End of training
Browse files- README.md +5 -3
- all_results.json +10 -10
- eval_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +84 -84
README.md
CHANGED
@@ -5,6 +5,8 @@ license: apache-2.0
|
|
5 |
metrics:
|
6 |
- accuracy
|
7 |
tags:
|
|
|
|
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: only-lora-beans-vit-base-patch16-224-in21k
|
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
# only-lora-beans-vit-base-patch16-224-in21k
|
18 |
|
19 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on
|
20 |
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 0.
|
22 |
-
- Accuracy: 0.
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
5 |
metrics:
|
6 |
- accuracy
|
7 |
tags:
|
8 |
+
- image-classification
|
9 |
+
- vision
|
10 |
- generated_from_trainer
|
11 |
model-index:
|
12 |
- name: only-lora-beans-vit-base-patch16-224-in21k
|
|
|
18 |
|
19 |
# only-lora-beans-vit-base-patch16-224-in21k
|
20 |
|
21 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
|
22 |
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 0.0752
|
24 |
+
- Accuracy: 0.9624
|
25 |
|
26 |
## Model description
|
27 |
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
"epoch": 10.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss":
|
5 |
-
"eval_runtime": 0.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
-
"total_flos": 8.
|
9 |
-
"train_loss":
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second":
|
13 |
}
|
|
|
1 |
{
|
2 |
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.9624060150375939,
|
4 |
+
"eval_loss": 0.0752067118883133,
|
5 |
+
"eval_runtime": 0.7899,
|
6 |
+
"eval_samples_per_second": 168.383,
|
7 |
+
"eval_steps_per_second": 21.523,
|
8 |
+
"total_flos": 8.508699063477043e+17,
|
9 |
+
"train_loss": 0.2738876885634202,
|
10 |
+
"train_runtime": 131.5423,
|
11 |
+
"train_samples_per_second": 78.606,
|
12 |
+
"train_steps_per_second": 9.883
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 10.0,
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_loss":
|
5 |
-
"eval_runtime": 0.
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 10.0,
|
3 |
+
"eval_accuracy": 0.9624060150375939,
|
4 |
+
"eval_loss": 0.0752067118883133,
|
5 |
+
"eval_runtime": 0.7899,
|
6 |
+
"eval_samples_per_second": 168.383,
|
7 |
+
"eval_steps_per_second": 21.523
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 10.0,
|
3 |
-
"total_flos": 8.
|
4 |
-
"train_loss":
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 10.0,
|
3 |
+
"total_flos": 8.508699063477043e+17,
|
4 |
+
"train_loss": 0.2738876885634202,
|
5 |
+
"train_runtime": 131.5423,
|
6 |
+
"train_samples_per_second": 78.606,
|
7 |
+
"train_steps_per_second": 9.883
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric":
|
3 |
-
"best_model_checkpoint": "./beans_outputs/checkpoint-
|
4 |
"epoch": 10.0,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 1300,
|
@@ -10,193 +10,193 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.7692307692307693,
|
13 |
-
"grad_norm":
|
14 |
"learning_rate": 0.005,
|
15 |
-
"loss": 0.
|
16 |
"step": 100
|
17 |
},
|
18 |
{
|
19 |
"epoch": 1.0,
|
20 |
-
"eval_accuracy": 0.
|
21 |
-
"eval_loss":
|
22 |
-
"eval_runtime": 0.
|
23 |
-
"eval_samples_per_second":
|
24 |
-
"eval_steps_per_second":
|
25 |
"step": 130
|
26 |
},
|
27 |
{
|
28 |
"epoch": 1.5384615384615383,
|
29 |
-
"grad_norm": 0.
|
30 |
"learning_rate": 0.005,
|
31 |
-
"loss":
|
32 |
"step": 200
|
33 |
},
|
34 |
{
|
35 |
"epoch": 2.0,
|
36 |
-
"eval_accuracy": 0.
|
37 |
-
"eval_loss":
|
38 |
-
"eval_runtime": 0.
|
39 |
-
"eval_samples_per_second":
|
40 |
-
"eval_steps_per_second":
|
41 |
"step": 260
|
42 |
},
|
43 |
{
|
44 |
"epoch": 2.3076923076923075,
|
45 |
-
"grad_norm":
|
46 |
"learning_rate": 0.005,
|
47 |
-
"loss":
|
48 |
"step": 300
|
49 |
},
|
50 |
{
|
51 |
"epoch": 3.0,
|
52 |
-
"eval_accuracy": 0.
|
53 |
-
"eval_loss":
|
54 |
-
"eval_runtime": 0.
|
55 |
-
"eval_samples_per_second":
|
56 |
-
"eval_steps_per_second":
|
57 |
"step": 390
|
58 |
},
|
59 |
{
|
60 |
"epoch": 3.076923076923077,
|
61 |
-
"grad_norm":
|
62 |
"learning_rate": 0.005,
|
63 |
-
"loss":
|
64 |
"step": 400
|
65 |
},
|
66 |
{
|
67 |
"epoch": 3.8461538461538463,
|
68 |
-
"grad_norm":
|
69 |
"learning_rate": 0.005,
|
70 |
-
"loss":
|
71 |
"step": 500
|
72 |
},
|
73 |
{
|
74 |
"epoch": 4.0,
|
75 |
-
"eval_accuracy": 0.
|
76 |
-
"eval_loss":
|
77 |
-
"eval_runtime": 0.
|
78 |
-
"eval_samples_per_second":
|
79 |
-
"eval_steps_per_second":
|
80 |
"step": 520
|
81 |
},
|
82 |
{
|
83 |
"epoch": 4.615384615384615,
|
84 |
-
"grad_norm":
|
85 |
"learning_rate": 0.005,
|
86 |
-
"loss":
|
87 |
"step": 600
|
88 |
},
|
89 |
{
|
90 |
"epoch": 5.0,
|
91 |
-
"eval_accuracy": 0.
|
92 |
-
"eval_loss":
|
93 |
-
"eval_runtime": 0.
|
94 |
-
"eval_samples_per_second":
|
95 |
-
"eval_steps_per_second":
|
96 |
"step": 650
|
97 |
},
|
98 |
{
|
99 |
"epoch": 5.384615384615385,
|
100 |
-
"grad_norm": 0.
|
101 |
"learning_rate": 0.005,
|
102 |
-
"loss":
|
103 |
"step": 700
|
104 |
},
|
105 |
{
|
106 |
"epoch": 6.0,
|
107 |
-
"eval_accuracy": 0.
|
108 |
-
"eval_loss":
|
109 |
-
"eval_runtime": 0.
|
110 |
-
"eval_samples_per_second":
|
111 |
-
"eval_steps_per_second":
|
112 |
"step": 780
|
113 |
},
|
114 |
{
|
115 |
"epoch": 6.153846153846154,
|
116 |
-
"grad_norm": 0.
|
117 |
"learning_rate": 0.005,
|
118 |
-
"loss":
|
119 |
"step": 800
|
120 |
},
|
121 |
{
|
122 |
"epoch": 6.923076923076923,
|
123 |
-
"grad_norm":
|
124 |
"learning_rate": 0.005,
|
125 |
-
"loss":
|
126 |
"step": 900
|
127 |
},
|
128 |
{
|
129 |
"epoch": 7.0,
|
130 |
-
"eval_accuracy": 0.
|
131 |
-
"eval_loss":
|
132 |
-
"eval_runtime": 0.
|
133 |
-
"eval_samples_per_second":
|
134 |
-
"eval_steps_per_second":
|
135 |
"step": 910
|
136 |
},
|
137 |
{
|
138 |
"epoch": 7.6923076923076925,
|
139 |
-
"grad_norm":
|
140 |
"learning_rate": 0.005,
|
141 |
-
"loss":
|
142 |
"step": 1000
|
143 |
},
|
144 |
{
|
145 |
"epoch": 8.0,
|
146 |
-
"eval_accuracy": 0.
|
147 |
-
"eval_loss":
|
148 |
-
"eval_runtime": 0.
|
149 |
-
"eval_samples_per_second":
|
150 |
-
"eval_steps_per_second":
|
151 |
"step": 1040
|
152 |
},
|
153 |
{
|
154 |
"epoch": 8.461538461538462,
|
155 |
-
"grad_norm": 0.
|
156 |
"learning_rate": 0.005,
|
157 |
-
"loss":
|
158 |
"step": 1100
|
159 |
},
|
160 |
{
|
161 |
"epoch": 9.0,
|
162 |
-
"eval_accuracy": 0.
|
163 |
-
"eval_loss":
|
164 |
-
"eval_runtime": 0.
|
165 |
-
"eval_samples_per_second":
|
166 |
-
"eval_steps_per_second":
|
167 |
"step": 1170
|
168 |
},
|
169 |
{
|
170 |
"epoch": 9.23076923076923,
|
171 |
-
"grad_norm":
|
172 |
"learning_rate": 0.005,
|
173 |
-
"loss":
|
174 |
"step": 1200
|
175 |
},
|
176 |
{
|
177 |
"epoch": 10.0,
|
178 |
-
"grad_norm": 0.
|
179 |
"learning_rate": 0.005,
|
180 |
-
"loss":
|
181 |
"step": 1300
|
182 |
},
|
183 |
{
|
184 |
"epoch": 10.0,
|
185 |
-
"eval_accuracy": 0.
|
186 |
-
"eval_loss":
|
187 |
-
"eval_runtime": 0.
|
188 |
-
"eval_samples_per_second":
|
189 |
-
"eval_steps_per_second":
|
190 |
"step": 1300
|
191 |
},
|
192 |
{
|
193 |
"epoch": 10.0,
|
194 |
"step": 1300,
|
195 |
-
"total_flos": 8.
|
196 |
-
"train_loss":
|
197 |
-
"train_runtime":
|
198 |
-
"train_samples_per_second":
|
199 |
-
"train_steps_per_second":
|
200 |
}
|
201 |
],
|
202 |
"logging_steps": 100,
|
@@ -216,7 +216,7 @@
|
|
216 |
"attributes": {}
|
217 |
}
|
218 |
},
|
219 |
-
"total_flos": 8.
|
220 |
"train_batch_size": 8,
|
221 |
"trial_name": null,
|
222 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.0752067118883133,
|
3 |
+
"best_model_checkpoint": "./beans_outputs/checkpoint-130",
|
4 |
"epoch": 10.0,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 1300,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.7692307692307693,
|
13 |
+
"grad_norm": 4.380956172943115,
|
14 |
"learning_rate": 0.005,
|
15 |
+
"loss": 0.4846,
|
16 |
"step": 100
|
17 |
},
|
18 |
{
|
19 |
"epoch": 1.0,
|
20 |
+
"eval_accuracy": 0.9624060150375939,
|
21 |
+
"eval_loss": 0.0752067118883133,
|
22 |
+
"eval_runtime": 0.7531,
|
23 |
+
"eval_samples_per_second": 176.6,
|
24 |
+
"eval_steps_per_second": 22.573,
|
25 |
"step": 130
|
26 |
},
|
27 |
{
|
28 |
"epoch": 1.5384615384615383,
|
29 |
+
"grad_norm": 0.7832342982292175,
|
30 |
"learning_rate": 0.005,
|
31 |
+
"loss": 0.3473,
|
32 |
"step": 200
|
33 |
},
|
34 |
{
|
35 |
"epoch": 2.0,
|
36 |
+
"eval_accuracy": 0.9548872180451128,
|
37 |
+
"eval_loss": 0.15986023843288422,
|
38 |
+
"eval_runtime": 0.7821,
|
39 |
+
"eval_samples_per_second": 170.05,
|
40 |
+
"eval_steps_per_second": 21.736,
|
41 |
"step": 260
|
42 |
},
|
43 |
{
|
44 |
"epoch": 2.3076923076923075,
|
45 |
+
"grad_norm": 1.9846404790878296,
|
46 |
"learning_rate": 0.005,
|
47 |
+
"loss": 0.2953,
|
48 |
"step": 300
|
49 |
},
|
50 |
{
|
51 |
"epoch": 3.0,
|
52 |
+
"eval_accuracy": 0.9548872180451128,
|
53 |
+
"eval_loss": 0.1191997155547142,
|
54 |
+
"eval_runtime": 0.8019,
|
55 |
+
"eval_samples_per_second": 165.864,
|
56 |
+
"eval_steps_per_second": 21.201,
|
57 |
"step": 390
|
58 |
},
|
59 |
{
|
60 |
"epoch": 3.076923076923077,
|
61 |
+
"grad_norm": 1.4397724866867065,
|
62 |
"learning_rate": 0.005,
|
63 |
+
"loss": 0.2891,
|
64 |
"step": 400
|
65 |
},
|
66 |
{
|
67 |
"epoch": 3.8461538461538463,
|
68 |
+
"grad_norm": 3.0908043384552,
|
69 |
"learning_rate": 0.005,
|
70 |
+
"loss": 0.2653,
|
71 |
"step": 500
|
72 |
},
|
73 |
{
|
74 |
"epoch": 4.0,
|
75 |
+
"eval_accuracy": 0.9398496240601504,
|
76 |
+
"eval_loss": 0.13930319249629974,
|
77 |
+
"eval_runtime": 0.7871,
|
78 |
+
"eval_samples_per_second": 168.974,
|
79 |
+
"eval_steps_per_second": 21.598,
|
80 |
"step": 520
|
81 |
},
|
82 |
{
|
83 |
"epoch": 4.615384615384615,
|
84 |
+
"grad_norm": 2.720701217651367,
|
85 |
"learning_rate": 0.005,
|
86 |
+
"loss": 0.2344,
|
87 |
"step": 600
|
88 |
},
|
89 |
{
|
90 |
"epoch": 5.0,
|
91 |
+
"eval_accuracy": 0.9624060150375939,
|
92 |
+
"eval_loss": 0.10011590272188187,
|
93 |
+
"eval_runtime": 0.8095,
|
94 |
+
"eval_samples_per_second": 164.306,
|
95 |
+
"eval_steps_per_second": 21.001,
|
96 |
"step": 650
|
97 |
},
|
98 |
{
|
99 |
"epoch": 5.384615384615385,
|
100 |
+
"grad_norm": 0.17460593581199646,
|
101 |
"learning_rate": 0.005,
|
102 |
+
"loss": 0.21,
|
103 |
"step": 700
|
104 |
},
|
105 |
{
|
106 |
"epoch": 6.0,
|
107 |
+
"eval_accuracy": 0.9624060150375939,
|
108 |
+
"eval_loss": 0.0893179401755333,
|
109 |
+
"eval_runtime": 0.7824,
|
110 |
+
"eval_samples_per_second": 169.993,
|
111 |
+
"eval_steps_per_second": 21.728,
|
112 |
"step": 780
|
113 |
},
|
114 |
{
|
115 |
"epoch": 6.153846153846154,
|
116 |
+
"grad_norm": 0.6050938963890076,
|
117 |
"learning_rate": 0.005,
|
118 |
+
"loss": 0.2397,
|
119 |
"step": 800
|
120 |
},
|
121 |
{
|
122 |
"epoch": 6.923076923076923,
|
123 |
+
"grad_norm": 1.7308460474014282,
|
124 |
"learning_rate": 0.005,
|
125 |
+
"loss": 0.3117,
|
126 |
"step": 900
|
127 |
},
|
128 |
{
|
129 |
"epoch": 7.0,
|
130 |
+
"eval_accuracy": 0.924812030075188,
|
131 |
+
"eval_loss": 0.1933087259531021,
|
132 |
+
"eval_runtime": 0.7988,
|
133 |
+
"eval_samples_per_second": 166.494,
|
134 |
+
"eval_steps_per_second": 21.281,
|
135 |
"step": 910
|
136 |
},
|
137 |
{
|
138 |
"epoch": 7.6923076923076925,
|
139 |
+
"grad_norm": 1.327934741973877,
|
140 |
"learning_rate": 0.005,
|
141 |
+
"loss": 0.2459,
|
142 |
"step": 1000
|
143 |
},
|
144 |
{
|
145 |
"epoch": 8.0,
|
146 |
+
"eval_accuracy": 0.924812030075188,
|
147 |
+
"eval_loss": 0.19011461734771729,
|
148 |
+
"eval_runtime": 0.7881,
|
149 |
+
"eval_samples_per_second": 168.75,
|
150 |
+
"eval_steps_per_second": 21.57,
|
151 |
"step": 1040
|
152 |
},
|
153 |
{
|
154 |
"epoch": 8.461538461538462,
|
155 |
+
"grad_norm": 0.8504745364189148,
|
156 |
"learning_rate": 0.005,
|
157 |
+
"loss": 0.25,
|
158 |
"step": 1100
|
159 |
},
|
160 |
{
|
161 |
"epoch": 9.0,
|
162 |
+
"eval_accuracy": 0.9699248120300752,
|
163 |
+
"eval_loss": 0.08675242960453033,
|
164 |
+
"eval_runtime": 0.7998,
|
165 |
+
"eval_samples_per_second": 166.287,
|
166 |
+
"eval_steps_per_second": 21.255,
|
167 |
"step": 1170
|
168 |
},
|
169 |
{
|
170 |
"epoch": 9.23076923076923,
|
171 |
+
"grad_norm": 3.838754177093506,
|
172 |
"learning_rate": 0.005,
|
173 |
+
"loss": 0.1834,
|
174 |
"step": 1200
|
175 |
},
|
176 |
{
|
177 |
"epoch": 10.0,
|
178 |
+
"grad_norm": 0.03880231827497482,
|
179 |
"learning_rate": 0.005,
|
180 |
+
"loss": 0.2038,
|
181 |
"step": 1300
|
182 |
},
|
183 |
{
|
184 |
"epoch": 10.0,
|
185 |
+
"eval_accuracy": 0.9473684210526315,
|
186 |
+
"eval_loss": 0.15278470516204834,
|
187 |
+
"eval_runtime": 0.8065,
|
188 |
+
"eval_samples_per_second": 164.915,
|
189 |
+
"eval_steps_per_second": 21.079,
|
190 |
"step": 1300
|
191 |
},
|
192 |
{
|
193 |
"epoch": 10.0,
|
194 |
"step": 1300,
|
195 |
+
"total_flos": 8.508699063477043e+17,
|
196 |
+
"train_loss": 0.2738876885634202,
|
197 |
+
"train_runtime": 131.5423,
|
198 |
+
"train_samples_per_second": 78.606,
|
199 |
+
"train_steps_per_second": 9.883
|
200 |
}
|
201 |
],
|
202 |
"logging_steps": 100,
|
|
|
216 |
"attributes": {}
|
217 |
}
|
218 |
},
|
219 |
+
"total_flos": 8.508699063477043e+17,
|
220 |
"train_batch_size": 8,
|
221 |
"trial_name": null,
|
222 |
"trial_params": null
|