nielsr HF staff commited on
Commit
1c8ca65
1 Parent(s): 273ff55

End of training

Browse files
all_results.json CHANGED
@@ -1,17 +1,17 @@
1
  {
2
  "epoch": 20.0,
3
- "eval_accuracy": 0.9384550084889643,
4
- "eval_f1": 0.9269016697588126,
5
- "eval_loss": 0.30563604831695557,
6
- "eval_precision": 0.9190581309786607,
7
- "eval_recall": 0.9348802395209581,
8
- "eval_runtime": 8.103,
9
  "eval_samples": 100,
10
- "eval_samples_per_second": 12.341,
11
- "eval_steps_per_second": 0.864,
12
- "train_loss": 0.7290411071777344,
13
- "train_runtime": 1777.9741,
14
  "train_samples": 800,
15
- "train_samples_per_second": 8.999,
16
- "train_steps_per_second": 0.562
17
  }
 
1
  {
2
  "epoch": 20.0,
3
+ "eval_accuracy": 0.9681663837011885,
4
+ "eval_f1": 0.9637654090399701,
5
+ "eval_loss": 0.18448814749717712,
6
+ "eval_precision": 0.9619686800894854,
7
+ "eval_recall": 0.9655688622754491,
8
+ "eval_runtime": 7.9644,
9
  "eval_samples": 100,
10
+ "eval_samples_per_second": 12.556,
11
+ "eval_steps_per_second": 0.879,
12
+ "train_loss": 0.2323043212890625,
13
+ "train_runtime": 1807.2991,
14
  "train_samples": 800,
15
+ "train_samples_per_second": 8.853,
16
+ "train_steps_per_second": 0.553
17
  }
eval_results.json CHANGED
@@ -1,12 +1,12 @@
1
  {
2
  "epoch": 20.0,
3
- "eval_accuracy": 0.9384550084889643,
4
- "eval_f1": 0.9269016697588126,
5
- "eval_loss": 0.30563604831695557,
6
- "eval_precision": 0.9190581309786607,
7
- "eval_recall": 0.9348802395209581,
8
- "eval_runtime": 8.103,
9
  "eval_samples": 100,
10
- "eval_samples_per_second": 12.341,
11
- "eval_steps_per_second": 0.864
12
  }
 
1
  {
2
  "epoch": 20.0,
3
+ "eval_accuracy": 0.9681663837011885,
4
+ "eval_f1": 0.9637654090399701,
5
+ "eval_loss": 0.18448814749717712,
6
+ "eval_precision": 0.9619686800894854,
7
+ "eval_recall": 0.9655688622754491,
8
+ "eval_runtime": 7.9644,
9
  "eval_samples": 100,
10
+ "eval_samples_per_second": 12.556,
11
+ "eval_steps_per_second": 0.879
12
  }
runs/May02_18-53-56_brutasse/events.out.tfevents.1651519687.brutasse.31018.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70da8dadbb44c976605002f3c4d73a5e5efae944bbd380049857b4aae33c6598
3
+ size 512
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 20.0,
3
- "train_loss": 0.7290411071777344,
4
- "train_runtime": 1777.9741,
5
  "train_samples": 800,
6
- "train_samples_per_second": 8.999,
7
- "train_steps_per_second": 0.562
8
  }
 
1
  {
2
  "epoch": 20.0,
3
+ "train_loss": 0.2323043212890625,
4
+ "train_runtime": 1807.2991,
5
  "train_samples": 800,
6
+ "train_samples_per_second": 8.853,
7
+ "train_steps_per_second": 0.553
8
  }
trainer_state.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "best_metric": 0.9269016697588126,
3
  "best_model_checkpoint": "layoutlmv3-test-cord/checkpoint-1000",
4
  "epoch": 20.0,
5
  "global_step": 1000,
@@ -9,144 +9,144 @@
9
  "log_history": [
10
  {
11
  "epoch": 2.0,
12
- "eval_accuracy": 0.6366723259762309,
13
- "eval_f1": 0.5626932325661285,
14
- "eval_loss": 1.6054191589355469,
15
- "eval_precision": 0.52,
16
- "eval_recall": 0.6130239520958084,
17
- "eval_runtime": 8.1419,
18
- "eval_samples_per_second": 12.282,
19
- "eval_steps_per_second": 0.86,
20
  "step": 100
21
  },
22
  {
23
  "epoch": 4.0,
24
- "eval_accuracy": 0.8314940577249575,
25
- "eval_f1": 0.8096632503660322,
26
- "eval_loss": 0.9171701669692993,
27
- "eval_precision": 0.7922636103151862,
28
- "eval_recall": 0.8278443113772455,
29
- "eval_runtime": 8.1716,
30
- "eval_samples_per_second": 12.238,
31
- "eval_steps_per_second": 0.857,
32
  "step": 200
33
  },
34
  {
35
  "epoch": 6.0,
36
- "eval_accuracy": 0.866723259762309,
37
- "eval_f1": 0.8496683861459101,
38
- "eval_loss": 0.6381942629814148,
39
- "eval_precision": 0.8367198838896952,
40
- "eval_recall": 0.8630239520958084,
41
- "eval_runtime": 8.0786,
42
- "eval_samples_per_second": 12.378,
43
- "eval_steps_per_second": 0.866,
44
  "step": 300
45
  },
46
  {
47
  "epoch": 8.0,
48
- "eval_accuracy": 0.8960101867572157,
49
- "eval_f1": 0.8775811209439529,
50
- "eval_loss": 0.4974280595779419,
51
- "eval_precision": 0.8648255813953488,
52
- "eval_recall": 0.8907185628742516,
53
- "eval_runtime": 8.0768,
54
- "eval_samples_per_second": 12.381,
55
- "eval_steps_per_second": 0.867,
56
  "step": 400
57
  },
58
  {
59
  "epoch": 10.0,
60
- "learning_rate": 5e-06,
61
- "loss": 1.1589,
62
  "step": 500
63
  },
64
  {
65
  "epoch": 10.0,
66
- "eval_accuracy": 0.916383701188455,
67
- "eval_f1": 0.8914243651085757,
68
- "eval_loss": 0.41238266229629517,
69
- "eval_precision": 0.8769007965242578,
70
- "eval_recall": 0.906437125748503,
71
- "eval_runtime": 8.1158,
72
- "eval_samples_per_second": 12.322,
73
- "eval_steps_per_second": 0.863,
74
  "step": 500
75
  },
76
  {
77
  "epoch": 12.0,
78
- "eval_accuracy": 0.9235993208828522,
79
- "eval_f1": 0.9064002959674436,
80
- "eval_loss": 0.37666961550712585,
81
- "eval_precision": 0.8961228968544257,
82
- "eval_recall": 0.9169161676646707,
83
- "eval_runtime": 8.1123,
84
- "eval_samples_per_second": 12.327,
85
- "eval_steps_per_second": 0.863,
86
  "step": 600
87
  },
88
  {
89
  "epoch": 14.0,
90
- "eval_accuracy": 0.933786078098472,
91
- "eval_f1": 0.9210818821785848,
92
- "eval_loss": 0.3387647271156311,
93
- "eval_precision": 0.9119589141599413,
94
- "eval_recall": 0.9303892215568862,
95
- "eval_runtime": 8.0833,
96
- "eval_samples_per_second": 12.371,
97
- "eval_steps_per_second": 0.866,
98
  "step": 700
99
  },
100
  {
101
  "epoch": 16.0,
102
- "eval_accuracy": 0.9393039049235993,
103
- "eval_f1": 0.927643784786642,
104
- "eval_loss": 0.3137519955635071,
105
- "eval_precision": 0.919793966151582,
106
- "eval_recall": 0.9356287425149701,
107
- "eval_runtime": 7.8384,
108
- "eval_samples_per_second": 12.758,
109
- "eval_steps_per_second": 0.893,
110
  "step": 800
111
  },
112
  {
113
  "epoch": 18.0,
114
- "eval_accuracy": 0.9376061120543294,
115
- "eval_f1": 0.9254174397031539,
116
- "eval_loss": 0.30727389454841614,
117
- "eval_precision": 0.9175864606328182,
118
- "eval_recall": 0.9333832335329342,
119
- "eval_runtime": 7.8694,
120
- "eval_samples_per_second": 12.708,
121
- "eval_steps_per_second": 0.89,
122
  "step": 900
123
  },
124
  {
125
  "epoch": 20.0,
126
  "learning_rate": 0.0,
127
- "loss": 0.2992,
128
  "step": 1000
129
  },
130
  {
131
  "epoch": 20.0,
132
- "eval_accuracy": 0.9384550084889643,
133
- "eval_f1": 0.9269016697588126,
134
- "eval_loss": 0.30563604831695557,
135
- "eval_precision": 0.9190581309786607,
136
- "eval_recall": 0.9348802395209581,
137
- "eval_runtime": 7.9092,
138
- "eval_samples_per_second": 12.644,
139
- "eval_steps_per_second": 0.885,
140
  "step": 1000
141
  },
142
  {
143
  "epoch": 20.0,
144
  "step": 1000,
145
  "total_flos": 4248755355648000.0,
146
- "train_loss": 0.7290411071777344,
147
- "train_runtime": 1777.9741,
148
- "train_samples_per_second": 8.999,
149
- "train_steps_per_second": 0.562
150
  }
151
  ],
152
  "max_steps": 1000,
 
1
  {
2
+ "best_metric": 0.9637654090399701,
3
  "best_model_checkpoint": "layoutlmv3-test-cord/checkpoint-1000",
4
  "epoch": 20.0,
5
  "global_step": 1000,
 
9
  "log_history": [
10
  {
11
  "epoch": 2.0,
12
+ "eval_accuracy": 0.8709677419354839,
13
+ "eval_f1": 0.838591342626559,
14
+ "eval_loss": 0.5257240533828735,
15
+ "eval_precision": 0.8223021582733813,
16
+ "eval_recall": 0.8555389221556886,
17
+ "eval_runtime": 8.1118,
18
+ "eval_samples_per_second": 12.328,
19
+ "eval_steps_per_second": 0.863,
20
  "step": 100
21
  },
22
  {
23
  "epoch": 4.0,
24
+ "eval_accuracy": 0.9316638370118846,
25
+ "eval_f1": 0.9198813056379822,
26
+ "eval_loss": 0.3200061023235321,
27
+ "eval_precision": 0.9117647058823529,
28
+ "eval_recall": 0.9281437125748503,
29
+ "eval_runtime": 7.9991,
30
+ "eval_samples_per_second": 12.501,
31
+ "eval_steps_per_second": 0.875,
32
  "step": 200
33
  },
34
  {
35
  "epoch": 6.0,
36
+ "eval_accuracy": 0.9465195246179966,
37
+ "eval_f1": 0.9360594795539035,
38
+ "eval_loss": 0.2448921650648117,
39
+ "eval_precision": 0.9298375184638109,
40
+ "eval_recall": 0.9423652694610778,
41
+ "eval_runtime": 7.8919,
42
+ "eval_samples_per_second": 12.671,
43
+ "eval_steps_per_second": 0.887,
44
  "step": 300
45
  },
46
  {
47
  "epoch": 8.0,
48
+ "eval_accuracy": 0.9596774193548387,
49
+ "eval_f1": 0.9503916449086163,
50
+ "eval_loss": 0.19232001900672913,
51
+ "eval_precision": 0.9472118959107807,
52
+ "eval_recall": 0.9535928143712575,
53
+ "eval_runtime": 8.0551,
54
+ "eval_samples_per_second": 12.415,
55
+ "eval_steps_per_second": 0.869,
56
  "step": 400
57
  },
58
  {
59
  "epoch": 10.0,
60
+ "learning_rate": 2.5e-05,
61
+ "loss": 0.4328,
62
  "step": 500
63
  },
64
  {
65
  "epoch": 10.0,
66
+ "eval_accuracy": 0.9681663837011885,
67
+ "eval_f1": 0.9623274897426333,
68
+ "eval_loss": 0.18568018078804016,
69
+ "eval_precision": 0.9591078066914498,
70
+ "eval_recall": 0.9655688622754491,
71
+ "eval_runtime": 8.1741,
72
+ "eval_samples_per_second": 12.234,
73
+ "eval_steps_per_second": 0.856,
74
  "step": 500
75
  },
76
  {
77
  "epoch": 12.0,
78
+ "eval_accuracy": 0.9656196943972836,
79
+ "eval_f1": 0.9607476635514018,
80
+ "eval_loss": 0.20726947486400604,
81
+ "eval_precision": 0.9596713965646004,
82
+ "eval_recall": 0.9618263473053892,
83
+ "eval_runtime": 8.2339,
84
+ "eval_samples_per_second": 12.145,
85
+ "eval_steps_per_second": 0.85,
86
  "step": 600
87
  },
88
  {
89
  "epoch": 14.0,
90
+ "eval_accuracy": 0.9702886247877759,
91
+ "eval_f1": 0.9648729446935724,
92
+ "eval_loss": 0.1803944706916809,
93
+ "eval_precision": 0.9634328358208956,
94
+ "eval_recall": 0.9663173652694611,
95
+ "eval_runtime": 7.9827,
96
+ "eval_samples_per_second": 12.527,
97
+ "eval_steps_per_second": 0.877,
98
  "step": 700
99
  },
100
  {
101
  "epoch": 16.0,
102
+ "eval_accuracy": 0.9664685908319185,
103
+ "eval_f1": 0.9640987284966342,
104
+ "eval_loss": 0.18818138539791107,
105
+ "eval_precision": 0.9633781763826607,
106
+ "eval_recall": 0.9648203592814372,
107
+ "eval_runtime": 8.0654,
108
+ "eval_samples_per_second": 12.399,
109
+ "eval_steps_per_second": 0.868,
110
  "step": 800
111
  },
112
  {
113
  "epoch": 18.0,
114
+ "eval_accuracy": 0.967741935483871,
115
+ "eval_f1": 0.9633781763826607,
116
+ "eval_loss": 0.1799575388431549,
117
+ "eval_precision": 0.9619402985074627,
118
+ "eval_recall": 0.9648203592814372,
119
+ "eval_runtime": 7.9449,
120
+ "eval_samples_per_second": 12.587,
121
+ "eval_steps_per_second": 0.881,
122
  "step": 900
123
  },
124
  {
125
  "epoch": 20.0,
126
  "learning_rate": 0.0,
127
+ "loss": 0.0318,
128
  "step": 1000
129
  },
130
  {
131
  "epoch": 20.0,
132
+ "eval_accuracy": 0.9681663837011885,
133
+ "eval_f1": 0.9637654090399701,
134
+ "eval_loss": 0.18448814749717712,
135
+ "eval_precision": 0.9619686800894854,
136
+ "eval_recall": 0.9655688622754491,
137
+ "eval_runtime": 8.0126,
138
+ "eval_samples_per_second": 12.48,
139
+ "eval_steps_per_second": 0.874,
140
  "step": 1000
141
  },
142
  {
143
  "epoch": 20.0,
144
  "step": 1000,
145
  "total_flos": 4248755355648000.0,
146
+ "train_loss": 0.2323043212890625,
147
+ "train_runtime": 1807.2991,
148
+ "train_samples_per_second": 8.853,
149
+ "train_steps_per_second": 0.553
150
  }
151
  ],
152
  "max_steps": 1000,