🍻 cheers
Browse files
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9924812030075187,
|
4 |
+
"eval_loss": 0.029002541676163673,
|
5 |
+
"eval_runtime": 3.0711,
|
6 |
+
"eval_samples_per_second": 43.308,
|
7 |
+
"eval_steps_per_second": 5.536,
|
8 |
+
"total_flos": 3.205097416476426e+17,
|
9 |
+
"train_loss": 0.128345954647431,
|
10 |
+
"train_runtime": 213.2669,
|
11 |
+
"train_samples_per_second": 19.394,
|
12 |
+
"train_steps_per_second": 1.219
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9924812030075187,
|
4 |
+
"eval_loss": 0.029002541676163673,
|
5 |
+
"eval_runtime": 3.0711,
|
6 |
+
"eval_samples_per_second": 43.308,
|
7 |
+
"eval_steps_per_second": 5.536
|
8 |
+
}
|
runs/Jul16_17-10-39_b4d52368873a/events.out.tfevents.1689527683.b4d52368873a.2423.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3704324925b39615277e74781d58522cc7777d207b9b7965b54365746120af52
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"total_flos": 3.205097416476426e+17,
|
4 |
+
"train_loss": 0.128345954647431,
|
5 |
+
"train_runtime": 213.2669,
|
6 |
+
"train_samples_per_second": 19.394,
|
7 |
+
"train_steps_per_second": 1.219
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,199 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.029002541676163673,
|
3 |
+
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"global_step": 260,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.15,
|
12 |
+
"learning_rate": 0.00019230769230769233,
|
13 |
+
"loss": 0.8284,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.31,
|
18 |
+
"learning_rate": 0.00018461538461538463,
|
19 |
+
"loss": 0.2956,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.46,
|
24 |
+
"learning_rate": 0.00017692307692307693,
|
25 |
+
"loss": 0.3667,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.62,
|
30 |
+
"learning_rate": 0.00016923076923076923,
|
31 |
+
"loss": 0.2713,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.77,
|
36 |
+
"learning_rate": 0.00016153846153846155,
|
37 |
+
"loss": 0.2125,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.92,
|
42 |
+
"learning_rate": 0.00015384615384615385,
|
43 |
+
"loss": 0.1852,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.08,
|
48 |
+
"learning_rate": 0.00014615384615384615,
|
49 |
+
"loss": 0.1254,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 1.23,
|
54 |
+
"learning_rate": 0.00013923076923076923,
|
55 |
+
"loss": 0.1552,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 1.38,
|
60 |
+
"learning_rate": 0.00013153846153846156,
|
61 |
+
"loss": 0.0474,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 1.54,
|
66 |
+
"learning_rate": 0.00012384615384615385,
|
67 |
+
"loss": 0.1265,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 1.54,
|
72 |
+
"eval_accuracy": 0.9774436090225563,
|
73 |
+
"eval_loss": 0.08259306848049164,
|
74 |
+
"eval_runtime": 2.625,
|
75 |
+
"eval_samples_per_second": 50.666,
|
76 |
+
"eval_steps_per_second": 6.476,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 1.69,
|
81 |
+
"learning_rate": 0.00011615384615384617,
|
82 |
+
"loss": 0.1154,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 1.85,
|
87 |
+
"learning_rate": 0.00010846153846153846,
|
88 |
+
"loss": 0.1548,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 2.0,
|
93 |
+
"learning_rate": 0.00010076923076923077,
|
94 |
+
"loss": 0.0739,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 2.15,
|
99 |
+
"learning_rate": 9.307692307692309e-05,
|
100 |
+
"loss": 0.0507,
|
101 |
+
"step": 140
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 2.31,
|
105 |
+
"learning_rate": 8.538461538461538e-05,
|
106 |
+
"loss": 0.0479,
|
107 |
+
"step": 150
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 2.46,
|
111 |
+
"learning_rate": 7.76923076923077e-05,
|
112 |
+
"loss": 0.0475,
|
113 |
+
"step": 160
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 2.62,
|
117 |
+
"learning_rate": 7e-05,
|
118 |
+
"loss": 0.0367,
|
119 |
+
"step": 170
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 2.77,
|
123 |
+
"learning_rate": 6.23076923076923e-05,
|
124 |
+
"loss": 0.0648,
|
125 |
+
"step": 180
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 2.92,
|
129 |
+
"learning_rate": 5.461538461538461e-05,
|
130 |
+
"loss": 0.0201,
|
131 |
+
"step": 190
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 3.08,
|
135 |
+
"learning_rate": 4.692307692307693e-05,
|
136 |
+
"loss": 0.0417,
|
137 |
+
"step": 200
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 3.08,
|
141 |
+
"eval_accuracy": 0.9924812030075187,
|
142 |
+
"eval_loss": 0.029002541676163673,
|
143 |
+
"eval_runtime": 2.6469,
|
144 |
+
"eval_samples_per_second": 50.247,
|
145 |
+
"eval_steps_per_second": 6.423,
|
146 |
+
"step": 200
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 3.23,
|
150 |
+
"learning_rate": 3.923076923076923e-05,
|
151 |
+
"loss": 0.0126,
|
152 |
+
"step": 210
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 3.38,
|
156 |
+
"learning_rate": 3.153846153846154e-05,
|
157 |
+
"loss": 0.0117,
|
158 |
+
"step": 220
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 3.54,
|
162 |
+
"learning_rate": 2.384615384615385e-05,
|
163 |
+
"loss": 0.0109,
|
164 |
+
"step": 230
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 3.69,
|
168 |
+
"learning_rate": 1.6153846153846154e-05,
|
169 |
+
"loss": 0.0123,
|
170 |
+
"step": 240
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 3.85,
|
174 |
+
"learning_rate": 8.461538461538462e-06,
|
175 |
+
"loss": 0.0113,
|
176 |
+
"step": 250
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 4.0,
|
180 |
+
"learning_rate": 7.692307692307694e-07,
|
181 |
+
"loss": 0.0106,
|
182 |
+
"step": 260
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 4.0,
|
186 |
+
"step": 260,
|
187 |
+
"total_flos": 3.205097416476426e+17,
|
188 |
+
"train_loss": 0.128345954647431,
|
189 |
+
"train_runtime": 213.2669,
|
190 |
+
"train_samples_per_second": 19.394,
|
191 |
+
"train_steps_per_second": 1.219
|
192 |
+
}
|
193 |
+
],
|
194 |
+
"max_steps": 260,
|
195 |
+
"num_train_epochs": 4,
|
196 |
+
"total_flos": 3.205097416476426e+17,
|
197 |
+
"trial_name": null,
|
198 |
+
"trial_params": null
|
199 |
+
}
|