End of training
Browse files- README.md +3 -3
- all_results.json +5 -5
- test_results.json +5 -6
- trainer_state.json +24 -150
README.md
CHANGED
@@ -17,9 +17,9 @@ This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingfac
|
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- eval_loss: 2.4447
|
19 |
- eval_accuracy: 0.0452
|
20 |
-
- eval_runtime:
|
21 |
-
- eval_samples_per_second:
|
22 |
-
- eval_steps_per_second: 2.
|
23 |
- step: 0
|
24 |
|
25 |
## Model description
|
|
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- eval_loss: 2.4447
|
19 |
- eval_accuracy: 0.0452
|
20 |
+
- eval_runtime: 27.1309
|
21 |
+
- eval_samples_per_second: 5.713
|
22 |
+
- eval_steps_per_second: 2.875
|
23 |
- step: 0
|
24 |
|
25 |
## Model description
|
all_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 3.23,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss":
|
5 |
-
"eval_runtime":
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.23,
|
3 |
+
"eval_accuracy": 0.04516129032258064,
|
4 |
+
"eval_loss": 2.444662094116211,
|
5 |
+
"eval_runtime": 27.1309,
|
6 |
+
"eval_samples_per_second": 5.713,
|
7 |
+
"eval_steps_per_second": 2.875
|
8 |
}
|
test_results.json
CHANGED
@@ -1,8 +1,7 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"
|
7 |
-
"eval_steps_per_second": 1.449
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"eval_accuracy": 0.04516129032258064,
|
3 |
+
"eval_loss": 2.444662094116211,
|
4 |
+
"eval_runtime": 27.1309,
|
5 |
+
"eval_samples_per_second": 5.713,
|
6 |
+
"eval_steps_per_second": 2.875
|
|
|
7 |
}
|
trainer_state.json
CHANGED
@@ -1,163 +1,37 @@
|
|
1 |
{
|
2 |
-
"best_metric":
|
3 |
-
"best_model_checkpoint":
|
4 |
-
"epoch":
|
5 |
-
"
|
|
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
-
"
|
12 |
-
"
|
13 |
-
"
|
14 |
-
"
|
|
|
|
|
15 |
},
|
16 |
{
|
17 |
-
"
|
18 |
-
"
|
19 |
-
"
|
20 |
-
"
|
21 |
-
|
22 |
-
|
23 |
-
"epoch": 0.2,
|
24 |
-
"learning_rate": 4.43609022556391e-05,
|
25 |
-
"loss": 2.1374,
|
26 |
-
"step": 30
|
27 |
-
},
|
28 |
-
{
|
29 |
-
"epoch": 0.26,
|
30 |
-
"eval_accuracy": 0.5714285714285714,
|
31 |
-
"eval_loss": 1.7412834167480469,
|
32 |
-
"eval_runtime": 6.1835,
|
33 |
-
"eval_samples_per_second": 11.32,
|
34 |
-
"eval_steps_per_second": 1.455,
|
35 |
-
"step": 38
|
36 |
-
},
|
37 |
-
{
|
38 |
-
"epoch": 1.01,
|
39 |
-
"learning_rate": 4.0601503759398494e-05,
|
40 |
-
"loss": 1.9183,
|
41 |
-
"step": 40
|
42 |
-
},
|
43 |
-
{
|
44 |
-
"epoch": 1.08,
|
45 |
-
"learning_rate": 3.6842105263157895e-05,
|
46 |
-
"loss": 1.4923,
|
47 |
-
"step": 50
|
48 |
-
},
|
49 |
-
{
|
50 |
-
"epoch": 1.15,
|
51 |
-
"learning_rate": 3.3082706766917295e-05,
|
52 |
-
"loss": 1.2149,
|
53 |
-
"step": 60
|
54 |
-
},
|
55 |
-
{
|
56 |
-
"epoch": 1.22,
|
57 |
-
"learning_rate": 2.9323308270676693e-05,
|
58 |
-
"loss": 0.7949,
|
59 |
-
"step": 70
|
60 |
-
},
|
61 |
-
{
|
62 |
-
"epoch": 1.26,
|
63 |
-
"eval_accuracy": 0.8,
|
64 |
-
"eval_loss": 0.7746667265892029,
|
65 |
-
"eval_runtime": 6.3565,
|
66 |
-
"eval_samples_per_second": 11.012,
|
67 |
-
"eval_steps_per_second": 1.416,
|
68 |
-
"step": 76
|
69 |
-
},
|
70 |
-
{
|
71 |
-
"epoch": 2.03,
|
72 |
-
"learning_rate": 2.556390977443609e-05,
|
73 |
-
"loss": 0.8088,
|
74 |
-
"step": 80
|
75 |
-
},
|
76 |
-
{
|
77 |
-
"epoch": 2.09,
|
78 |
-
"learning_rate": 2.1804511278195487e-05,
|
79 |
-
"loss": 0.535,
|
80 |
-
"step": 90
|
81 |
-
},
|
82 |
-
{
|
83 |
-
"epoch": 2.16,
|
84 |
-
"learning_rate": 1.8045112781954888e-05,
|
85 |
-
"loss": 0.5405,
|
86 |
-
"step": 100
|
87 |
-
},
|
88 |
-
{
|
89 |
-
"epoch": 2.23,
|
90 |
-
"learning_rate": 1.4285714285714285e-05,
|
91 |
-
"loss": 0.4279,
|
92 |
-
"step": 110
|
93 |
-
},
|
94 |
-
{
|
95 |
-
"epoch": 2.26,
|
96 |
-
"eval_accuracy": 0.9142857142857143,
|
97 |
-
"eval_loss": 0.4053409993648529,
|
98 |
-
"eval_runtime": 6.2263,
|
99 |
-
"eval_samples_per_second": 11.243,
|
100 |
-
"eval_steps_per_second": 1.445,
|
101 |
-
"step": 114
|
102 |
-
},
|
103 |
-
{
|
104 |
-
"epoch": 3.04,
|
105 |
-
"learning_rate": 1.0526315789473684e-05,
|
106 |
-
"loss": 0.3461,
|
107 |
-
"step": 120
|
108 |
-
},
|
109 |
-
{
|
110 |
-
"epoch": 3.11,
|
111 |
-
"learning_rate": 6.766917293233083e-06,
|
112 |
-
"loss": 0.2509,
|
113 |
-
"step": 130
|
114 |
-
},
|
115 |
-
{
|
116 |
-
"epoch": 3.18,
|
117 |
-
"learning_rate": 3.007518796992481e-06,
|
118 |
-
"loss": 0.291,
|
119 |
-
"step": 140
|
120 |
-
},
|
121 |
-
{
|
122 |
-
"epoch": 3.23,
|
123 |
-
"eval_accuracy": 0.9285714285714286,
|
124 |
-
"eval_loss": 0.3428815007209778,
|
125 |
-
"eval_runtime": 6.583,
|
126 |
-
"eval_samples_per_second": 10.633,
|
127 |
-
"eval_steps_per_second": 1.367,
|
128 |
-
"step": 148
|
129 |
-
},
|
130 |
-
{
|
131 |
-
"epoch": 3.23,
|
132 |
-
"step": 148,
|
133 |
-
"total_flos": 1.460491890402263e+18,
|
134 |
-
"train_loss": 1.0492687128685616,
|
135 |
-
"train_runtime": 239.2161,
|
136 |
-
"train_samples_per_second": 4.949,
|
137 |
-
"train_steps_per_second": 0.619
|
138 |
-
},
|
139 |
-
{
|
140 |
-
"epoch": 3.23,
|
141 |
-
"eval_accuracy": 0.864516129032258,
|
142 |
-
"eval_loss": 0.3992488980293274,
|
143 |
-
"eval_runtime": 14.1288,
|
144 |
-
"eval_samples_per_second": 10.971,
|
145 |
-
"eval_steps_per_second": 1.416,
|
146 |
-
"step": 148
|
147 |
-
},
|
148 |
-
{
|
149 |
-
"epoch": 3.23,
|
150 |
-
"eval_accuracy": 0.864516129032258,
|
151 |
-
"eval_loss": 0.3992488980293274,
|
152 |
-
"eval_runtime": 13.7986,
|
153 |
-
"eval_samples_per_second": 11.233,
|
154 |
-
"eval_steps_per_second": 1.449,
|
155 |
-
"step": 148
|
156 |
}
|
157 |
],
|
158 |
-
"
|
159 |
-
"
|
160 |
-
"
|
|
|
|
|
|
|
|
|
161 |
"trial_name": null,
|
162 |
"trial_params": null
|
163 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": null,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 0,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
+
"eval_accuracy": 0.04516129032258064,
|
13 |
+
"eval_loss": 2.444662094116211,
|
14 |
+
"eval_runtime": 31.3777,
|
15 |
+
"eval_samples_per_second": 4.94,
|
16 |
+
"eval_steps_per_second": 2.486,
|
17 |
+
"step": 0
|
18 |
},
|
19 |
{
|
20 |
+
"eval_accuracy": 0.04516129032258064,
|
21 |
+
"eval_loss": 2.444662094116211,
|
22 |
+
"eval_runtime": 27.1309,
|
23 |
+
"eval_samples_per_second": 5.713,
|
24 |
+
"eval_steps_per_second": 2.875,
|
25 |
+
"step": 0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
}
|
27 |
],
|
28 |
+
"logging_steps": 500,
|
29 |
+
"max_steps": 0,
|
30 |
+
"num_input_tokens_seen": 0,
|
31 |
+
"num_train_epochs": 0,
|
32 |
+
"save_steps": 500,
|
33 |
+
"total_flos": 0,
|
34 |
+
"train_batch_size": null,
|
35 |
"trial_name": null,
|
36 |
"trial_params": null
|
37 |
}
|