tuning Epoch 15/50 - Train Acc: 1.0000, Val Acc: 0.7586
660
+ [2025-04-11 14:04:41] [INFO] 已保存第 15 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_15.pth
661
+ [2025-04-11 14:04:43] [INFO] Fine-tuning Epoch 16/50 - Train Acc: 0.9789, Val Acc: 0.7586
662
+ [2025-04-11 14:04:43] [INFO] 已保存第 16 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_16.pth
663
+ [2025-04-11 14:04:45] [INFO] Fine-tuning Epoch 17/50 - Train Acc: 0.9789, Val Acc: 0.7586
664
+ [2025-04-11 14:04:46] [INFO] 已保存第 17 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_17.pth
665
+ [2025-04-11 14:04:47] [INFO] Fine-tuning Epoch 18/50 - Train Acc: 0.9718, Val Acc: 0.7586
666
+ [2025-04-11 14:04:48] [INFO] 已保存第 18 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_18.pth
667
+ [2025-04-11 14:04:49] [INFO] Fine-tuning Epoch 19/50 - Train Acc: 0.9718, Val Acc: 0.7241
668
+ [2025-04-11 14:04:50] [INFO] 已保存第 19 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_19.pth
669
+ [2025-04-11 14:04:51] [INFO] Fine-tuning Epoch 20/50 - Train Acc: 1.0000, Val Acc: 0.6552
670
+ [2025-04-11 14:04:52] [INFO] 已保存第 20 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_20.pth
671
+ [2025-04-11 14:04:53] [INFO] Fine-tuning Epoch 21/50 - Train Acc: 0.9789, Val Acc: 0.6897
672
+ [2025-04-11 14:04:54] [INFO] 已保存第 21 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_21.pth
673
+ [2025-04-11 14:04:55] [INFO] Fine-tuning Epoch 22/50 - Train Acc: 1.0000, Val Acc: 0.8276
674
+ [2025-04-11 14:04:56] [INFO] 已保存第 22 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_22.pth
675
+ [2025-04-11 14:04:57] [INFO] Fine-tuning Epoch 23/50 - Train Acc: 0.9859, Val Acc: 0.8621
676
+ [2025-04-11 14:04:58] [INFO] 已保存第 23 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_23.pth
677
+ [2025-04-11 14:04:59] [INFO] Fine-tuning Epoch 24/50 - Train Acc: 0.9859, Val Acc: 0.8621
678
+ [2025-04-11 14:05:00] [INFO] 已保存第 24 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_24.pth
679
+ [2025-04-11 14:05:01] [INFO] Fine-tuning Epoch 25/50 - Train Acc: 0.9718, Val Acc: 0.8276
680
+ [2025-04-11 14:05:02] [INFO] 已保存第 25 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_25.pth
681
+ [2025-04-11 14:05:03] [INFO] Fine-tuning Epoch 26/50 - Train Acc: 0.9859, Val Acc: 0.7241
682
+ [2025-04-11 14:05:04] [INFO] 已保存第 26 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_26.pth
683
+ [2025-04-11 14:05:05] [INFO] Fine-tuning Epoch 27/50 - Train Acc: 1.0000, Val Acc: 0.7241
684
+ [2025-04-11 14:05:06] [INFO] 已保存第 27 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_27.pth
685
+ [2025-04-11 14:05:08] [INFO] Fine-tuning Epoch 28/50 - Train Acc: 0.9930, Val Acc: 0.7241
686
+ [2025-04-11 14:05:08] [INFO] 已保存第 28 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_28.pth
687
+ [2025-04-11 14:05:10] [INFO] Fine-tuning Epoch 29/50 - Train Acc: 1.0000, Val Acc: 0.7241
688
+ [2025-04-11 14:05:11] [INFO] 已保存第 29 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_29.pth
689
+ [2025-04-11 14:05:12] [INFO] Fine-tuning Epoch 30/50 - Train Acc: 0.9930, Val Acc: 0.7241
690
+ [2025-04-11 14:05:13] [INFO] 已保存第 30 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_30.pth
691
+ [2025-04-11 14:05:14] [INFO] Fine-tuning Epoch 31/50 - Train Acc: 0.9930, Val Acc: 0.7586
692
+ [2025-04-11 14:05:15] [INFO] 已保存第 31 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_31.pth
693
+ [2025-04-11 14:05:16] [INFO] Fine-tuning Epoch 32/50 - Train Acc: 0.9930, Val Acc: 0.6897
694
+ [2025-04-11 14:05:17] [INFO] 已保存第 32 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_32.pth
695
+ [2025-04-11 14:05:18] [INFO] Fine-tuning Epoch 33/50 - Train Acc: 1.0000, Val Acc: 0.6897
696
+ [2025-04-11 14:05:19] [INFO] 已保存第 33 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_33.pth
697
+ [2025-04-11 14:05:20] [INFO] Fine-tuning Epoch 34/50 - Train Acc: 1.0000, Val Acc: 0.6897
698
+ [2025-04-11 14:05:21] [INFO] 已保存第 34 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_34.pth
699
+ [2025-04-11 14:05:22] [INFO] Fine-tuning Epoch 35/50 - Train Acc: 1.0000, Val Acc: 0.8621
700
+ [2025-04-11 14:05:23] [INFO] 已保存第 35 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_35.pth
701
+ [2025-04-11 14:05:24] [INFO] Fine-tuning Epoch 36/50 - Train Acc: 0.9930, Val Acc: 0.8276
702
+ [2025-04-11 14:05:25] [INFO] 已保存第 36 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_36.pth
703
+ [2025-04-11 14:05:26] [INFO] Fine-tuning Epoch 37/50 - Train Acc: 1.0000, Val Acc: 0.7931
704
+ [2025-04-11 14:05:27] [INFO] 已保存第 37 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_37.pth
705
+ [2025-04-11 14:05:28] [INFO] Fine-tuning Epoch 38/50 - Train Acc: 1.0000, Val Acc: 0.7931
706
+ [2025-04-11 14:05:29] [INFO] 已保存第 38 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_38.pth
707
+ [2025-04-11 14:05:30] [INFO] Fine-tuning Epoch 39/50 - Train Acc: 0.9930, Val Acc: 0.7931
708
+ [2025-04-11 14:05:31] [INFO] 已保存第 39 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_39.pth
709
+ [2025-04-11 14:05:33] [INFO] Fine-tuning Epoch 40/50 - Train Acc: 0.9859, Val Acc: 0.8966
710
+ [2025-04-11 14:05:33] [INFO] 已保存第 40 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_40.pth
711
+ [2025-04-11 14:05:35] [INFO] Fine-tuning Epoch 41/50 - Train Acc: 0.9930, Val Acc: 0.8966
712
+ [2025-04-11 14:05:35] [INFO] 已保存第 41 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_41.pth
713
+ [2025-04-11 14:05:37] [INFO] Fine-tuning Epoch 42/50 - Train Acc: 1.0000, Val Acc: 0.8621
714
+ [2025-04-11 14:05:37] [INFO] 已保存第 42 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_42.pth
715
+ [2025-04-11 14:05:39] [INFO] Fine-tuning Epoch 43/50 - Train Acc: 1.0000, Val Acc: 0.8966
716
+ [2025-04-11 14:05:39] [INFO] 已保存第 43 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_43.pth
717
+ [2025-04-11 14:05:41] [INFO] Fine-tuning Epoch 44/50 - Train Acc: 0.9859, Val Acc: 0.8966
718
+ [2025-04-11 14:05:42] [INFO] 已保存第 44 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_44.pth
719
+ [2025-04-11 14:05:43] [INFO] Fine-tuning Epoch 45/50 - Train Acc: 1.0000, Val Acc: 0.8966
720
+ [2025-04-11 14:05:44] [INFO] 已保存第 45 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_45.pth
721
+ [2025-04-11 14:05:45] [INFO] Fine-tuning Epoch 46/50 - Train Acc: 1.0000, Val Acc: 0.8966
722
+ [2025-04-11 14:05:46] [INFO] 已保存第 46 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_46.pth
723
+ [2025-04-11 14:05:47] [INFO] Fine-tuning Epoch 47/50 - Train Acc: 0.9930, Val Acc: 0.8966
724
+ [2025-04-11 14:05:48] [INFO] 已保存第 47 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_47.pth
725
+ [2025-04-11 14:05:49] [INFO] Fine-tuning Epoch 48/50 - Train Acc: 1.0000, Val Acc: 0.9310
726
+ [2025-04-11 14:05:50] [INFO] 已保存第 48 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_48.pth
727
+ [2025-04-11 14:05:51] [INFO] Fine-tuning Epoch 49/50 - Train Acc: 0.9718, Val Acc: 0.8966
728
+ [2025-04-11 14:05:52] [INFO] 已保存第 49 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_49.pth
729
+ [2025-04-11 14:05:53] [INFO] Fine-tuning Epoch 50/50 - Train Acc: 1.0000, Val Acc: 0.8621
730
+ [2025-04-11 14:05:54] [INFO] 已保存第 50 轮微调模型到: checkpoints/fine_tune_checkpoints/fine_tuned_model_epoch_50.pth
731
+ [2025-04-11 14:05:54] [INFO] 评估结果 - Loss: 0.4411, Accuracy: 0.8621
732
+ [2025-04-11 14:05:55] [INFO] 微调模型已保存到 checkpoints/fine_tuned_model.pth
733
+ [2025-04-11 14:05:55] [INFO] 微调前后精度对比: RRAM映射 0.4483 vs 微调后 0.8621, 变化: 0.4138
734
+ [2025-04-11 14:05:55] [INFO] 所有处理完成!
checkpoints_v2m_part2/base_training_metrics.csv ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ epoch,train_loss,train_acc,val_loss,val_acc
2
+ 1,1.3665,0.3803,1.3432,0.4483
3
+ 2,1.2475,0.669,1.3103,0.4483
4
+ 3,1.1308,0.7042,1.2931,0.4483
5
+ 4,1.0179,0.7042,1.2997,0.4483
6
+ 5,0.9342,0.7042,1.325,0.4483
7
+ 6,0.8822,0.7042,1.3368,0.4483
8
+ 7,0.8348,0.7042,1.3392,0.4483
9
+ 8,0.7929,0.7042,1.3446,0.4483
10
+ 9,0.7403,0.7042,1.4892,0.4483
11
+ 10,0.6734,0.7113,1.7414,0.4483
12
+ 11,0.6278,0.7183,1.2002,0.4828
13
+ 12,0.5761,0.7324,1.1805,0.4483
14
+ 13,0.5605,0.7606,1.2032,0.5862
15
+ 14,0.5306,0.7887,1.5191,0.4483
16
+ 15,0.4984,0.8028,1.1684,0.6207
17
+ 16,0.4516,0.838,1.1488,0.3793
18
+ 17,0.4322,0.831,1.1146,0.4138
19
+ 18,0.3943,0.8662,1.1947,0.4483
20
+ 19,0.3648,0.8873,1.1186,0.4138
21
+ 20,0.3696,0.9085,1.072,0.6552
22
+ 21,0.3843,0.8521,1.0828,0.5862
23
+ 22,0.344,0.8944,1.0306,0.4483
24
+ 23,0.3167,0.9085,1.0836,0.5172
25
+ 24,0.3757,0.8662,1.0405,0.5172
26
+ 25,0.2902,0.8803,1.0648,0.4828
27
+ 26,0.3106,0.8873,1.0448,0.5862
28
+ 27,0.3001,0.9014,1.0292,0.5172
29
+ 28,0.3327,0.8803,0.9527,0.5517
30
+ 29,0.3186,0.9225,1.0192,0.5517
31
+ 30,0.2801,0.9366,0.9282,0.5172
32
+ 31,0.2773,0.9155,0.9423,0.5862
33
+ 32,0.292,0.9014,1.0064,0.5862
34
+ 33,0.2702,0.9296,0.9927,0.6552
35
+ 34,0.2532,0.9225,1.0585,0.6552
36
+ 35,0.2324,0.9577,1.0661,0.6207
37
+ 36,0.2451,0.9085,0.9552,0.6552
38
+ 37,0.2731,0.9437,0.9102,0.6897
39
+ 38,0.2497,0.9437,0.8856,0.6552
40
+ 39,0.2333,0.9648,0.8399,0.6897
41
+ 40,0.2431,0.9507,0.8593,0.6552
42
+ 41,0.2487,0.9507,0.8634,0.6207
43
+ 42,0.2074,0.9718,0.8873,0.6897
44
+ 43,0.2321,0.9437,0.8828,0.6207
45
+ 44,0.2352,0.9648,0.8669,0.6552
46
+ 45,0.2563,0.9648,0.8832,0.6897
47
+ 46,0.2093,0.9859,0.8795,0.6552
48
+ 47,0.2569,0.9437,0.9025,0.7241
49
+ 48,0.2156,0.9648,0.9006,0.7241
50
+ 49,0.2208,0.9648,0.9028,0.6552
51
+ 50,0.2216,0.9577,0.8998,0.6552
checkpoints_v2m_part2/fine_tuning_metrics.csv ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ epoch,train_acc,val_acc
2
+ 1,0.9859,0.6207
3
+ 2,0.9437,0.7586
4
+ 3,0.9648,0.7586
5
+ 4,0.9859,0.7241
6
+ 5,0.9507,0.7241
7
+ 6,0.9789,0.6552
8
+ 7,0.9859,0.6552
9
+ 8,0.9648,0.6552
10
+ 9,0.9577,0.6897
11
+ 10,1.0,0.6897
12
+ 11,0.9789,0.7586
13
+ 12,0.9859,0.8276
14
+ 13,0.993,0.7241
15
+ 14,0.993,0.7586
16
+ 15,1.0,0.7586
17
+ 16,0.9789,0.7586
18
+ 17,0.9789,0.7586
19
+ 18,0.9718,0.7586
20
+ 19,0.9718,0.7241
21
+ 20,1.0,0.6552
22
+ 21,0.9789,0.6897
23
+ 22,1.0,0.8276
24
+ 23,0.9859,0.8621
25
+ 24,0.9859,0.8621
26
+ 25,0.9718,0.8276
27
+ 26,0.9859,0.7241
28
+ 27,1.0,0.7241
29
+ 28,0.993,0.7241
30
+ 29,1.0,0.7241
31
+ 30,0.993,0.7241
32
+ 31,0.993,0.7586
33
+ 32,0.993,0.6897
34
+ 33,1.0,0.6897
35
+ 34,1.0,0.6897
36
+ 35,1.0,0.8621
37
+ 36,0.993,0.8276
38
+ 37,1.0,0.7931
39
+ 38,1.0,0.7931
40
+ 39,0.993,0.7931
41
+ 40,0.9859,0.8966
42
+ 41,0.993,0.8966
43
+ 42,1.0,0.8621
44
+ 43,1.0,0.8966
45
+ 44,0.9859,0.8966
46
+ 45,1.0,0.8966
47
+ 46,1.0,0.8966
48
+ 47,0.993,0.8966
49
+ 48,1.0,0.931
50
+ 49,0.9718,0.8966
51
+ 50,1.0,0.8621