djbp commited on
Commit
4b901da
1 Parent(s): 029acc5

End of training

Browse files
README.md CHANGED
@@ -23,7 +23,7 @@ model-index:
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
- value: 0.8344402277039848
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,12 +33,12 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  This model is a fine-tuned version of [microsoft/swin-base-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-base-patch4-window7-224-in22k) on the imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 0.4016
37
- - Accuracy: 0.8344
38
- - Auc Overall: 0.9393
39
- - Auc Class 0: 0.9626
40
- - Auc Class 1: 0.9328
41
- - Auc Class 2: 0.9224
42
 
43
  ## Model description
44
 
 
23
  metrics:
24
  - name: Accuracy
25
  type: accuracy
26
+ value: 0.8349146110056926
27
  ---
28
 
29
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
33
 
34
  This model is a fine-tuned version of [microsoft/swin-base-patch4-window7-224-in22k](https://huggingface.co/microsoft/swin-base-patch4-window7-224-in22k) on the imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
+ - Loss: 0.4066
37
+ - Accuracy: 0.8349
38
+ - Auc Overall: 0.9379
39
+ - Auc Class 0: 0.9614
40
+ - Auc Class 1: 0.9315
41
+ - Auc Class 2: 0.9207
42
 
43
  ## Model description
44
 
all_results.json CHANGED
@@ -1,17 +1,17 @@
1
  {
2
- "epoch": 7.0,
3
- "eval_accuracy": 0.8600206611570248,
4
- "eval_auc_class_0": 0.9623798076923077,
5
- "eval_auc_class_1": 0.9106044580248364,
6
- "eval_auc_class_2": 0.9252550195053197,
7
- "eval_auc_overall": 0.932746428407488,
8
- "eval_loss": 0.358356237411499,
9
- "eval_runtime": 205.5157,
10
- "eval_samples_per_second": 9.42,
11
- "eval_steps_per_second": 0.078,
12
- "total_flos": 4.143902092898771e+18,
13
- "train_loss": 0.37870263712746755,
14
- "train_runtime": 11542.069,
15
- "train_samples_per_second": 4.583,
16
  "train_steps_per_second": 0.009
17
  }
 
1
  {
2
+ "epoch": 6.892307692307693,
3
+ "eval_accuracy": 0.8349146110056926,
4
+ "eval_auc_class_0": 0.9613559582309583,
5
+ "eval_auc_class_1": 0.9315019107361742,
6
+ "eval_auc_class_2": 0.920726636279075,
7
+ "eval_auc_overall": 0.9378615017487358,
8
+ "eval_loss": 0.40655088424682617,
9
+ "eval_runtime": 226.5129,
10
+ "eval_samples_per_second": 9.306,
11
+ "eval_steps_per_second": 0.075,
12
+ "total_flos": 4.4565045535859466e+18,
13
+ "train_loss": 0.45718009237732204,
14
+ "train_runtime": 12452.0227,
15
+ "train_samples_per_second": 4.634,
16
  "train_steps_per_second": 0.009
17
  }
eval_results.json CHANGED
@@ -1,12 +1,12 @@
1
  {
2
- "epoch": 7.0,
3
- "eval_accuracy": 0.8600206611570248,
4
- "eval_auc_class_0": 0.9623798076923077,
5
- "eval_auc_class_1": 0.9106044580248364,
6
- "eval_auc_class_2": 0.9252550195053197,
7
- "eval_auc_overall": 0.932746428407488,
8
- "eval_loss": 0.358356237411499,
9
- "eval_runtime": 205.5157,
10
- "eval_samples_per_second": 9.42,
11
- "eval_steps_per_second": 0.078
12
  }
 
1
  {
2
+ "epoch": 6.892307692307693,
3
+ "eval_accuracy": 0.8349146110056926,
4
+ "eval_auc_class_0": 0.9613559582309583,
5
+ "eval_auc_class_1": 0.9315019107361742,
6
+ "eval_auc_class_2": 0.920726636279075,
7
+ "eval_auc_overall": 0.9378615017487358,
8
+ "eval_loss": 0.40655088424682617,
9
+ "eval_runtime": 226.5129,
10
+ "eval_samples_per_second": 9.306,
11
+ "eval_steps_per_second": 0.075
12
  }
runs/Sep24_06-17-42_data-science-wbi/events.out.tfevents.1727171357.data-science-wbi ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:047bd9e0830e3ad29e864c1d2e8102a3d401da16c1375f6a49fd3870c829092d
3
+ size 138
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
- "epoch": 7.0,
3
- "total_flos": 4.143902092898771e+18,
4
- "train_loss": 0.37870263712746755,
5
- "train_runtime": 11542.069,
6
- "train_samples_per_second": 4.583,
7
  "train_steps_per_second": 0.009
8
  }
 
1
  {
2
+ "epoch": 6.892307692307693,
3
+ "total_flos": 4.4565045535859466e+18,
4
+ "train_loss": 0.45718009237732204,
5
+ "train_runtime": 12452.0227,
6
+ "train_samples_per_second": 4.634,
7
  "train_steps_per_second": 0.009
8
  }
trainer_state.json CHANGED
@@ -1,277 +1,284 @@
1
  {
2
- "best_metric": 0.8600206611570248,
3
- "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM/NMM_Classification_base_V10/checkpoint-90",
4
- "epoch": 7.0,
5
  "eval_steps": 500,
6
- "global_step": 105,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.6666666666666666,
13
- "grad_norm": 4.08638334274292,
14
- "learning_rate": 4.545454545454546e-05,
15
- "loss": 0.8429,
16
  "step": 10
17
  },
18
  {
19
- "epoch": 1.0,
20
- "step": 15,
21
- "train_accuracy": 0.828215987294865,
22
- "train_auc_class_0": 0.9164869255677786,
23
- "train_auc_class_1": 0.770497463263411,
24
- "train_auc_class_2": 0.8879816559016815,
25
- "train_auc_overall": 0.858322014910957,
26
- "train_loss": 0.4563606083393097,
27
- "train_runtime": 687.587,
28
- "train_samples_per_second": 10.989,
29
- "train_steps_per_second": 0.087
30
  },
31
  {
32
- "epoch": 1.0,
33
- "eval_accuracy": 0.8274793388429752,
34
- "eval_auc_class_0": 0.9340616414835164,
35
- "eval_auc_class_1": 0.7507519271533845,
36
- "eval_auc_class_2": 0.8937217290626166,
37
- "eval_auc_overall": 0.8595117658998391,
38
- "eval_loss": 0.49912166595458984,
39
- "eval_runtime": 208.8157,
40
- "eval_samples_per_second": 9.271,
41
- "eval_steps_per_second": 0.077,
42
- "step": 15
43
  },
44
  {
45
- "epoch": 1.3333333333333333,
46
- "grad_norm": 1.5998913049697876,
47
- "learning_rate": 4.5212765957446815e-05,
48
- "loss": 0.4716,
49
  "step": 20
50
  },
51
  {
52
- "epoch": 2.0,
53
- "grad_norm": 2.4967057704925537,
54
- "learning_rate": 3.9893617021276594e-05,
55
- "loss": 0.3887,
56
  "step": 30
57
  },
58
  {
59
- "epoch": 2.0,
60
- "step": 30,
61
- "train_accuracy": 0.8553467443091582,
62
- "train_auc_class_0": 0.9472708924769339,
63
- "train_auc_class_1": 0.9283657851250956,
64
- "train_auc_class_2": 0.9247901532600991,
65
- "train_auc_overall": 0.9334756102873761,
66
- "train_loss": 0.35952258110046387,
67
- "train_runtime": 685.7612,
68
- "train_samples_per_second": 11.018,
69
- "train_steps_per_second": 0.087
70
  },
71
  {
72
- "epoch": 2.0,
73
- "eval_accuracy": 0.8502066115702479,
74
- "eval_auc_class_0": 0.9477621336996336,
75
- "eval_auc_class_1": 0.9035008683616039,
76
- "eval_auc_class_2": 0.9096346486873113,
77
- "eval_auc_overall": 0.9202992169161829,
78
- "eval_loss": 0.3957720398902893,
79
- "eval_runtime": 208.9311,
80
- "eval_samples_per_second": 9.266,
81
- "eval_steps_per_second": 0.077,
82
- "step": 30
83
  },
84
  {
85
- "epoch": 2.6666666666666665,
86
- "grad_norm": 1.716306209564209,
87
- "learning_rate": 3.4574468085106386e-05,
88
- "loss": 0.3533,
89
  "step": 40
90
  },
91
  {
92
- "epoch": 3.0,
93
- "step": 45,
94
- "train_accuracy": 0.8750661725780836,
95
- "train_auc_class_0": 0.9582223466702152,
96
- "train_auc_class_1": 0.9392220276138971,
97
- "train_auc_class_2": 0.9390688589921182,
98
- "train_auc_overall": 0.9455044110920768,
99
- "train_loss": 0.3131944239139557,
100
- "train_runtime": 681.8829,
101
- "train_samples_per_second": 11.081,
102
- "train_steps_per_second": 0.088
103
  },
104
  {
105
- "epoch": 3.0,
106
- "eval_accuracy": 0.8589876033057852,
107
- "eval_auc_class_0": 0.9578196543040294,
108
- "eval_auc_class_1": 0.9092855930322143,
109
- "eval_auc_class_2": 0.920674647298944,
110
- "eval_auc_overall": 0.9292599648783959,
111
- "eval_loss": 0.37213757634162903,
112
- "eval_runtime": 207.6482,
113
- "eval_samples_per_second": 9.323,
114
- "eval_steps_per_second": 0.077,
115
- "step": 45
116
  },
117
  {
118
- "epoch": 3.3333333333333335,
119
- "grad_norm": 1.727489948272705,
120
- "learning_rate": 2.925531914893617e-05,
121
- "loss": 0.3208,
122
  "step": 50
123
  },
124
  {
125
- "epoch": 4.0,
126
- "grad_norm": 10.33216381072998,
127
- "learning_rate": 2.393617021276596e-05,
128
- "loss": 0.3092,
129
  "step": 60
130
  },
131
  {
132
  "epoch": 4.0,
133
- "step": 60,
134
- "train_accuracy": 0.8848597141344626,
135
- "train_auc_class_0": 0.9659353652856636,
136
- "train_auc_class_1": 0.9407955383480827,
137
- "train_auc_class_2": 0.946874731546507,
138
- "train_auc_overall": 0.9512018783934177,
139
- "train_loss": 0.29102450609207153,
140
- "train_runtime": 680.3084,
141
- "train_samples_per_second": 11.107,
142
- "train_steps_per_second": 0.088
143
  },
144
  {
145
  "epoch": 4.0,
146
- "eval_accuracy": 0.8579545454545454,
147
- "eval_auc_class_0": 0.9606026785714286,
148
- "eval_auc_class_1": 0.9083715281858423,
149
- "eval_auc_class_2": 0.9231885836547031,
150
- "eval_auc_overall": 0.9307209301373248,
151
- "eval_loss": 0.37202903628349304,
152
- "eval_runtime": 208.7004,
153
- "eval_samples_per_second": 9.276,
154
- "eval_steps_per_second": 0.077,
155
- "step": 60
156
  },
157
  {
158
- "epoch": 4.666666666666667,
159
- "grad_norm": 2.016211748123169,
160
- "learning_rate": 1.8617021276595745e-05,
161
- "loss": 0.2891,
162
  "step": 70
163
  },
164
  {
165
- "epoch": 5.0,
166
- "step": 75,
167
- "train_accuracy": 0.8913446267866596,
168
- "train_auc_class_0": 0.9681433160338301,
169
- "train_auc_class_1": 0.9542337553261225,
170
- "train_auc_class_2": 0.9506045992123575,
171
- "train_auc_overall": 0.9576605568574367,
172
- "train_loss": 0.2791108191013336,
173
- "train_runtime": 690.7325,
174
- "train_samples_per_second": 10.939,
175
- "train_steps_per_second": 0.087
176
  },
177
  {
178
- "epoch": 5.0,
179
- "eval_accuracy": 0.8584710743801653,
180
- "eval_auc_class_0": 0.9613524496336998,
181
- "eval_auc_class_1": 0.9090984273731954,
182
- "eval_auc_class_2": 0.9242633286351978,
183
- "eval_auc_overall": 0.9315714018806975,
184
- "eval_loss": 0.36684560775756836,
185
- "eval_runtime": 211.7912,
186
- "eval_samples_per_second": 9.141,
187
- "eval_steps_per_second": 0.076,
188
- "step": 75
189
  },
190
  {
191
- "epoch": 5.333333333333333,
192
- "grad_norm": 2.7578961849212646,
193
- "learning_rate": 1.3297872340425532e-05,
194
- "loss": 0.301,
195
- "step": 80
 
 
 
 
 
 
196
  },
197
  {
198
- "epoch": 6.0,
199
- "grad_norm": 2.526148557662964,
200
- "learning_rate": 7.97872340425532e-06,
201
- "loss": 0.2852,
202
  "step": 90
203
  },
204
  {
205
- "epoch": 6.0,
206
- "step": 90,
207
- "train_accuracy": 0.8955796717840127,
208
- "train_auc_class_0": 0.9719376552519517,
209
- "train_auc_class_1": 0.9603400183000109,
210
- "train_auc_class_2": 0.9560110428344394,
211
- "train_auc_overall": 0.9627629054621339,
212
- "train_loss": 0.2651444673538208,
213
- "train_runtime": 689.8251,
214
- "train_samples_per_second": 10.954,
215
  "train_steps_per_second": 0.087
216
  },
217
  {
218
- "epoch": 6.0,
219
- "eval_accuracy": 0.8600206611570248,
220
- "eval_auc_class_0": 0.9623798076923077,
221
- "eval_auc_class_1": 0.9106044580248364,
222
- "eval_auc_class_2": 0.9252550195053197,
223
- "eval_auc_overall": 0.932746428407488,
224
- "eval_loss": 0.358356237411499,
225
- "eval_runtime": 209.7236,
226
- "eval_samples_per_second": 9.231,
227
- "eval_steps_per_second": 0.076,
228
- "step": 90
229
  },
230
  {
231
- "epoch": 6.666666666666667,
232
- "grad_norm": 1.6473604440689087,
233
- "learning_rate": 2.6595744680851065e-06,
234
- "loss": 0.2804,
235
  "step": 100
236
  },
237
  {
238
- "epoch": 7.0,
239
- "step": 105,
240
- "train_accuracy": 0.8926680783483325,
241
- "train_auc_class_0": 0.9692773945765316,
242
- "train_auc_class_1": 0.9595504889107397,
243
- "train_auc_class_2": 0.9538150093690271,
244
- "train_auc_overall": 0.9608809642854328,
245
- "train_loss": 0.2725497782230377,
246
- "train_runtime": 674.8314,
247
- "train_samples_per_second": 11.197,
248
- "train_steps_per_second": 0.089
 
 
 
 
 
 
 
249
  },
250
  {
251
- "epoch": 7.0,
252
- "eval_accuracy": 0.8600206611570248,
253
- "eval_auc_class_0": 0.9628448374542126,
254
- "eval_auc_class_1": 0.90828447439095,
255
- "eval_auc_class_2": 0.924919084223066,
256
- "eval_auc_overall": 0.9320161320227428,
257
- "eval_loss": 0.3619421422481537,
258
- "eval_runtime": 204.9165,
259
- "eval_samples_per_second": 9.448,
260
- "eval_steps_per_second": 0.078,
261
- "step": 105
262
  },
263
  {
264
- "epoch": 7.0,
265
- "step": 105,
266
- "total_flos": 4.143902092898771e+18,
267
- "train_loss": 0.37870263712746755,
268
- "train_runtime": 11542.069,
269
- "train_samples_per_second": 4.583,
270
  "train_steps_per_second": 0.009
271
  }
272
  ],
273
  "logging_steps": 10,
274
- "max_steps": 105,
275
  "num_input_tokens_seen": 0,
276
  "num_train_epochs": 7,
277
  "save_steps": 500,
@@ -287,7 +294,7 @@
287
  "attributes": {}
288
  }
289
  },
290
- "total_flos": 4.143902092898771e+18,
291
  "train_batch_size": 128,
292
  "trial_name": null,
293
  "trial_params": null
 
1
  {
2
+ "best_metric": 0.8349146110056926,
3
+ "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM/NMM_Classification_base_V10/checkpoint-81",
4
+ "epoch": 6.892307692307693,
5
  "eval_steps": 500,
6
+ "global_step": 112,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.6153846153846154,
13
+ "grad_norm": 3.4730300903320312,
14
+ "learning_rate": 4.166666666666667e-05,
15
+ "loss": 0.8416,
16
  "step": 10
17
  },
18
  {
19
+ "epoch": 0.9846153846153847,
20
+ "step": 16,
21
+ "train_accuracy": 0.7838165716365401,
22
+ "train_auc_class_0": 0.9185630172913702,
23
+ "train_auc_class_1": 0.8867337604929844,
24
+ "train_auc_class_2": 0.8750511256900684,
25
+ "train_auc_overall": 0.8934493011581409,
26
+ "train_loss": 0.5212814211845398,
27
+ "train_runtime": 738.2478,
28
+ "train_samples_per_second": 11.166,
29
+ "train_steps_per_second": 0.088
30
  },
31
  {
32
+ "epoch": 0.9846153846153847,
33
+ "eval_accuracy": 0.7751423149905123,
34
+ "eval_auc_class_0": 0.9351696867321869,
35
+ "eval_auc_class_1": 0.9013257754334977,
36
+ "eval_auc_class_2": 0.896788439192459,
37
+ "eval_auc_overall": 0.9110946337860479,
38
+ "eval_loss": 0.5271921157836914,
39
+ "eval_runtime": 226.0415,
40
+ "eval_samples_per_second": 9.326,
41
+ "eval_steps_per_second": 0.075,
42
+ "step": 16
43
  },
44
  {
45
+ "epoch": 1.2307692307692308,
46
+ "grad_norm": 2.644275426864624,
47
+ "learning_rate": 4.600000000000001e-05,
48
+ "loss": 0.5698,
49
  "step": 20
50
  },
51
  {
52
+ "epoch": 1.8461538461538463,
53
+ "grad_norm": 2.8130509853363037,
54
+ "learning_rate": 4.1e-05,
55
+ "loss": 0.4812,
56
  "step": 30
57
  },
58
  {
59
+ "epoch": 1.9692307692307693,
60
+ "step": 32,
61
+ "train_accuracy": 0.8125682397185491,
62
+ "train_auc_class_0": 0.9453865517423163,
63
+ "train_auc_class_1": 0.9307891812846302,
64
+ "train_auc_class_2": 0.9068255493827921,
65
+ "train_auc_overall": 0.9276670941365796,
66
+ "train_loss": 0.45027220249176025,
67
+ "train_runtime": 742.3599,
68
+ "train_samples_per_second": 11.104,
69
+ "train_steps_per_second": 0.088
70
  },
71
  {
72
+ "epoch": 1.9692307692307693,
73
+ "eval_accuracy": 0.8088235294117647,
74
+ "eval_auc_class_0": 0.9525644963144962,
75
+ "eval_auc_class_1": 0.92725521408195,
76
+ "eval_auc_class_2": 0.9013002141426902,
77
+ "eval_auc_overall": 0.9270399748463788,
78
+ "eval_loss": 0.4628311097621918,
79
+ "eval_runtime": 227.1982,
80
+ "eval_samples_per_second": 9.278,
81
+ "eval_steps_per_second": 0.075,
82
+ "step": 32
83
  },
84
  {
85
+ "epoch": 2.4615384615384617,
86
+ "grad_norm": 4.537336349487305,
87
+ "learning_rate": 3.6e-05,
88
+ "loss": 0.4444,
89
  "step": 40
90
  },
91
  {
92
+ "epoch": 2.953846153846154,
93
+ "step": 48,
94
+ "train_accuracy": 0.8330704840470703,
95
+ "train_auc_class_0": 0.9573463611682219,
96
+ "train_auc_class_1": 0.9445183457629789,
97
+ "train_auc_class_2": 0.9234336270362926,
98
+ "train_auc_overall": 0.9417661113224979,
99
+ "train_loss": 0.3945094048976898,
100
+ "train_runtime": 749.6504,
101
+ "train_samples_per_second": 10.996,
102
+ "train_steps_per_second": 0.087
103
  },
104
  {
105
+ "epoch": 2.953846153846154,
106
+ "eval_accuracy": 0.8273244781783681,
107
+ "eval_auc_class_0": 0.9561783374283375,
108
+ "eval_auc_class_1": 0.9278154008848364,
109
+ "eval_auc_class_2": 0.9137749676178002,
110
+ "eval_auc_overall": 0.932589568643658,
111
+ "eval_loss": 0.42404672503471375,
112
+ "eval_runtime": 226.6025,
113
+ "eval_samples_per_second": 9.303,
114
+ "eval_steps_per_second": 0.075,
115
+ "step": 48
116
  },
117
  {
118
+ "epoch": 3.076923076923077,
119
+ "grad_norm": 1.965461254119873,
120
+ "learning_rate": 3.1e-05,
121
+ "loss": 0.4265,
122
  "step": 50
123
  },
124
  {
125
+ "epoch": 3.6923076923076925,
126
+ "grad_norm": 3.6663880348205566,
127
+ "learning_rate": 2.6000000000000002e-05,
128
+ "loss": 0.4055,
129
  "step": 60
130
  },
131
  {
132
  "epoch": 4.0,
133
+ "step": 65,
134
+ "train_accuracy": 0.8460511949532937,
135
+ "train_auc_class_0": 0.9634787762270021,
136
+ "train_auc_class_1": 0.9469802033822643,
137
+ "train_auc_class_2": 0.9308556192528558,
138
+ "train_auc_overall": 0.9471048662873741,
139
+ "train_loss": 0.37029632925987244,
140
+ "train_runtime": 746.5442,
141
+ "train_samples_per_second": 11.042,
142
+ "train_steps_per_second": 0.087
143
  },
144
  {
145
  "epoch": 4.0,
146
+ "eval_accuracy": 0.8287476280834914,
147
+ "eval_auc_class_0": 0.9613994676494677,
148
+ "eval_auc_class_1": 0.9314889691268666,
149
+ "eval_auc_class_2": 0.9206087950151214,
150
+ "eval_auc_overall": 0.9378324105971519,
151
+ "eval_loss": 0.41696038842201233,
152
+ "eval_runtime": 226.2202,
153
+ "eval_samples_per_second": 9.318,
154
+ "eval_steps_per_second": 0.075,
155
+ "step": 65
156
  },
157
  {
158
+ "epoch": 4.3076923076923075,
159
+ "grad_norm": 3.821587324142456,
160
+ "learning_rate": 2.1e-05,
161
+ "loss": 0.3875,
162
  "step": 70
163
  },
164
  {
165
+ "epoch": 4.923076923076923,
166
+ "grad_norm": 3.1884751319885254,
167
+ "learning_rate": 1.6000000000000003e-05,
168
+ "loss": 0.3914,
169
+ "step": 80
 
 
 
 
 
 
170
  },
171
  {
172
+ "epoch": 4.984615384615385,
173
+ "step": 81,
174
+ "train_accuracy": 0.8536940434307898,
175
+ "train_auc_class_0": 0.9671529040803416,
176
+ "train_auc_class_1": 0.9496355903643986,
177
+ "train_auc_class_2": 0.9367910485339515,
178
+ "train_auc_overall": 0.9511931809928972,
179
+ "train_loss": 0.3558380603790283,
180
+ "train_runtime": 752.3511,
181
+ "train_samples_per_second": 10.956,
182
+ "train_steps_per_second": 0.086
183
  },
184
  {
185
+ "epoch": 4.984615384615385,
186
+ "eval_accuracy": 0.8349146110056926,
187
+ "eval_auc_class_0": 0.9613559582309583,
188
+ "eval_auc_class_1": 0.9315019107361742,
189
+ "eval_auc_class_2": 0.920726636279075,
190
+ "eval_auc_overall": 0.9378615017487358,
191
+ "eval_loss": 0.40655088424682617,
192
+ "eval_runtime": 227.6818,
193
+ "eval_samples_per_second": 9.259,
194
+ "eval_steps_per_second": 0.075,
195
+ "step": 81
196
  },
197
  {
198
+ "epoch": 5.538461538461538,
199
+ "grad_norm": 2.0869638919830322,
200
+ "learning_rate": 1.1000000000000001e-05,
201
+ "loss": 0.3731,
202
  "step": 90
203
  },
204
  {
205
+ "epoch": 5.969230769230769,
206
+ "step": 97,
207
+ "train_accuracy": 0.8507824821060294,
208
+ "train_auc_class_0": 0.9680180491855133,
209
+ "train_auc_class_1": 0.9494889730661834,
210
+ "train_auc_class_2": 0.9379508502279332,
211
+ "train_auc_overall": 0.9518192908265433,
212
+ "train_loss": 0.35146117210388184,
213
+ "train_runtime": 745.296,
214
+ "train_samples_per_second": 11.06,
215
  "train_steps_per_second": 0.087
216
  },
217
  {
218
+ "epoch": 5.969230769230769,
219
+ "eval_accuracy": 0.8334914611005693,
220
+ "eval_auc_class_0": 0.9624193796068795,
221
+ "eval_auc_class_1": 0.9319585646645997,
222
+ "eval_auc_class_2": 0.9218326055186417,
223
+ "eval_auc_overall": 0.9387368499300403,
224
+ "eval_loss": 0.40432390570640564,
225
+ "eval_runtime": 226.3347,
226
+ "eval_samples_per_second": 9.314,
227
+ "eval_steps_per_second": 0.075,
228
+ "step": 97
229
  },
230
  {
231
+ "epoch": 6.153846153846154,
232
+ "grad_norm": 2.3311638832092285,
233
+ "learning_rate": 6e-06,
234
+ "loss": 0.357,
235
  "step": 100
236
  },
237
  {
238
+ "epoch": 6.769230769230769,
239
+ "grad_norm": 2.3013312816619873,
240
+ "learning_rate": 1.0000000000000002e-06,
241
+ "loss": 0.3687,
242
+ "step": 110
243
+ },
244
+ {
245
+ "epoch": 6.892307692307693,
246
+ "step": 112,
247
+ "train_accuracy": 0.8533300982651947,
248
+ "train_auc_class_0": 0.9690346749960462,
249
+ "train_auc_class_1": 0.9513212789842702,
250
+ "train_auc_class_2": 0.9389334461695297,
251
+ "train_auc_overall": 0.9530964667166154,
252
+ "train_loss": 0.34802600741386414,
253
+ "train_runtime": 740.1343,
254
+ "train_samples_per_second": 11.137,
255
+ "train_steps_per_second": 0.088
256
  },
257
  {
258
+ "epoch": 6.892307692307693,
259
+ "eval_accuracy": 0.8344402277039848,
260
+ "eval_auc_class_0": 0.9626446048321048,
261
+ "eval_auc_class_1": 0.9327683396469899,
262
+ "eval_auc_class_2": 0.9223551639104368,
263
+ "eval_auc_overall": 0.9392560361298439,
264
+ "eval_loss": 0.40155309438705444,
265
+ "eval_runtime": 226.5604,
266
+ "eval_samples_per_second": 9.304,
267
+ "eval_steps_per_second": 0.075,
268
+ "step": 112
269
  },
270
  {
271
+ "epoch": 6.892307692307693,
272
+ "step": 112,
273
+ "total_flos": 4.4565045535859466e+18,
274
+ "train_loss": 0.45718009237732204,
275
+ "train_runtime": 12452.0227,
276
+ "train_samples_per_second": 4.634,
277
  "train_steps_per_second": 0.009
278
  }
279
  ],
280
  "logging_steps": 10,
281
+ "max_steps": 112,
282
  "num_input_tokens_seen": 0,
283
  "num_train_epochs": 7,
284
  "save_steps": 500,
 
294
  "attributes": {}
295
  }
296
  },
297
+ "total_flos": 4.4565045535859466e+18,
298
  "train_batch_size": 128,
299
  "trial_name": null,
300
  "trial_params": null