🍻 cheers
Browse files- README.md +7 -6
- all_results.json +16 -0
- eval_results.json +11 -0
- runs/Mar21_08-16-46_45242e0ea5c0/events.out.tfevents.1711011309.45242e0ea5c0.20273.9 +3 -0
- train_results.json +8 -0
- trainer_state.json +391 -0
README.md
CHANGED
@@ -2,6 +2,7 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
metrics:
|
7 |
- accuracy
|
@@ -18,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
18 |
|
19 |
# vit-dropout-0.4
|
20 |
|
21 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on
|
22 |
It achieves the following results on the evaluation set:
|
23 |
-
- Loss: 0.
|
24 |
-
- Accuracy: 0.
|
25 |
-
- Precision: 0.
|
26 |
-
- Recall: 0.
|
27 |
-
- F1: 0.
|
28 |
|
29 |
## Model description
|
30 |
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: google/vit-base-patch16-224
|
4 |
tags:
|
5 |
+
- image-classification
|
6 |
- generated_from_trainer
|
7 |
metrics:
|
8 |
- accuracy
|
|
|
19 |
|
20 |
# vit-dropout-0.4
|
21 |
|
22 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the skin-cancer dataset.
|
23 |
It achieves the following results on the evaluation set:
|
24 |
+
- Loss: 0.4953
|
25 |
+
- Accuracy: 0.8402
|
26 |
+
- Precision: 0.8424
|
27 |
+
- Recall: 0.8402
|
28 |
+
- F1: 0.8404
|
29 |
|
30 |
## Model description
|
31 |
|
all_results.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.0,
|
3 |
+
"eval_accuracy": 0.8401525658807212,
|
4 |
+
"eval_f1": 0.8403607055183363,
|
5 |
+
"eval_loss": 0.49532002210617065,
|
6 |
+
"eval_precision": 0.8423956140728975,
|
7 |
+
"eval_recall": 0.8401525658807212,
|
8 |
+
"eval_runtime": 23.1598,
|
9 |
+
"eval_samples_per_second": 124.526,
|
10 |
+
"eval_steps_per_second": 15.587,
|
11 |
+
"total_flos": 7.550537882380222e+18,
|
12 |
+
"train_loss": 0.579371564758471,
|
13 |
+
"train_runtime": 2226.9019,
|
14 |
+
"train_samples_per_second": 230.275,
|
15 |
+
"train_steps_per_second": 14.415
|
16 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.0,
|
3 |
+
"eval_accuracy": 0.8401525658807212,
|
4 |
+
"eval_f1": 0.8403607055183363,
|
5 |
+
"eval_loss": 0.49532002210617065,
|
6 |
+
"eval_precision": 0.8423956140728975,
|
7 |
+
"eval_recall": 0.8401525658807212,
|
8 |
+
"eval_runtime": 23.1598,
|
9 |
+
"eval_samples_per_second": 124.526,
|
10 |
+
"eval_steps_per_second": 15.587
|
11 |
+
}
|
runs/Mar21_08-16-46_45242e0ea5c0/events.out.tfevents.1711011309.45242e0ea5c0.20273.9
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5925d00a503d0155e4efdd53d3216da08eb8189be62e8d5c8c10fd10b5ce0be5
|
3 |
+
size 560
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 19.0,
|
3 |
+
"total_flos": 7.550537882380222e+18,
|
4 |
+
"train_loss": 0.579371564758471,
|
5 |
+
"train_runtime": 2226.9019,
|
6 |
+
"train_samples_per_second": 230.275,
|
7 |
+
"train_steps_per_second": 14.415
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,391 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.49532002210617065,
|
3 |
+
"best_model_checkpoint": "./vit-dropout-0.4/checkpoint-2889",
|
4 |
+
"epoch": 19.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 6099,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 1.0,
|
13 |
+
"grad_norm": 20.27271842956543,
|
14 |
+
"learning_rate": 9.008498583569406e-05,
|
15 |
+
"loss": 1.5638,
|
16 |
+
"step": 321
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 1.0,
|
20 |
+
"eval_accuracy": 0.625866851595007,
|
21 |
+
"eval_f1": 0.6634599560471133,
|
22 |
+
"eval_loss": 0.917059063911438,
|
23 |
+
"eval_precision": 0.7843432532062684,
|
24 |
+
"eval_recall": 0.625866851595007,
|
25 |
+
"eval_runtime": 22.9894,
|
26 |
+
"eval_samples_per_second": 125.449,
|
27 |
+
"eval_steps_per_second": 15.703,
|
28 |
+
"step": 321
|
29 |
+
},
|
30 |
+
{
|
31 |
+
"epoch": 2.0,
|
32 |
+
"grad_norm": 2.8246476650238037,
|
33 |
+
"learning_rate": 9.967994652315143e-05,
|
34 |
+
"loss": 1.1092,
|
35 |
+
"step": 642
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 2.0,
|
39 |
+
"eval_accuracy": 0.7007628294036061,
|
40 |
+
"eval_f1": 0.7192900679194619,
|
41 |
+
"eval_loss": 0.7738657593727112,
|
42 |
+
"eval_precision": 0.7903482253631083,
|
43 |
+
"eval_recall": 0.7007628294036061,
|
44 |
+
"eval_runtime": 22.8225,
|
45 |
+
"eval_samples_per_second": 126.367,
|
46 |
+
"eval_steps_per_second": 15.818,
|
47 |
+
"step": 642
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 3.0,
|
51 |
+
"grad_norm": 4.247750282287598,
|
52 |
+
"learning_rate": 9.856371224544644e-05,
|
53 |
+
"loss": 0.9892,
|
54 |
+
"step": 963
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 3.0,
|
58 |
+
"eval_accuracy": 0.7780859916782247,
|
59 |
+
"eval_f1": 0.778423467149636,
|
60 |
+
"eval_loss": 0.6145934462547302,
|
61 |
+
"eval_precision": 0.790423999944827,
|
62 |
+
"eval_recall": 0.7780859916782247,
|
63 |
+
"eval_runtime": 22.6731,
|
64 |
+
"eval_samples_per_second": 127.199,
|
65 |
+
"eval_steps_per_second": 15.922,
|
66 |
+
"step": 963
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 4.0,
|
70 |
+
"grad_norm": 13.25033950805664,
|
71 |
+
"learning_rate": 9.666449571439611e-05,
|
72 |
+
"loss": 0.8735,
|
73 |
+
"step": 1284
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 4.0,
|
77 |
+
"eval_accuracy": 0.7454923717059639,
|
78 |
+
"eval_f1": 0.7615557397273891,
|
79 |
+
"eval_loss": 0.6262115836143494,
|
80 |
+
"eval_precision": 0.807493921529863,
|
81 |
+
"eval_recall": 0.7454923717059639,
|
82 |
+
"eval_runtime": 22.9609,
|
83 |
+
"eval_samples_per_second": 125.605,
|
84 |
+
"eval_steps_per_second": 15.722,
|
85 |
+
"step": 1284
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 5.0,
|
89 |
+
"grad_norm": 7.960892677307129,
|
90 |
+
"learning_rate": 9.401291758831328e-05,
|
91 |
+
"loss": 0.8118,
|
92 |
+
"step": 1605
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 5.0,
|
96 |
+
"eval_accuracy": 0.7163661581137309,
|
97 |
+
"eval_f1": 0.7311152376922296,
|
98 |
+
"eval_loss": 0.7255703806877136,
|
99 |
+
"eval_precision": 0.8185184282178283,
|
100 |
+
"eval_recall": 0.7163661581137309,
|
101 |
+
"eval_runtime": 22.9078,
|
102 |
+
"eval_samples_per_second": 125.896,
|
103 |
+
"eval_steps_per_second": 15.759,
|
104 |
+
"step": 1605
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 6.0,
|
108 |
+
"grad_norm": 4.861917972564697,
|
109 |
+
"learning_rate": 9.065172868916128e-05,
|
110 |
+
"loss": 0.7794,
|
111 |
+
"step": 1926
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 6.0,
|
115 |
+
"eval_accuracy": 0.7819001386962552,
|
116 |
+
"eval_f1": 0.7924960521456311,
|
117 |
+
"eval_loss": 0.608811616897583,
|
118 |
+
"eval_precision": 0.8200552259690934,
|
119 |
+
"eval_recall": 0.7819001386962552,
|
120 |
+
"eval_runtime": 22.5121,
|
121 |
+
"eval_samples_per_second": 128.109,
|
122 |
+
"eval_steps_per_second": 16.036,
|
123 |
+
"step": 1926
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 7.0,
|
127 |
+
"grad_norm": 8.067416191101074,
|
128 |
+
"learning_rate": 8.663512074027613e-05,
|
129 |
+
"loss": 0.6835,
|
130 |
+
"step": 2247
|
131 |
+
},
|
132 |
+
{
|
133 |
+
"epoch": 7.0,
|
134 |
+
"eval_accuracy": 0.7624826629680999,
|
135 |
+
"eval_f1": 0.778275921221607,
|
136 |
+
"eval_loss": 0.5834646821022034,
|
137 |
+
"eval_precision": 0.8169986016936643,
|
138 |
+
"eval_recall": 0.7624826629680999,
|
139 |
+
"eval_runtime": 23.3953,
|
140 |
+
"eval_samples_per_second": 123.273,
|
141 |
+
"eval_steps_per_second": 15.43,
|
142 |
+
"step": 2247
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 8.0,
|
146 |
+
"grad_norm": 3.7579257488250732,
|
147 |
+
"learning_rate": 8.202785264490498e-05,
|
148 |
+
"loss": 0.5997,
|
149 |
+
"step": 2568
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 8.0,
|
153 |
+
"eval_accuracy": 0.765256588072122,
|
154 |
+
"eval_f1": 0.7821382649754356,
|
155 |
+
"eval_loss": 0.6475569605827332,
|
156 |
+
"eval_precision": 0.826387274966978,
|
157 |
+
"eval_recall": 0.765256588072122,
|
158 |
+
"eval_runtime": 22.6325,
|
159 |
+
"eval_samples_per_second": 127.427,
|
160 |
+
"eval_steps_per_second": 15.95,
|
161 |
+
"step": 2568
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 9.0,
|
165 |
+
"grad_norm": 10.09793758392334,
|
166 |
+
"learning_rate": 7.692088630529957e-05,
|
167 |
+
"loss": 0.5814,
|
168 |
+
"step": 2889
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 9.0,
|
172 |
+
"eval_accuracy": 0.8401525658807212,
|
173 |
+
"eval_f1": 0.8403607055183363,
|
174 |
+
"eval_loss": 0.49532002210617065,
|
175 |
+
"eval_precision": 0.8423956140728975,
|
176 |
+
"eval_recall": 0.8401525658807212,
|
177 |
+
"eval_runtime": 25.1073,
|
178 |
+
"eval_samples_per_second": 114.867,
|
179 |
+
"eval_steps_per_second": 14.378,
|
180 |
+
"step": 2889
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 10.0,
|
184 |
+
"grad_norm": 7.585777282714844,
|
185 |
+
"learning_rate": 7.136468478571602e-05,
|
186 |
+
"loss": 0.5134,
|
187 |
+
"step": 3210
|
188 |
+
},
|
189 |
+
{
|
190 |
+
"epoch": 10.0,
|
191 |
+
"eval_accuracy": 0.8103328710124826,
|
192 |
+
"eval_f1": 0.8220041299869107,
|
193 |
+
"eval_loss": 0.5335156321525574,
|
194 |
+
"eval_precision": 0.8534486868065884,
|
195 |
+
"eval_recall": 0.8103328710124826,
|
196 |
+
"eval_runtime": 24.7217,
|
197 |
+
"eval_samples_per_second": 116.659,
|
198 |
+
"eval_steps_per_second": 14.603,
|
199 |
+
"step": 3210
|
200 |
+
},
|
201 |
+
{
|
202 |
+
"epoch": 11.0,
|
203 |
+
"grad_norm": 20.268178939819336,
|
204 |
+
"learning_rate": 6.546402506345253e-05,
|
205 |
+
"loss": 0.5109,
|
206 |
+
"step": 3531
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 11.0,
|
210 |
+
"eval_accuracy": 0.8124133148404993,
|
211 |
+
"eval_f1": 0.8191514417244506,
|
212 |
+
"eval_loss": 0.5497095584869385,
|
213 |
+
"eval_precision": 0.8414741317511313,
|
214 |
+
"eval_recall": 0.8124133148404993,
|
215 |
+
"eval_runtime": 23.8572,
|
216 |
+
"eval_samples_per_second": 120.886,
|
217 |
+
"eval_steps_per_second": 15.132,
|
218 |
+
"step": 3531
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 12.0,
|
222 |
+
"grad_norm": 3.8051164150238037,
|
223 |
+
"learning_rate": 5.931404220195339e-05,
|
224 |
+
"loss": 0.4073,
|
225 |
+
"step": 3852
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 12.0,
|
229 |
+
"eval_accuracy": 0.8311373092926491,
|
230 |
+
"eval_f1": 0.831620079776665,
|
231 |
+
"eval_loss": 0.5754452347755432,
|
232 |
+
"eval_precision": 0.8347554735526493,
|
233 |
+
"eval_recall": 0.8311373092926491,
|
234 |
+
"eval_runtime": 24.6815,
|
235 |
+
"eval_samples_per_second": 116.849,
|
236 |
+
"eval_steps_per_second": 14.626,
|
237 |
+
"step": 3852
|
238 |
+
},
|
239 |
+
{
|
240 |
+
"epoch": 13.0,
|
241 |
+
"grad_norm": 1.562414526939392,
|
242 |
+
"learning_rate": 5.30138910478373e-05,
|
243 |
+
"loss": 0.3255,
|
244 |
+
"step": 4173
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 13.0,
|
248 |
+
"eval_accuracy": 0.8574895977808599,
|
249 |
+
"eval_f1": 0.8525988995704912,
|
250 |
+
"eval_loss": 0.5593818426132202,
|
251 |
+
"eval_precision": 0.8620968298732496,
|
252 |
+
"eval_recall": 0.8574895977808599,
|
253 |
+
"eval_runtime": 24.0309,
|
254 |
+
"eval_samples_per_second": 120.012,
|
255 |
+
"eval_steps_per_second": 15.022,
|
256 |
+
"step": 4173
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 14.0,
|
260 |
+
"grad_norm": 0.7332964539527893,
|
261 |
+
"learning_rate": 4.666514757870119e-05,
|
262 |
+
"loss": 0.3288,
|
263 |
+
"step": 4494
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 14.0,
|
267 |
+
"eval_accuracy": 0.8332177531206657,
|
268 |
+
"eval_f1": 0.8401776032043191,
|
269 |
+
"eval_loss": 0.6329751014709473,
|
270 |
+
"eval_precision": 0.8607345037298127,
|
271 |
+
"eval_recall": 0.8332177531206657,
|
272 |
+
"eval_runtime": 23.7481,
|
273 |
+
"eval_samples_per_second": 121.441,
|
274 |
+
"eval_steps_per_second": 15.201,
|
275 |
+
"step": 4494
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 15.0,
|
279 |
+
"grad_norm": 5.967740535736084,
|
280 |
+
"learning_rate": 4.038959286483306e-05,
|
281 |
+
"loss": 0.2434,
|
282 |
+
"step": 4815
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 15.0,
|
286 |
+
"eval_accuracy": 0.8606102635228848,
|
287 |
+
"eval_f1": 0.861866915129384,
|
288 |
+
"eval_loss": 0.5198509693145752,
|
289 |
+
"eval_precision": 0.8645868176470374,
|
290 |
+
"eval_recall": 0.8606102635228848,
|
291 |
+
"eval_runtime": 23.048,
|
292 |
+
"eval_samples_per_second": 125.13,
|
293 |
+
"eval_steps_per_second": 15.663,
|
294 |
+
"step": 4815
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 16.0,
|
298 |
+
"grad_norm": 19.630001068115234,
|
299 |
+
"learning_rate": 3.424923705075116e-05,
|
300 |
+
"loss": 0.2185,
|
301 |
+
"step": 5136
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 16.0,
|
305 |
+
"eval_accuracy": 0.858876560332871,
|
306 |
+
"eval_f1": 0.860451647543394,
|
307 |
+
"eval_loss": 0.5324673652648926,
|
308 |
+
"eval_precision": 0.8647326759799664,
|
309 |
+
"eval_recall": 0.858876560332871,
|
310 |
+
"eval_runtime": 22.8629,
|
311 |
+
"eval_samples_per_second": 126.143,
|
312 |
+
"eval_steps_per_second": 15.79,
|
313 |
+
"step": 5136
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 17.0,
|
317 |
+
"grad_norm": 0.011605961248278618,
|
318 |
+
"learning_rate": 2.8362827388964596e-05,
|
319 |
+
"loss": 0.1707,
|
320 |
+
"step": 5457
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 17.0,
|
324 |
+
"eval_accuracy": 0.8640776699029126,
|
325 |
+
"eval_f1": 0.8597682191227608,
|
326 |
+
"eval_loss": 0.5524240136146545,
|
327 |
+
"eval_precision": 0.8638803900922614,
|
328 |
+
"eval_recall": 0.8640776699029126,
|
329 |
+
"eval_runtime": 22.516,
|
330 |
+
"eval_samples_per_second": 128.087,
|
331 |
+
"eval_steps_per_second": 16.033,
|
332 |
+
"step": 5457
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 18.0,
|
336 |
+
"grad_norm": 25.897785186767578,
|
337 |
+
"learning_rate": 2.2825269192267617e-05,
|
338 |
+
"loss": 0.1702,
|
339 |
+
"step": 5778
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 18.0,
|
343 |
+
"eval_accuracy": 0.8522884882108183,
|
344 |
+
"eval_f1": 0.8551640876798166,
|
345 |
+
"eval_loss": 0.5471655130386353,
|
346 |
+
"eval_precision": 0.8612118088331184,
|
347 |
+
"eval_recall": 0.8522884882108183,
|
348 |
+
"eval_runtime": 22.9729,
|
349 |
+
"eval_samples_per_second": 125.539,
|
350 |
+
"eval_steps_per_second": 15.714,
|
351 |
+
"step": 5778
|
352 |
+
},
|
353 |
+
{
|
354 |
+
"epoch": 19.0,
|
355 |
+
"grad_norm": 0.07399608939886093,
|
356 |
+
"learning_rate": 1.7725843316558628e-05,
|
357 |
+
"loss": 0.128,
|
358 |
+
"step": 6099
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 19.0,
|
362 |
+
"eval_accuracy": 0.8751733703190014,
|
363 |
+
"eval_f1": 0.8745515169095409,
|
364 |
+
"eval_loss": 0.559937596321106,
|
365 |
+
"eval_precision": 0.8758481332673859,
|
366 |
+
"eval_recall": 0.8751733703190014,
|
367 |
+
"eval_runtime": 22.9935,
|
368 |
+
"eval_samples_per_second": 125.427,
|
369 |
+
"eval_steps_per_second": 15.7,
|
370 |
+
"step": 6099
|
371 |
+
},
|
372 |
+
{
|
373 |
+
"epoch": 19.0,
|
374 |
+
"step": 6099,
|
375 |
+
"total_flos": 7.550537882380222e+18,
|
376 |
+
"train_loss": 0.579371564758471,
|
377 |
+
"train_runtime": 2226.9019,
|
378 |
+
"train_samples_per_second": 230.275,
|
379 |
+
"train_steps_per_second": 14.415
|
380 |
+
}
|
381 |
+
],
|
382 |
+
"logging_steps": 500,
|
383 |
+
"max_steps": 32100,
|
384 |
+
"num_input_tokens_seen": 0,
|
385 |
+
"num_train_epochs": 100,
|
386 |
+
"save_steps": 500,
|
387 |
+
"total_flos": 7.550537882380222e+18,
|
388 |
+
"train_batch_size": 16,
|
389 |
+
"trial_name": null,
|
390 |
+
"trial_params": null
|
391 |
+
}
|