pavitemple
commited on
Commit
•
dc5c09c
1
Parent(s):
0132317
Model save
Browse files- README.md +2 -2
- all_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +208 -0
README.md
CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss:
|
21 |
-
- Accuracy: 0.
|
22 |
|
23 |
## Model description
|
24 |
|
|
|
17 |
|
18 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 1.7434
|
21 |
+
- Accuracy: 0.3333
|
22 |
|
23 |
## Model description
|
24 |
|
all_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 17.03,
|
3 |
+
"train_loss": 1.4402054105486188,
|
4 |
+
"train_runtime": 5962.8154,
|
5 |
+
"train_samples_per_second": 0.188,
|
6 |
+
"train_steps_per_second": 0.006
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 17.03,
|
3 |
+
"train_loss": 1.4402054105486188,
|
4 |
+
"train_runtime": 5962.8154,
|
5 |
+
"train_samples_per_second": 0.188,
|
6 |
+
"train_steps_per_second": 0.006
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,208 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.328125,
|
3 |
+
"best_model_checkpoint": "finetuned-Accident-MultipleLabels-Video-subset-v2-new3/checkpoint-4",
|
4 |
+
"epoch": 17.02857142857143,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 35,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.06,
|
13 |
+
"eval_accuracy": 0.171875,
|
14 |
+
"eval_loss": 1.9011348485946655,
|
15 |
+
"eval_runtime": 164.5439,
|
16 |
+
"eval_samples_per_second": 0.389,
|
17 |
+
"eval_steps_per_second": 0.049,
|
18 |
+
"step": 2
|
19 |
+
},
|
20 |
+
{
|
21 |
+
"epoch": 1.06,
|
22 |
+
"eval_accuracy": 0.328125,
|
23 |
+
"eval_loss": 1.8468830585479736,
|
24 |
+
"eval_runtime": 166.5047,
|
25 |
+
"eval_samples_per_second": 0.384,
|
26 |
+
"eval_steps_per_second": 0.048,
|
27 |
+
"step": 4
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 2.06,
|
31 |
+
"eval_accuracy": 0.328125,
|
32 |
+
"eval_loss": 1.853258728981018,
|
33 |
+
"eval_runtime": 165.3119,
|
34 |
+
"eval_samples_per_second": 0.387,
|
35 |
+
"eval_steps_per_second": 0.048,
|
36 |
+
"step": 6
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 3.06,
|
40 |
+
"eval_accuracy": 0.328125,
|
41 |
+
"eval_loss": 1.912185788154602,
|
42 |
+
"eval_runtime": 164.3063,
|
43 |
+
"eval_samples_per_second": 0.39,
|
44 |
+
"eval_steps_per_second": 0.049,
|
45 |
+
"step": 8
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 4.06,
|
49 |
+
"learning_rate": 4.032258064516129e-05,
|
50 |
+
"loss": 1.6429,
|
51 |
+
"step": 10
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 4.06,
|
55 |
+
"eval_accuracy": 0.3125,
|
56 |
+
"eval_loss": 1.9869797229766846,
|
57 |
+
"eval_runtime": 165.275,
|
58 |
+
"eval_samples_per_second": 0.387,
|
59 |
+
"eval_steps_per_second": 0.048,
|
60 |
+
"step": 10
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 5.06,
|
64 |
+
"eval_accuracy": 0.296875,
|
65 |
+
"eval_loss": 1.9921482801437378,
|
66 |
+
"eval_runtime": 164.3454,
|
67 |
+
"eval_samples_per_second": 0.389,
|
68 |
+
"eval_steps_per_second": 0.049,
|
69 |
+
"step": 12
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 6.06,
|
73 |
+
"eval_accuracy": 0.3125,
|
74 |
+
"eval_loss": 1.9682811498641968,
|
75 |
+
"eval_runtime": 165.1323,
|
76 |
+
"eval_samples_per_second": 0.388,
|
77 |
+
"eval_steps_per_second": 0.048,
|
78 |
+
"step": 14
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 7.06,
|
82 |
+
"eval_accuracy": 0.3125,
|
83 |
+
"eval_loss": 1.9313127994537354,
|
84 |
+
"eval_runtime": 165.2473,
|
85 |
+
"eval_samples_per_second": 0.387,
|
86 |
+
"eval_steps_per_second": 0.048,
|
87 |
+
"step": 16
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 8.06,
|
91 |
+
"eval_accuracy": 0.3125,
|
92 |
+
"eval_loss": 1.9203096628189087,
|
93 |
+
"eval_runtime": 165.3473,
|
94 |
+
"eval_samples_per_second": 0.387,
|
95 |
+
"eval_steps_per_second": 0.048,
|
96 |
+
"step": 18
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 9.06,
|
100 |
+
"learning_rate": 2.4193548387096777e-05,
|
101 |
+
"loss": 1.4144,
|
102 |
+
"step": 20
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 9.06,
|
106 |
+
"eval_accuracy": 0.3125,
|
107 |
+
"eval_loss": 1.9251445531845093,
|
108 |
+
"eval_runtime": 165.2928,
|
109 |
+
"eval_samples_per_second": 0.387,
|
110 |
+
"eval_steps_per_second": 0.048,
|
111 |
+
"step": 20
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 10.06,
|
115 |
+
"eval_accuracy": 0.296875,
|
116 |
+
"eval_loss": 1.931593656539917,
|
117 |
+
"eval_runtime": 164.0819,
|
118 |
+
"eval_samples_per_second": 0.39,
|
119 |
+
"eval_steps_per_second": 0.049,
|
120 |
+
"step": 22
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 11.06,
|
124 |
+
"eval_accuracy": 0.296875,
|
125 |
+
"eval_loss": 1.945461630821228,
|
126 |
+
"eval_runtime": 165.6035,
|
127 |
+
"eval_samples_per_second": 0.386,
|
128 |
+
"eval_steps_per_second": 0.048,
|
129 |
+
"step": 24
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 12.06,
|
133 |
+
"eval_accuracy": 0.296875,
|
134 |
+
"eval_loss": 1.972325325012207,
|
135 |
+
"eval_runtime": 164.9627,
|
136 |
+
"eval_samples_per_second": 0.388,
|
137 |
+
"eval_steps_per_second": 0.048,
|
138 |
+
"step": 26
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 13.06,
|
142 |
+
"eval_accuracy": 0.296875,
|
143 |
+
"eval_loss": 1.989672064781189,
|
144 |
+
"eval_runtime": 164.6204,
|
145 |
+
"eval_samples_per_second": 0.389,
|
146 |
+
"eval_steps_per_second": 0.049,
|
147 |
+
"step": 28
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 14.06,
|
151 |
+
"learning_rate": 8.064516129032258e-06,
|
152 |
+
"loss": 1.3307,
|
153 |
+
"step": 30
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 14.06,
|
157 |
+
"eval_accuracy": 0.296875,
|
158 |
+
"eval_loss": 2.000044584274292,
|
159 |
+
"eval_runtime": 163.2711,
|
160 |
+
"eval_samples_per_second": 0.392,
|
161 |
+
"eval_steps_per_second": 0.049,
|
162 |
+
"step": 30
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 15.06,
|
166 |
+
"eval_accuracy": 0.296875,
|
167 |
+
"eval_loss": 2.0062506198883057,
|
168 |
+
"eval_runtime": 163.8508,
|
169 |
+
"eval_samples_per_second": 0.391,
|
170 |
+
"eval_steps_per_second": 0.049,
|
171 |
+
"step": 32
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 16.06,
|
175 |
+
"eval_accuracy": 0.296875,
|
176 |
+
"eval_loss": 2.0072290897369385,
|
177 |
+
"eval_runtime": 164.7282,
|
178 |
+
"eval_samples_per_second": 0.389,
|
179 |
+
"eval_steps_per_second": 0.049,
|
180 |
+
"step": 34
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 17.03,
|
184 |
+
"eval_accuracy": 0.296875,
|
185 |
+
"eval_loss": 2.008248805999756,
|
186 |
+
"eval_runtime": 165.6489,
|
187 |
+
"eval_samples_per_second": 0.386,
|
188 |
+
"eval_steps_per_second": 0.048,
|
189 |
+
"step": 35
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 17.03,
|
193 |
+
"step": 35,
|
194 |
+
"total_flos": 1.3956545706511565e+18,
|
195 |
+
"train_loss": 1.4402054105486188,
|
196 |
+
"train_runtime": 5962.8154,
|
197 |
+
"train_samples_per_second": 0.188,
|
198 |
+
"train_steps_per_second": 0.006
|
199 |
+
}
|
200 |
+
],
|
201 |
+
"logging_steps": 10,
|
202 |
+
"max_steps": 35,
|
203 |
+
"num_train_epochs": 9223372036854775807,
|
204 |
+
"save_steps": 500,
|
205 |
+
"total_flos": 1.3956545706511565e+18,
|
206 |
+
"trial_name": null,
|
207 |
+
"trial_params": null
|
208 |
+
}
|