synergyai-jaeung commited on
Commit
c1ebc49
1 Parent(s): 3031ed7

Training in progress, epoch 1

Browse files
all_results.json CHANGED
@@ -1,17 +1,17 @@
1
  {
2
  "epoch": 12.0,
3
- "eval_AUC": 0.9545454545454545,
4
- "eval_F1": 0.9735863095238095,
5
- "eval_Precision": 0.9896373056994818,
6
  "eval_Recall": 1.0,
7
- "eval_accuracy": 0.9906103286384976,
8
- "eval_loss": 0.048169028013944626,
9
- "eval_runtime": 4.3355,
10
- "eval_samples_per_second": 49.129,
11
- "eval_steps_per_second": 3.229,
12
- "total_flos": 7.894883901934633e+17,
13
- "train_loss": 0.09885871769101531,
14
- "train_runtime": 224.7245,
15
- "train_samples_per_second": 45.336,
16
- "train_steps_per_second": 2.884
17
  }
 
1
  {
2
  "epoch": 12.0,
3
+ "eval_AUC": 0.5,
4
+ "eval_F1": 0.47572815533980584,
5
+ "eval_Precision": 0.9074074074074074,
6
  "eval_Recall": 1.0,
7
+ "eval_accuracy": 0.9074074074074074,
8
+ "eval_loss": 0.3346473276615143,
9
+ "eval_runtime": 5.474,
10
+ "eval_samples_per_second": 59.189,
11
+ "eval_steps_per_second": 3.836,
12
+ "total_flos": 6.862690600268268e+17,
13
+ "train_loss": 0.03526242822408676,
14
+ "train_runtime": 251.233,
15
+ "train_samples_per_second": 35.25,
16
+ "train_steps_per_second": 2.245
17
  }
eval_results.json CHANGED
@@ -1,12 +1,12 @@
1
  {
2
  "epoch": 12.0,
3
- "eval_AUC": 0.9545454545454545,
4
- "eval_F1": 0.9735863095238095,
5
- "eval_Precision": 0.9896373056994818,
6
  "eval_Recall": 1.0,
7
- "eval_accuracy": 0.9906103286384976,
8
- "eval_loss": 0.048169028013944626,
9
- "eval_runtime": 4.3355,
10
- "eval_samples_per_second": 49.129,
11
- "eval_steps_per_second": 3.229
12
  }
 
1
  {
2
  "epoch": 12.0,
3
+ "eval_AUC": 0.5,
4
+ "eval_F1": 0.47572815533980584,
5
+ "eval_Precision": 0.9074074074074074,
6
  "eval_Recall": 1.0,
7
+ "eval_accuracy": 0.9074074074074074,
8
+ "eval_loss": 0.3346473276615143,
9
+ "eval_runtime": 5.474,
10
+ "eval_samples_per_second": 59.189,
11
+ "eval_steps_per_second": 3.836
12
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:af75b5d80d0a947cbd9ccf5fce3cf17bcc53c8b31f24138f4c52358b89462ed7
3
  size 343223968
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:760373362287b69643cb264294c3cdca3c69105ed8734ed2b2cb181e5b3d2411
3
  size 343223968
runs/May28_14-57-40_RTX3090/events.out.tfevents.1716876157.RTX3090.403009.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f12bff649f6e82654f424ea5b85d1e418260fe8ed73237ec9b7d9baddfbbe000
3
+ size 607
runs/May28_16-57-52_RTX3090/events.out.tfevents.1716883073.RTX3090.462376.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e62d86463f7c212e69f5465846e3d3f7c8215cf17e293ba833e7ddfc7a674f2c
3
+ size 5634
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 12.0,
3
- "total_flos": 7.894883901934633e+17,
4
- "train_loss": 0.09885871769101531,
5
- "train_runtime": 224.7245,
6
- "train_samples_per_second": 45.336,
7
- "train_steps_per_second": 2.884
8
  }
 
1
  {
2
  "epoch": 12.0,
3
+ "total_flos": 6.862690600268268e+17,
4
+ "train_loss": 0.03526242822408676,
5
+ "train_runtime": 251.233,
6
+ "train_samples_per_second": 35.25,
7
+ "train_steps_per_second": 2.245
8
  }
trainer_state.json CHANGED
@@ -1,195 +1,195 @@
1
  {
2
- "best_metric": 0.048169028013944626,
3
- "best_model_checkpoint": "google/vit-base-patch16-224-in21k_covid_19_ct_scans/checkpoint-432",
4
  "epoch": 12.0,
5
  "eval_steps": 500,
6
- "global_step": 648,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.018518518518518517,
13
- "grad_norm": 3.8348426818847656,
14
- "learning_rate": 0.00019969135802469138,
15
- "loss": 0.6804,
16
  "step": 1
17
  },
18
  {
19
  "epoch": 1.0,
20
  "eval_AUC": 0.5,
21
- "eval_F1": 0.47277227722772275,
22
- "eval_Precision": 0.8967136150234741,
23
  "eval_Recall": 1.0,
24
- "eval_accuracy": 0.8967136150234741,
25
- "eval_loss": 0.38209885358810425,
26
- "eval_runtime": 4.2504,
27
- "eval_samples_per_second": 50.113,
28
- "eval_steps_per_second": 3.294,
29
- "step": 54
30
  },
31
  {
32
  "epoch": 2.0,
33
- "eval_AUC": 0.5,
34
- "eval_F1": 0.47277227722772275,
35
- "eval_Precision": 0.8967136150234741,
36
- "eval_Recall": 1.0,
37
- "eval_accuracy": 0.8967136150234741,
38
- "eval_loss": 0.413409024477005,
39
- "eval_runtime": 4.3554,
40
- "eval_samples_per_second": 48.904,
41
- "eval_steps_per_second": 3.214,
42
- "step": 108
43
  },
44
  {
45
  "epoch": 3.0,
46
- "eval_AUC": 0.5454545454545454,
47
- "eval_F1": 0.5584577114427861,
48
- "eval_Precision": 0.9052132701421801,
49
  "eval_Recall": 1.0,
50
- "eval_accuracy": 0.9061032863849765,
51
- "eval_loss": 0.27077189087867737,
52
- "eval_runtime": 4.2214,
53
- "eval_samples_per_second": 50.458,
54
- "eval_steps_per_second": 3.316,
55
- "step": 162
56
  },
57
  {
58
  "epoch": 4.0,
59
- "eval_AUC": 0.7272727272727273,
60
- "eval_F1": 0.797271573604061,
61
- "eval_Precision": 0.9408866995073891,
62
- "eval_Recall": 1.0,
63
- "eval_accuracy": 0.9436619718309859,
64
- "eval_loss": 0.2405475229024887,
65
- "eval_runtime": 4.1493,
66
- "eval_samples_per_second": 51.335,
67
- "eval_steps_per_second": 3.374,
68
- "step": 216
69
  },
70
  {
71
  "epoch": 5.0,
72
- "eval_AUC": 0.7272727272727273,
73
- "eval_F1": 0.797271573604061,
74
- "eval_Precision": 0.9408866995073891,
75
- "eval_Recall": 1.0,
76
- "eval_accuracy": 0.9436619718309859,
77
- "eval_loss": 0.21931496262550354,
78
- "eval_runtime": 4.3053,
79
- "eval_samples_per_second": 49.474,
80
- "eval_steps_per_second": 3.252,
81
- "step": 270
82
  },
83
  {
84
  "epoch": 6.0,
85
- "eval_AUC": 0.9309852451213707,
86
- "eval_F1": 0.8774901960784314,
87
- "eval_Precision": 0.9891304347826086,
88
- "eval_Recall": 0.9528795811518325,
89
- "eval_accuracy": 0.9483568075117371,
90
- "eval_loss": 0.1719074547290802,
91
- "eval_runtime": 4.1609,
92
- "eval_samples_per_second": 51.191,
93
- "eval_steps_per_second": 3.365,
94
- "step": 324
95
  },
96
  {
97
  "epoch": 7.0,
98
- "eval_AUC": 0.9519276534983341,
99
- "eval_F1": 0.9611998299836055,
100
- "eval_Precision": 0.9895833333333334,
101
- "eval_Recall": 0.9947643979057592,
102
- "eval_accuracy": 0.9859154929577465,
103
- "eval_loss": 0.05248570814728737,
104
- "eval_runtime": 4.1623,
105
- "eval_samples_per_second": 51.174,
106
- "eval_steps_per_second": 3.364,
107
- "step": 378
108
  },
109
  {
110
  "epoch": 8.0,
111
- "eval_AUC": 0.9545454545454545,
112
- "eval_F1": 0.9735863095238095,
113
- "eval_Precision": 0.9896373056994818,
114
- "eval_Recall": 1.0,
115
- "eval_accuracy": 0.9906103286384976,
116
- "eval_loss": 0.048169028013944626,
117
- "eval_runtime": 4.2766,
118
- "eval_samples_per_second": 49.806,
119
- "eval_steps_per_second": 3.274,
120
- "step": 432
121
  },
122
  {
123
  "epoch": 9.0,
124
- "eval_AUC": 0.8863636363636364,
125
- "eval_F1": 0.9294374875770225,
126
- "eval_Precision": 0.9744897959183674,
127
- "eval_Recall": 1.0,
128
- "eval_accuracy": 0.9765258215962441,
129
- "eval_loss": 0.09067811071872711,
130
- "eval_runtime": 4.2163,
131
- "eval_samples_per_second": 50.519,
132
- "eval_steps_per_second": 3.32,
133
- "step": 486
134
  },
135
  {
136
- "epoch": 9.25925925925926,
137
- "grad_norm": 0.03957719728350639,
138
- "learning_rate": 4.567901234567901e-05,
139
- "loss": 0.1258,
140
- "step": 500
 
 
 
 
 
 
141
  },
142
  {
143
- "epoch": 10.0,
144
- "eval_AUC": 0.8863636363636364,
145
- "eval_F1": 0.9294374875770225,
146
- "eval_Precision": 0.9744897959183674,
147
- "eval_Recall": 1.0,
148
- "eval_accuracy": 0.9765258215962441,
149
- "eval_loss": 0.10092553496360779,
150
- "eval_runtime": 4.2027,
151
- "eval_samples_per_second": 50.682,
152
- "eval_steps_per_second": 3.331,
153
- "step": 540
154
  },
155
  {
156
  "epoch": 11.0,
157
- "eval_AUC": 0.8863636363636364,
158
- "eval_F1": 0.9294374875770225,
159
- "eval_Precision": 0.9744897959183674,
160
- "eval_Recall": 1.0,
161
- "eval_accuracy": 0.9765258215962441,
162
- "eval_loss": 0.10514290630817413,
163
- "eval_runtime": 4.2216,
164
- "eval_samples_per_second": 50.454,
165
- "eval_steps_per_second": 3.316,
166
- "step": 594
167
  },
168
  {
169
  "epoch": 12.0,
170
- "eval_AUC": 0.8863636363636364,
171
- "eval_F1": 0.9294374875770225,
172
- "eval_Precision": 0.9744897959183674,
173
- "eval_Recall": 1.0,
174
- "eval_accuracy": 0.9765258215962441,
175
- "eval_loss": 0.10642894357442856,
176
- "eval_runtime": 4.2244,
177
- "eval_samples_per_second": 50.421,
178
- "eval_steps_per_second": 3.314,
179
- "step": 648
180
  },
181
  {
182
  "epoch": 12.0,
183
- "step": 648,
184
- "total_flos": 7.894883901934633e+17,
185
- "train_loss": 0.09885871769101531,
186
- "train_runtime": 224.7245,
187
- "train_samples_per_second": 45.336,
188
- "train_steps_per_second": 2.884
189
  }
190
  ],
191
  "logging_steps": 500,
192
- "max_steps": 648,
193
  "num_input_tokens_seen": 0,
194
  "num_train_epochs": 12,
195
  "save_steps": 500,
@@ -205,7 +205,7 @@
205
  "attributes": {}
206
  }
207
  },
208
- "total_flos": 7.894883901934633e+17,
209
  "train_batch_size": 16,
210
  "trial_name": null,
211
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.3346473276615143,
3
+ "best_model_checkpoint": "google/vit-base-patch16-224-in21k_covid_19_ct_scans/checkpoint-47",
4
  "epoch": 12.0,
5
  "eval_steps": 500,
6
+ "global_step": 564,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.02127659574468085,
13
+ "grad_norm": 4.243431091308594,
14
+ "learning_rate": 0.000199645390070922,
15
+ "loss": 0.7767,
16
  "step": 1
17
  },
18
  {
19
  "epoch": 1.0,
20
  "eval_AUC": 0.5,
21
+ "eval_F1": 0.47572815533980584,
22
+ "eval_Precision": 0.9074074074074074,
23
  "eval_Recall": 1.0,
24
+ "eval_accuracy": 0.9074074074074074,
25
+ "eval_loss": 0.3346473276615143,
26
+ "eval_runtime": 5.5299,
27
+ "eval_samples_per_second": 58.591,
28
+ "eval_steps_per_second": 3.798,
29
+ "step": 47
30
  },
31
  {
32
  "epoch": 2.0,
33
+ "eval_AUC": 0.6204081632653061,
34
+ "eval_F1": 0.5919028340080972,
35
+ "eval_Precision": 0.9311594202898551,
36
+ "eval_Recall": 0.8741496598639455,
37
+ "eval_accuracy": 0.8271604938271605,
38
+ "eval_loss": 0.5513330698013306,
39
+ "eval_runtime": 5.5902,
40
+ "eval_samples_per_second": 57.959,
41
+ "eval_steps_per_second": 3.757,
42
+ "step": 94
43
  },
44
  {
45
  "epoch": 3.0,
46
+ "eval_AUC": 0.5,
47
+ "eval_F1": 0.47572815533980584,
48
+ "eval_Precision": 0.9074074074074074,
49
  "eval_Recall": 1.0,
50
+ "eval_accuracy": 0.9074074074074074,
51
+ "eval_loss": 0.4290040135383606,
52
+ "eval_runtime": 5.491,
53
+ "eval_samples_per_second": 59.005,
54
+ "eval_steps_per_second": 3.824,
55
+ "step": 141
56
  },
57
  {
58
  "epoch": 4.0,
59
+ "eval_AUC": 0.5965986394557823,
60
+ "eval_F1": 0.6365832614322692,
61
+ "eval_Precision": 0.9240506329113924,
62
+ "eval_Recall": 0.9931972789115646,
63
+ "eval_accuracy": 0.9197530864197531,
64
+ "eval_loss": 0.433300256729126,
65
+ "eval_runtime": 5.3718,
66
+ "eval_samples_per_second": 60.315,
67
+ "eval_steps_per_second": 3.909,
68
+ "step": 188
69
  },
70
  {
71
  "epoch": 5.0,
72
+ "eval_AUC": 0.5897959183673469,
73
+ "eval_F1": 0.618104667609618,
74
+ "eval_Precision": 0.9230769230769231,
75
+ "eval_Recall": 0.9795918367346939,
76
+ "eval_accuracy": 0.9074074074074074,
77
+ "eval_loss": 0.5040868520736694,
78
+ "eval_runtime": 5.759,
79
+ "eval_samples_per_second": 56.26,
80
+ "eval_steps_per_second": 3.646,
81
+ "step": 235
82
  },
83
  {
84
  "epoch": 6.0,
85
+ "eval_AUC": 0.5948979591836735,
86
+ "eval_F1": 0.6316786661614249,
87
+ "eval_Precision": 0.9238095238095239,
88
+ "eval_Recall": 0.9897959183673469,
89
+ "eval_accuracy": 0.9166666666666666,
90
+ "eval_loss": 0.4848092496395111,
91
+ "eval_runtime": 13.7532,
92
+ "eval_samples_per_second": 23.558,
93
+ "eval_steps_per_second": 1.527,
94
+ "step": 282
95
  },
96
  {
97
  "epoch": 7.0,
98
+ "eval_AUC": 0.5965986394557823,
99
+ "eval_F1": 0.6365832614322692,
100
+ "eval_Precision": 0.9240506329113924,
101
+ "eval_Recall": 0.9931972789115646,
102
+ "eval_accuracy": 0.9197530864197531,
103
+ "eval_loss": 0.48772257566452026,
104
+ "eval_runtime": 5.5683,
105
+ "eval_samples_per_second": 58.186,
106
+ "eval_steps_per_second": 3.771,
107
+ "step": 329
108
  },
109
  {
110
  "epoch": 8.0,
111
+ "eval_AUC": 0.5965986394557823,
112
+ "eval_F1": 0.6365832614322692,
113
+ "eval_Precision": 0.9240506329113924,
114
+ "eval_Recall": 0.9931972789115646,
115
+ "eval_accuracy": 0.9197530864197531,
116
+ "eval_loss": 0.49257075786590576,
117
+ "eval_runtime": 5.2567,
118
+ "eval_samples_per_second": 61.635,
119
+ "eval_steps_per_second": 3.995,
120
+ "step": 376
121
  },
122
  {
123
  "epoch": 9.0,
124
+ "eval_AUC": 0.5965986394557823,
125
+ "eval_F1": 0.6365832614322692,
126
+ "eval_Precision": 0.9240506329113924,
127
+ "eval_Recall": 0.9931972789115646,
128
+ "eval_accuracy": 0.9197530864197531,
129
+ "eval_loss": 0.49581947922706604,
130
+ "eval_runtime": 5.6218,
131
+ "eval_samples_per_second": 57.633,
132
+ "eval_steps_per_second": 3.735,
133
+ "step": 423
134
  },
135
  {
136
+ "epoch": 10.0,
137
+ "eval_AUC": 0.5965986394557823,
138
+ "eval_F1": 0.6365832614322692,
139
+ "eval_Precision": 0.9240506329113924,
140
+ "eval_Recall": 0.9931972789115646,
141
+ "eval_accuracy": 0.9197530864197531,
142
+ "eval_loss": 0.4980551600456238,
143
+ "eval_runtime": 5.5112,
144
+ "eval_samples_per_second": 58.79,
145
+ "eval_steps_per_second": 3.81,
146
+ "step": 470
147
  },
148
  {
149
+ "epoch": 10.638297872340425,
150
+ "grad_norm": 0.012480547651648521,
151
+ "learning_rate": 2.269503546099291e-05,
152
+ "loss": 0.0381,
153
+ "step": 500
 
 
 
 
 
 
154
  },
155
  {
156
  "epoch": 11.0,
157
+ "eval_AUC": 0.5965986394557823,
158
+ "eval_F1": 0.6365832614322692,
159
+ "eval_Precision": 0.9240506329113924,
160
+ "eval_Recall": 0.9931972789115646,
161
+ "eval_accuracy": 0.9197530864197531,
162
+ "eval_loss": 0.4996081292629242,
163
+ "eval_runtime": 5.5015,
164
+ "eval_samples_per_second": 58.893,
165
+ "eval_steps_per_second": 3.817,
166
+ "step": 517
167
  },
168
  {
169
  "epoch": 12.0,
170
+ "eval_AUC": 0.5965986394557823,
171
+ "eval_F1": 0.6365832614322692,
172
+ "eval_Precision": 0.9240506329113924,
173
+ "eval_Recall": 0.9931972789115646,
174
+ "eval_accuracy": 0.9197530864197531,
175
+ "eval_loss": 0.5001835823059082,
176
+ "eval_runtime": 5.5423,
177
+ "eval_samples_per_second": 58.46,
178
+ "eval_steps_per_second": 3.789,
179
+ "step": 564
180
  },
181
  {
182
  "epoch": 12.0,
183
+ "step": 564,
184
+ "total_flos": 6.862690600268268e+17,
185
+ "train_loss": 0.03526242822408676,
186
+ "train_runtime": 251.233,
187
+ "train_samples_per_second": 35.25,
188
+ "train_steps_per_second": 2.245
189
  }
190
  ],
191
  "logging_steps": 500,
192
+ "max_steps": 564,
193
  "num_input_tokens_seen": 0,
194
  "num_train_epochs": 12,
195
  "save_steps": 500,
 
205
  "attributes": {}
206
  }
207
  },
208
+ "total_flos": 6.862690600268268e+17,
209
  "train_batch_size": 16,
210
  "trial_name": null,
211
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:357df7b508e81337a045bc900ffa166dacc0301c45e85247db7b83a80952fefb
3
  size 4731
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:860ac9cc2700a6acca98170286a713aa05b047089895c0cc3d2037638dcd625e
3
  size 4731