heado
Browse files- README.md +4 -4
- all_results.json +4 -9
- model.safetensors +1 -1
- runs/Jul04_07-32-00_03d0faa5d47a/events.out.tfevents.1720078574.03d0faa5d47a.1790.0 +3 -0
- train_results.json +4 -4
- trainer_state.json +67 -67
- training_args.bin +1 -1
README.md
CHANGED
@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
19 |
|
20 |
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.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
-
- Loss: 0.
|
23 |
-
- Accuracy: 0
|
24 |
|
25 |
## Model description
|
26 |
|
@@ -52,8 +52,8 @@ The following hyperparameters were used during training:
|
|
52 |
|
53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
54 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
55 |
-
| 0.
|
56 |
-
| 0.
|
57 |
|
58 |
|
59 |
### Framework versions
|
|
|
19 |
|
20 |
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.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.0205
|
23 |
+
- Accuracy: 1.0
|
24 |
|
25 |
## Model description
|
26 |
|
|
|
52 |
|
53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
54 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
55 |
+
| 0.091 | 1.5385 | 100 | 0.0747 | 0.9774 |
|
56 |
+
| 0.036 | 3.0769 | 200 | 0.0205 | 1.0 |
|
57 |
|
58 |
|
59 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,13 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 4.0,
|
3 |
-
"eval_accuracy": 0.953125,
|
4 |
-
"eval_loss": 0.1842205822467804,
|
5 |
-
"eval_runtime": 1.4634,
|
6 |
-
"eval_samples_per_second": 87.465,
|
7 |
-
"eval_steps_per_second": 10.933,
|
8 |
"total_flos": 3.205097416476426e+17,
|
9 |
-
"train_loss": 0.
|
10 |
-
"train_runtime":
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 2.
|
13 |
}
|
|
|
1 |
{
|
2 |
"epoch": 4.0,
|
|
|
|
|
|
|
|
|
|
|
3 |
"total_flos": 3.205097416476426e+17,
|
4 |
+
"train_loss": 0.11695209254439061,
|
5 |
+
"train_runtime": 89.4129,
|
6 |
+
"train_samples_per_second": 46.257,
|
7 |
+
"train_steps_per_second": 2.908
|
8 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 343227052
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef21ae29af6781b332bf4ae417d3e2c6af87f6dfcb14dccb2bb0a78d6e8925f6
|
3 |
size 343227052
|
runs/Jul04_07-32-00_03d0faa5d47a/events.out.tfevents.1720078574.03d0faa5d47a.1790.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:92648a71a91c9c421018e6e20271a1293de1dea06e3ee31b3f603aecd7e4e9c7
|
3 |
+
size 11323
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 4.0,
|
3 |
"total_flos": 3.205097416476426e+17,
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 2.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 4.0,
|
3 |
"total_flos": 3.205097416476426e+17,
|
4 |
+
"train_loss": 0.11695209254439061,
|
5 |
+
"train_runtime": 89.4129,
|
6 |
+
"train_samples_per_second": 46.257,
|
7 |
+
"train_steps_per_second": 2.908
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-
|
4 |
"epoch": 4.0,
|
5 |
"eval_steps": 100,
|
6 |
"global_step": 260,
|
@@ -10,212 +10,212 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.15384615384615385,
|
13 |
-
"grad_norm":
|
14 |
"learning_rate": 0.00019230769230769233,
|
15 |
-
"loss": 0.
|
16 |
"step": 10
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.3076923076923077,
|
20 |
-
"grad_norm":
|
21 |
"learning_rate": 0.00018461538461538463,
|
22 |
-
"loss": 0.
|
23 |
"step": 20
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.46153846153846156,
|
27 |
-
"grad_norm":
|
28 |
"learning_rate": 0.00017692307692307693,
|
29 |
-
"loss": 0.
|
30 |
"step": 30
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.6153846153846154,
|
34 |
-
"grad_norm":
|
35 |
"learning_rate": 0.00016923076923076923,
|
36 |
-
"loss": 0.
|
37 |
"step": 40
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.7692307692307693,
|
41 |
-
"grad_norm": 0.
|
42 |
"learning_rate": 0.00016153846153846155,
|
43 |
-
"loss": 0.
|
44 |
"step": 50
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.9230769230769231,
|
48 |
-
"grad_norm": 4.
|
49 |
"learning_rate": 0.00015384615384615385,
|
50 |
-
"loss": 0.
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
"epoch": 1.0769230769230769,
|
55 |
-
"grad_norm":
|
56 |
"learning_rate": 0.00014615384615384615,
|
57 |
-
"loss": 0.
|
58 |
"step": 70
|
59 |
},
|
60 |
{
|
61 |
"epoch": 1.2307692307692308,
|
62 |
-
"grad_norm": 0.
|
63 |
"learning_rate": 0.00013846153846153847,
|
64 |
-
"loss": 0.
|
65 |
"step": 80
|
66 |
},
|
67 |
{
|
68 |
"epoch": 1.3846153846153846,
|
69 |
-
"grad_norm": 0.
|
70 |
"learning_rate": 0.00013076923076923077,
|
71 |
-
"loss": 0.
|
72 |
"step": 90
|
73 |
},
|
74 |
{
|
75 |
"epoch": 1.5384615384615383,
|
76 |
-
"grad_norm":
|
77 |
"learning_rate": 0.0001230769230769231,
|
78 |
-
"loss": 0.
|
79 |
"step": 100
|
80 |
},
|
81 |
{
|
82 |
"epoch": 1.5384615384615383,
|
83 |
"eval_accuracy": 0.9774436090225563,
|
84 |
-
"eval_loss": 0.
|
85 |
-
"eval_runtime": 1.
|
86 |
-
"eval_samples_per_second":
|
87 |
-
"eval_steps_per_second": 13.
|
88 |
"step": 100
|
89 |
},
|
90 |
{
|
91 |
"epoch": 1.6923076923076923,
|
92 |
-
"grad_norm": 0.
|
93 |
"learning_rate": 0.00011538461538461538,
|
94 |
-
"loss": 0.
|
95 |
"step": 110
|
96 |
},
|
97 |
{
|
98 |
"epoch": 1.8461538461538463,
|
99 |
-
"grad_norm": 0.
|
100 |
"learning_rate": 0.0001076923076923077,
|
101 |
-
"loss": 0.
|
102 |
"step": 120
|
103 |
},
|
104 |
{
|
105 |
"epoch": 2.0,
|
106 |
-
"grad_norm":
|
107 |
"learning_rate": 0.0001,
|
108 |
-
"loss": 0.
|
109 |
"step": 130
|
110 |
},
|
111 |
{
|
112 |
"epoch": 2.1538461538461537,
|
113 |
-
"grad_norm":
|
114 |
"learning_rate": 9.230769230769232e-05,
|
115 |
-
"loss": 0.
|
116 |
"step": 140
|
117 |
},
|
118 |
{
|
119 |
"epoch": 2.3076923076923075,
|
120 |
-
"grad_norm":
|
121 |
"learning_rate": 8.461538461538461e-05,
|
122 |
-
"loss": 0.
|
123 |
"step": 150
|
124 |
},
|
125 |
{
|
126 |
"epoch": 2.4615384615384617,
|
127 |
-
"grad_norm": 0.
|
128 |
"learning_rate": 7.692307692307693e-05,
|
129 |
-
"loss": 0.
|
130 |
"step": 160
|
131 |
},
|
132 |
{
|
133 |
"epoch": 2.6153846153846154,
|
134 |
-
"grad_norm": 0.
|
135 |
"learning_rate": 6.923076923076924e-05,
|
136 |
-
"loss": 0.
|
137 |
"step": 170
|
138 |
},
|
139 |
{
|
140 |
"epoch": 2.769230769230769,
|
141 |
-
"grad_norm": 0.
|
142 |
"learning_rate": 6.153846153846155e-05,
|
143 |
-
"loss": 0.
|
144 |
"step": 180
|
145 |
},
|
146 |
{
|
147 |
"epoch": 2.9230769230769234,
|
148 |
-
"grad_norm":
|
149 |
"learning_rate": 5.384615384615385e-05,
|
150 |
-
"loss": 0.
|
151 |
"step": 190
|
152 |
},
|
153 |
{
|
154 |
"epoch": 3.076923076923077,
|
155 |
-
"grad_norm": 0.
|
156 |
"learning_rate": 4.615384615384616e-05,
|
157 |
-
"loss": 0.
|
158 |
"step": 200
|
159 |
},
|
160 |
{
|
161 |
"epoch": 3.076923076923077,
|
162 |
-
"eval_accuracy": 0
|
163 |
-
"eval_loss": 0.
|
164 |
-
"eval_runtime": 1.
|
165 |
-
"eval_samples_per_second":
|
166 |
-
"eval_steps_per_second":
|
167 |
"step": 200
|
168 |
},
|
169 |
{
|
170 |
"epoch": 3.230769230769231,
|
171 |
-
"grad_norm": 0.
|
172 |
"learning_rate": 3.846153846153846e-05,
|
173 |
-
"loss": 0.
|
174 |
"step": 210
|
175 |
},
|
176 |
{
|
177 |
"epoch": 3.3846153846153846,
|
178 |
-
"grad_norm": 0.
|
179 |
"learning_rate": 3.0769230769230774e-05,
|
180 |
-
"loss": 0.
|
181 |
"step": 220
|
182 |
},
|
183 |
{
|
184 |
"epoch": 3.5384615384615383,
|
185 |
-
"grad_norm": 0.
|
186 |
"learning_rate": 2.307692307692308e-05,
|
187 |
-
"loss": 0.
|
188 |
"step": 230
|
189 |
},
|
190 |
{
|
191 |
"epoch": 3.6923076923076925,
|
192 |
-
"grad_norm": 0.
|
193 |
"learning_rate": 1.5384615384615387e-05,
|
194 |
-
"loss": 0.
|
195 |
"step": 240
|
196 |
},
|
197 |
{
|
198 |
"epoch": 3.8461538461538463,
|
199 |
-
"grad_norm": 0.
|
200 |
"learning_rate": 7.692307692307694e-06,
|
201 |
-
"loss": 0.
|
202 |
"step": 250
|
203 |
},
|
204 |
{
|
205 |
"epoch": 4.0,
|
206 |
-
"grad_norm": 0.
|
207 |
"learning_rate": 0.0,
|
208 |
-
"loss": 0.
|
209 |
"step": 260
|
210 |
},
|
211 |
{
|
212 |
"epoch": 4.0,
|
213 |
"step": 260,
|
214 |
"total_flos": 3.205097416476426e+17,
|
215 |
-
"train_loss": 0.
|
216 |
-
"train_runtime":
|
217 |
-
"train_samples_per_second":
|
218 |
-
"train_steps_per_second": 2.
|
219 |
}
|
220 |
],
|
221 |
"logging_steps": 10,
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.020527832210063934,
|
3 |
+
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
|
4 |
"epoch": 4.0,
|
5 |
"eval_steps": 100,
|
6 |
"global_step": 260,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.15384615384615385,
|
13 |
+
"grad_norm": 1.3131574392318726,
|
14 |
"learning_rate": 0.00019230769230769233,
|
15 |
+
"loss": 0.8368,
|
16 |
"step": 10
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.3076923076923077,
|
20 |
+
"grad_norm": 1.571279525756836,
|
21 |
"learning_rate": 0.00018461538461538463,
|
22 |
+
"loss": 0.3474,
|
23 |
"step": 20
|
24 |
},
|
25 |
{
|
26 |
"epoch": 0.46153846153846156,
|
27 |
+
"grad_norm": 3.746014356613159,
|
28 |
"learning_rate": 0.00017692307692307693,
|
29 |
+
"loss": 0.2139,
|
30 |
"step": 30
|
31 |
},
|
32 |
{
|
33 |
"epoch": 0.6153846153846154,
|
34 |
+
"grad_norm": 1.4054049253463745,
|
35 |
"learning_rate": 0.00016923076923076923,
|
36 |
+
"loss": 0.1799,
|
37 |
"step": 40
|
38 |
},
|
39 |
{
|
40 |
"epoch": 0.7692307692307693,
|
41 |
+
"grad_norm": 0.2104187309741974,
|
42 |
"learning_rate": 0.00016153846153846155,
|
43 |
+
"loss": 0.1475,
|
44 |
"step": 50
|
45 |
},
|
46 |
{
|
47 |
"epoch": 0.9230769230769231,
|
48 |
+
"grad_norm": 4.926904678344727,
|
49 |
"learning_rate": 0.00015384615384615385,
|
50 |
+
"loss": 0.2264,
|
51 |
"step": 60
|
52 |
},
|
53 |
{
|
54 |
"epoch": 1.0769230769230769,
|
55 |
+
"grad_norm": 3.4138243198394775,
|
56 |
"learning_rate": 0.00014615384615384615,
|
57 |
+
"loss": 0.186,
|
58 |
"step": 70
|
59 |
},
|
60 |
{
|
61 |
"epoch": 1.2307692307692308,
|
62 |
+
"grad_norm": 0.18232814967632294,
|
63 |
"learning_rate": 0.00013846153846153847,
|
64 |
+
"loss": 0.1536,
|
65 |
"step": 80
|
66 |
},
|
67 |
{
|
68 |
"epoch": 1.3846153846153846,
|
69 |
+
"grad_norm": 0.23882938921451569,
|
70 |
"learning_rate": 0.00013076923076923077,
|
71 |
+
"loss": 0.0561,
|
72 |
"step": 90
|
73 |
},
|
74 |
{
|
75 |
"epoch": 1.5384615384615383,
|
76 |
+
"grad_norm": 1.603104829788208,
|
77 |
"learning_rate": 0.0001230769230769231,
|
78 |
+
"loss": 0.091,
|
79 |
"step": 100
|
80 |
},
|
81 |
{
|
82 |
"epoch": 1.5384615384615383,
|
83 |
"eval_accuracy": 0.9774436090225563,
|
84 |
+
"eval_loss": 0.07465264946222305,
|
85 |
+
"eval_runtime": 1.2611,
|
86 |
+
"eval_samples_per_second": 105.465,
|
87 |
+
"eval_steps_per_second": 13.48,
|
88 |
"step": 100
|
89 |
},
|
90 |
{
|
91 |
"epoch": 1.6923076923076923,
|
92 |
+
"grad_norm": 0.11675859242677689,
|
93 |
"learning_rate": 0.00011538461538461538,
|
94 |
+
"loss": 0.0952,
|
95 |
"step": 110
|
96 |
},
|
97 |
{
|
98 |
"epoch": 1.8461538461538463,
|
99 |
+
"grad_norm": 0.11038266867399216,
|
100 |
"learning_rate": 0.0001076923076923077,
|
101 |
+
"loss": 0.1683,
|
102 |
"step": 120
|
103 |
},
|
104 |
{
|
105 |
"epoch": 2.0,
|
106 |
+
"grad_norm": 1.329042673110962,
|
107 |
"learning_rate": 0.0001,
|
108 |
+
"loss": 0.0489,
|
109 |
"step": 130
|
110 |
},
|
111 |
{
|
112 |
"epoch": 2.1538461538461537,
|
113 |
+
"grad_norm": 0.46014711260795593,
|
114 |
"learning_rate": 9.230769230769232e-05,
|
115 |
+
"loss": 0.0307,
|
116 |
"step": 140
|
117 |
},
|
118 |
{
|
119 |
"epoch": 2.3076923076923075,
|
120 |
+
"grad_norm": 0.6358932852745056,
|
121 |
"learning_rate": 8.461538461538461e-05,
|
122 |
+
"loss": 0.0254,
|
123 |
"step": 150
|
124 |
},
|
125 |
{
|
126 |
"epoch": 2.4615384615384617,
|
127 |
+
"grad_norm": 0.07531006634235382,
|
128 |
"learning_rate": 7.692307692307693e-05,
|
129 |
+
"loss": 0.0175,
|
130 |
"step": 160
|
131 |
},
|
132 |
{
|
133 |
"epoch": 2.6153846153846154,
|
134 |
+
"grad_norm": 0.07003136724233627,
|
135 |
"learning_rate": 6.923076923076924e-05,
|
136 |
+
"loss": 0.0172,
|
137 |
"step": 170
|
138 |
},
|
139 |
{
|
140 |
"epoch": 2.769230769230769,
|
141 |
+
"grad_norm": 0.08160967379808426,
|
142 |
"learning_rate": 6.153846153846155e-05,
|
143 |
+
"loss": 0.0435,
|
144 |
"step": 180
|
145 |
},
|
146 |
{
|
147 |
"epoch": 2.9230769230769234,
|
148 |
+
"grad_norm": 0.09463630616664886,
|
149 |
"learning_rate": 5.384615384615385e-05,
|
150 |
+
"loss": 0.0181,
|
151 |
"step": 190
|
152 |
},
|
153 |
{
|
154 |
"epoch": 3.076923076923077,
|
155 |
+
"grad_norm": 0.06692061573266983,
|
156 |
"learning_rate": 4.615384615384616e-05,
|
157 |
+
"loss": 0.036,
|
158 |
"step": 200
|
159 |
},
|
160 |
{
|
161 |
"epoch": 3.076923076923077,
|
162 |
+
"eval_accuracy": 1.0,
|
163 |
+
"eval_loss": 0.020527832210063934,
|
164 |
+
"eval_runtime": 1.2254,
|
165 |
+
"eval_samples_per_second": 108.536,
|
166 |
+
"eval_steps_per_second": 13.873,
|
167 |
"step": 200
|
168 |
},
|
169 |
{
|
170 |
"epoch": 3.230769230769231,
|
171 |
+
"grad_norm": 0.057732608169317245,
|
172 |
"learning_rate": 3.846153846153846e-05,
|
173 |
+
"loss": 0.0228,
|
174 |
"step": 210
|
175 |
},
|
176 |
{
|
177 |
"epoch": 3.3846153846153846,
|
178 |
+
"grad_norm": 0.06201878562569618,
|
179 |
"learning_rate": 3.0769230769230774e-05,
|
180 |
+
"loss": 0.0116,
|
181 |
"step": 220
|
182 |
},
|
183 |
{
|
184 |
"epoch": 3.5384615384615383,
|
185 |
+
"grad_norm": 0.06420481950044632,
|
186 |
"learning_rate": 2.307692307692308e-05,
|
187 |
+
"loss": 0.0325,
|
188 |
"step": 230
|
189 |
},
|
190 |
{
|
191 |
"epoch": 3.6923076923076925,
|
192 |
+
"grad_norm": 0.060832858085632324,
|
193 |
"learning_rate": 1.5384615384615387e-05,
|
194 |
+
"loss": 0.0114,
|
195 |
"step": 240
|
196 |
},
|
197 |
{
|
198 |
"epoch": 3.8461538461538463,
|
199 |
+
"grad_norm": 0.061510924249887466,
|
200 |
"learning_rate": 7.692307692307694e-06,
|
201 |
+
"loss": 0.0115,
|
202 |
"step": 250
|
203 |
},
|
204 |
{
|
205 |
"epoch": 4.0,
|
206 |
+
"grad_norm": 0.05591738596558571,
|
207 |
"learning_rate": 0.0,
|
208 |
+
"loss": 0.0114,
|
209 |
"step": 260
|
210 |
},
|
211 |
{
|
212 |
"epoch": 4.0,
|
213 |
"step": 260,
|
214 |
"total_flos": 3.205097416476426e+17,
|
215 |
+
"train_loss": 0.11695209254439061,
|
216 |
+
"train_runtime": 89.4129,
|
217 |
+
"train_samples_per_second": 46.257,
|
218 |
+
"train_steps_per_second": 2.908
|
219 |
}
|
220 |
],
|
221 |
"logging_steps": 10,
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5112
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:87354a911506538377ec10b629caf2f0beaa9bfa97389a2982938b09900d7a15
|
3 |
size 5112
|