pkr7098 commited on
Commit
a00438b
·
verified ·
1 Parent(s): 9f62ad1

End of training

Browse files
Files changed (5) hide show
  1. README.md +3 -1
  2. all_results.json +9 -9
  3. eval_results.json +5 -5
  4. train_results.json +4 -4
  5. trainer_state.json +93 -93
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
@@ -16,7 +18,7 @@ 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 an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.9381
22
  - Accuracy: 0.5414
 
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: 0.9381
24
  - Accuracy: 0.5414
all_results.json CHANGED
@@ -1,13 +1,13 @@
1
  {
2
  "epoch": 10.0,
3
- "eval_accuracy": 0.6541353383458647,
4
- "eval_loss": 0.751246988773346,
5
- "eval_runtime": 0.982,
6
- "eval_samples_per_second": 135.444,
7
- "eval_steps_per_second": 17.312,
8
  "total_flos": 8.772706474360013e+17,
9
- "train_loss": 0.9225140674297626,
10
- "train_runtime": 175.6098,
11
- "train_samples_per_second": 58.881,
12
- "train_steps_per_second": 7.403
13
  }
 
1
  {
2
  "epoch": 10.0,
3
+ "eval_accuracy": 0.5413533834586466,
4
+ "eval_loss": 0.9381032586097717,
5
+ "eval_runtime": 0.9636,
6
+ "eval_samples_per_second": 138.022,
7
+ "eval_steps_per_second": 17.642,
8
  "total_flos": 8.772706474360013e+17,
9
+ "train_loss": 0.9952024547870343,
10
+ "train_runtime": 174.8367,
11
+ "train_samples_per_second": 59.141,
12
+ "train_steps_per_second": 7.436
13
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 10.0,
3
- "eval_accuracy": 0.6541353383458647,
4
- "eval_loss": 0.751246988773346,
5
- "eval_runtime": 0.982,
6
- "eval_samples_per_second": 135.444,
7
- "eval_steps_per_second": 17.312
8
  }
 
1
  {
2
  "epoch": 10.0,
3
+ "eval_accuracy": 0.5413533834586466,
4
+ "eval_loss": 0.9381032586097717,
5
+ "eval_runtime": 0.9636,
6
+ "eval_samples_per_second": 138.022,
7
+ "eval_steps_per_second": 17.642
8
  }
train_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 10.0,
3
  "total_flos": 8.772706474360013e+17,
4
- "train_loss": 0.9225140674297626,
5
- "train_runtime": 175.6098,
6
- "train_samples_per_second": 58.881,
7
- "train_steps_per_second": 7.403
8
  }
 
1
  {
2
  "epoch": 10.0,
3
  "total_flos": 8.772706474360013e+17,
4
+ "train_loss": 0.9952024547870343,
5
+ "train_runtime": 174.8367,
6
+ "train_samples_per_second": 59.141,
7
+ "train_steps_per_second": 7.436
8
  }
trainer_state.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "best_metric": 0.751246988773346,
3
  "best_model_checkpoint": "./beans_outputs/checkpoint-1300",
4
  "epoch": 10.0,
5
  "eval_steps": 500,
@@ -10,193 +10,193 @@
10
  "log_history": [
11
  {
12
  "epoch": 0.7692307692307693,
13
- "grad_norm": 2.51617431640625,
14
- "learning_rate": 5e-05,
15
  "loss": 1.1019,
16
  "step": 100
17
  },
18
  {
19
  "epoch": 1.0,
20
- "eval_accuracy": 0.41353383458646614,
21
- "eval_loss": 1.094665288925171,
22
- "eval_runtime": 0.9302,
23
- "eval_samples_per_second": 142.976,
24
- "eval_steps_per_second": 18.275,
25
  "step": 130
26
  },
27
  {
28
  "epoch": 1.5384615384615383,
29
- "grad_norm": 0.6315650343894958,
30
- "learning_rate": 5e-05,
31
- "loss": 1.091,
32
  "step": 200
33
  },
34
  {
35
  "epoch": 2.0,
36
- "eval_accuracy": 0.518796992481203,
37
- "eval_loss": 1.002743124961853,
38
- "eval_runtime": 0.9515,
39
- "eval_samples_per_second": 139.773,
40
- "eval_steps_per_second": 17.866,
41
  "step": 260
42
  },
43
  {
44
  "epoch": 2.3076923076923075,
45
- "grad_norm": 1.4020596742630005,
46
- "learning_rate": 5e-05,
47
- "loss": 1.0433,
48
  "step": 300
49
  },
50
  {
51
  "epoch": 3.0,
52
- "eval_accuracy": 0.518796992481203,
53
- "eval_loss": 0.9840062856674194,
54
- "eval_runtime": 0.9456,
55
- "eval_samples_per_second": 140.656,
56
- "eval_steps_per_second": 17.979,
57
  "step": 390
58
  },
59
  {
60
  "epoch": 3.076923076923077,
61
- "grad_norm": 1.387670636177063,
62
- "learning_rate": 5e-05,
63
- "loss": 0.9915,
64
  "step": 400
65
  },
66
  {
67
  "epoch": 3.8461538461538463,
68
- "grad_norm": 0.7062584161758423,
69
- "learning_rate": 5e-05,
70
- "loss": 0.9823,
71
  "step": 500
72
  },
73
  {
74
  "epoch": 4.0,
75
- "eval_accuracy": 0.5639097744360902,
76
- "eval_loss": 0.9470232129096985,
77
- "eval_runtime": 0.9624,
78
- "eval_samples_per_second": 138.194,
79
- "eval_steps_per_second": 17.664,
80
  "step": 520
81
  },
82
  {
83
  "epoch": 4.615384615384615,
84
- "grad_norm": 0.5388019680976868,
85
- "learning_rate": 5e-05,
86
- "loss": 0.9523,
87
  "step": 600
88
  },
89
  {
90
  "epoch": 5.0,
91
- "eval_accuracy": 0.5789473684210527,
92
- "eval_loss": 0.924637496471405,
93
- "eval_runtime": 0.9596,
94
- "eval_samples_per_second": 138.602,
95
- "eval_steps_per_second": 17.716,
96
  "step": 650
97
  },
98
  {
99
  "epoch": 5.384615384615385,
100
- "grad_norm": 1.5363554954528809,
101
- "learning_rate": 5e-05,
102
- "loss": 0.9431,
103
  "step": 700
104
  },
105
  {
106
  "epoch": 6.0,
107
- "eval_accuracy": 0.556390977443609,
108
- "eval_loss": 0.890453577041626,
109
- "eval_runtime": 0.9858,
110
- "eval_samples_per_second": 134.909,
111
- "eval_steps_per_second": 17.244,
112
  "step": 780
113
  },
114
  {
115
  "epoch": 6.153846153846154,
116
- "grad_norm": 1.7380107641220093,
117
- "learning_rate": 5e-05,
118
- "loss": 0.8877,
119
  "step": 800
120
  },
121
  {
122
  "epoch": 6.923076923076923,
123
- "grad_norm": 2.0791220664978027,
124
- "learning_rate": 5e-05,
125
- "loss": 0.8587,
126
  "step": 900
127
  },
128
  {
129
  "epoch": 7.0,
130
- "eval_accuracy": 0.5939849624060151,
131
- "eval_loss": 0.8433966636657715,
132
- "eval_runtime": 0.9653,
133
- "eval_samples_per_second": 137.788,
134
- "eval_steps_per_second": 17.612,
135
  "step": 910
136
  },
137
  {
138
  "epoch": 7.6923076923076925,
139
- "grad_norm": 1.4432713985443115,
140
- "learning_rate": 5e-05,
141
- "loss": 0.834,
142
  "step": 1000
143
  },
144
  {
145
  "epoch": 8.0,
146
- "eval_accuracy": 0.5714285714285714,
147
- "eval_loss": 0.8318859934806824,
148
- "eval_runtime": 0.9617,
149
- "eval_samples_per_second": 138.302,
150
- "eval_steps_per_second": 17.678,
151
  "step": 1040
152
  },
153
  {
154
  "epoch": 8.461538461538462,
155
- "grad_norm": 2.059706449508667,
156
- "learning_rate": 5e-05,
157
- "loss": 0.8149,
158
  "step": 1100
159
  },
160
  {
161
  "epoch": 9.0,
162
- "eval_accuracy": 0.6766917293233082,
163
- "eval_loss": 0.7831493616104126,
164
- "eval_runtime": 0.9673,
165
- "eval_samples_per_second": 137.496,
166
- "eval_steps_per_second": 17.575,
167
  "step": 1170
168
  },
169
  {
170
  "epoch": 9.23076923076923,
171
- "grad_norm": 1.22969651222229,
172
- "learning_rate": 5e-05,
173
- "loss": 0.7628,
174
  "step": 1200
175
  },
176
  {
177
  "epoch": 10.0,
178
- "grad_norm": 3.152574300765991,
179
- "learning_rate": 5e-05,
180
- "loss": 0.729,
181
  "step": 1300
182
  },
183
  {
184
  "epoch": 10.0,
185
- "eval_accuracy": 0.6541353383458647,
186
- "eval_loss": 0.751246988773346,
187
- "eval_runtime": 0.9753,
188
- "eval_samples_per_second": 136.369,
189
- "eval_steps_per_second": 17.431,
190
  "step": 1300
191
  },
192
  {
193
  "epoch": 10.0,
194
  "step": 1300,
195
  "total_flos": 8.772706474360013e+17,
196
- "train_loss": 0.9225140674297626,
197
- "train_runtime": 175.6098,
198
- "train_samples_per_second": 58.881,
199
- "train_steps_per_second": 7.403
200
  }
201
  ],
202
  "logging_steps": 100,
 
1
  {
2
+ "best_metric": 0.9381032586097717,
3
  "best_model_checkpoint": "./beans_outputs/checkpoint-1300",
4
  "epoch": 10.0,
5
  "eval_steps": 500,
 
10
  "log_history": [
11
  {
12
  "epoch": 0.7692307692307693,
13
+ "grad_norm": 2.513921022415161,
14
+ "learning_rate": 4.615384615384616e-05,
15
  "loss": 1.1019,
16
  "step": 100
17
  },
18
  {
19
  "epoch": 1.0,
20
+ "eval_accuracy": 0.3007518796992481,
21
+ "eval_loss": 1.0955818891525269,
22
+ "eval_runtime": 0.9308,
23
+ "eval_samples_per_second": 142.882,
24
+ "eval_steps_per_second": 18.263,
25
  "step": 130
26
  },
27
  {
28
  "epoch": 1.5384615384615383,
29
+ "grad_norm": 0.6020777225494385,
30
+ "learning_rate": 4.230769230769231e-05,
31
+ "loss": 1.0934,
32
  "step": 200
33
  },
34
  {
35
  "epoch": 2.0,
36
+ "eval_accuracy": 0.5413533834586466,
37
+ "eval_loss": 1.0218424797058105,
38
+ "eval_runtime": 0.9394,
39
+ "eval_samples_per_second": 141.582,
40
+ "eval_steps_per_second": 18.097,
41
  "step": 260
42
  },
43
  {
44
  "epoch": 2.3076923076923075,
45
+ "grad_norm": 1.3475981950759888,
46
+ "learning_rate": 3.846153846153846e-05,
47
+ "loss": 1.0541,
48
  "step": 300
49
  },
50
  {
51
  "epoch": 3.0,
52
+ "eval_accuracy": 0.5037593984962406,
53
+ "eval_loss": 0.9863384366035461,
54
+ "eval_runtime": 0.961,
55
+ "eval_samples_per_second": 138.393,
56
+ "eval_steps_per_second": 17.689,
57
  "step": 390
58
  },
59
  {
60
  "epoch": 3.076923076923077,
61
+ "grad_norm": 1.3711971044540405,
62
+ "learning_rate": 3.461538461538462e-05,
63
+ "loss": 0.9991,
64
  "step": 400
65
  },
66
  {
67
  "epoch": 3.8461538461538463,
68
+ "grad_norm": 0.6074744462966919,
69
+ "learning_rate": 3.0769230769230774e-05,
70
+ "loss": 0.9911,
71
  "step": 500
72
  },
73
  {
74
  "epoch": 4.0,
75
+ "eval_accuracy": 0.5338345864661654,
76
+ "eval_loss": 0.9627529382705688,
77
+ "eval_runtime": 0.9661,
78
+ "eval_samples_per_second": 137.661,
79
+ "eval_steps_per_second": 17.596,
80
  "step": 520
81
  },
82
  {
83
  "epoch": 4.615384615384615,
84
+ "grad_norm": 0.4636833667755127,
85
+ "learning_rate": 2.6923076923076923e-05,
86
+ "loss": 0.9776,
87
  "step": 600
88
  },
89
  {
90
  "epoch": 5.0,
91
+ "eval_accuracy": 0.5413533834586466,
92
+ "eval_loss": 0.9637588858604431,
93
+ "eval_runtime": 0.96,
94
+ "eval_samples_per_second": 138.544,
95
+ "eval_steps_per_second": 17.709,
96
  "step": 650
97
  },
98
  {
99
  "epoch": 5.384615384615385,
100
+ "grad_norm": 1.6218713521957397,
101
+ "learning_rate": 2.307692307692308e-05,
102
+ "loss": 0.9852,
103
  "step": 700
104
  },
105
  {
106
  "epoch": 6.0,
107
+ "eval_accuracy": 0.5413533834586466,
108
+ "eval_loss": 0.9598925113677979,
109
+ "eval_runtime": 0.956,
110
+ "eval_samples_per_second": 139.127,
111
+ "eval_steps_per_second": 17.783,
112
  "step": 780
113
  },
114
  {
115
  "epoch": 6.153846153846154,
116
+ "grad_norm": 1.7121293544769287,
117
+ "learning_rate": 1.923076923076923e-05,
118
+ "loss": 0.9753,
119
  "step": 800
120
  },
121
  {
122
  "epoch": 6.923076923076923,
123
+ "grad_norm": 1.5765596628189087,
124
+ "learning_rate": 1.5384615384615387e-05,
125
+ "loss": 0.9666,
126
  "step": 900
127
  },
128
  {
129
  "epoch": 7.0,
130
+ "eval_accuracy": 0.5639097744360902,
131
+ "eval_loss": 0.9482960104942322,
132
+ "eval_runtime": 0.9859,
133
+ "eval_samples_per_second": 134.905,
134
+ "eval_steps_per_second": 17.244,
135
  "step": 910
136
  },
137
  {
138
  "epoch": 7.6923076923076925,
139
+ "grad_norm": 1.4977715015411377,
140
+ "learning_rate": 1.153846153846154e-05,
141
+ "loss": 0.9528,
142
  "step": 1000
143
  },
144
  {
145
  "epoch": 8.0,
146
+ "eval_accuracy": 0.5413533834586466,
147
+ "eval_loss": 0.9446954727172852,
148
+ "eval_runtime": 0.9431,
149
+ "eval_samples_per_second": 141.029,
150
+ "eval_steps_per_second": 18.026,
151
  "step": 1040
152
  },
153
  {
154
  "epoch": 8.461538461538462,
155
+ "grad_norm": 1.6241708993911743,
156
+ "learning_rate": 7.692307692307694e-06,
157
+ "loss": 0.9583,
158
  "step": 1100
159
  },
160
  {
161
  "epoch": 9.0,
162
+ "eval_accuracy": 0.5338345864661654,
163
+ "eval_loss": 0.9394896030426025,
164
+ "eval_runtime": 0.9745,
165
+ "eval_samples_per_second": 136.479,
166
+ "eval_steps_per_second": 17.445,
167
  "step": 1170
168
  },
169
  {
170
  "epoch": 9.23076923076923,
171
+ "grad_norm": 0.5392513871192932,
172
+ "learning_rate": 3.846153846153847e-06,
173
+ "loss": 0.9421,
174
  "step": 1200
175
  },
176
  {
177
  "epoch": 10.0,
178
+ "grad_norm": 4.028888702392578,
179
+ "learning_rate": 0.0,
180
+ "loss": 0.9401,
181
  "step": 1300
182
  },
183
  {
184
  "epoch": 10.0,
185
+ "eval_accuracy": 0.5413533834586466,
186
+ "eval_loss": 0.9381032586097717,
187
+ "eval_runtime": 0.9652,
188
+ "eval_samples_per_second": 137.798,
189
+ "eval_steps_per_second": 17.613,
190
  "step": 1300
191
  },
192
  {
193
  "epoch": 10.0,
194
  "step": 1300,
195
  "total_flos": 8.772706474360013e+17,
196
+ "train_loss": 0.9952024547870343,
197
+ "train_runtime": 174.8367,
198
+ "train_samples_per_second": 59.141,
199
+ "train_steps_per_second": 7.436
200
  }
201
  ],
202
  "logging_steps": 100,