moooji commited on
Commit
1643c96
1 Parent(s): 8207dad

End of training

Browse files
Files changed (5) hide show
  1. README.md +7 -5
  2. all_results.json +12 -0
  3. eval_results.json +8 -0
  4. train_results.json +7 -0
  5. trainer_state.json +169 -0
README.md CHANGED
@@ -2,6 +2,8 @@
2
  license: apache-2.0
3
  base_model: google/vit-large-patch16-224-in21k
4
  tags:
 
 
5
  - generated_from_trainer
6
  datasets:
7
  - imagefolder
@@ -14,7 +16,7 @@ model-index:
14
  name: Image Classification
15
  type: image-classification
16
  dataset:
17
- name: imagefolder
18
  type: imagefolder
19
  config: default
20
  split: train
@@ -22,7 +24,7 @@ model-index:
22
  metrics:
23
  - name: Accuracy
24
  type: accuracy
25
- value: 0.9915469146238377
26
  ---
27
 
28
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,10 +32,10 @@ should probably proofread and complete it, then remove this comment. -->
30
 
31
  # fashion-images-pack-types-vit-large-patch16-224-in21k
32
 
33
- This model is a fine-tuned version of [google/vit-large-patch16-224-in21k](https://huggingface.co/google/vit-large-patch16-224-in21k) on the imagefolder dataset.
34
  It achieves the following results on the evaluation set:
35
- - Loss: 0.0362
36
- - Accuracy: 0.9915
37
 
38
  ## Model description
39
 
 
2
  license: apache-2.0
3
  base_model: google/vit-large-patch16-224-in21k
4
  tags:
5
+ - image-classification
6
+ - vision
7
  - generated_from_trainer
8
  datasets:
9
  - imagefolder
 
16
  name: Image Classification
17
  type: image-classification
18
  dataset:
19
+ name: touchtech/fashion-images-pack-types
20
  type: imagefolder
21
  config: default
22
  split: train
 
24
  metrics:
25
  - name: Accuracy
26
  type: accuracy
27
+ value: 0.9894336432797971
28
  ---
29
 
30
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
32
 
33
  # fashion-images-pack-types-vit-large-patch16-224-in21k
34
 
35
+ This model is a fine-tuned version of [google/vit-large-patch16-224-in21k](https://huggingface.co/google/vit-large-patch16-224-in21k) on the touchtech/fashion-images-pack-types dataset.
36
  It achieves the following results on the evaluation set:
37
+ - Loss: 0.0343
38
+ - Accuracy: 0.9894
39
 
40
  ## Model description
41
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.9894336432797971,
4
+ "eval_loss": 0.03428684547543526,
5
+ "eval_runtime": 162.823,
6
+ "eval_samples_per_second": 14.531,
7
+ "eval_steps_per_second": 1.818,
8
+ "train_loss": 0.10461939445258894,
9
+ "train_runtime": 6074.546,
10
+ "train_samples_per_second": 11.034,
11
+ "train_steps_per_second": 1.38
12
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.9894336432797971,
4
+ "eval_loss": 0.03428684547543526,
5
+ "eval_runtime": 162.823,
6
+ "eval_samples_per_second": 14.531,
7
+ "eval_steps_per_second": 1.818
8
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "train_loss": 0.10461939445258894,
4
+ "train_runtime": 6074.546,
5
+ "train_samples_per_second": 11.034,
6
+ "train_steps_per_second": 1.38
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,169 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.03428684547543526,
3
+ "best_model_checkpoint": "/workspace/training_output/pack-types-vit-large-patch16-224-in21k/checkpoint-3352",
4
+ "epoch": 5.0,
5
+ "eval_steps": 500,
6
+ "global_step": 8380,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.3,
13
+ "learning_rate": 1.8806682577565632e-05,
14
+ "loss": 0.2863,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.6,
19
+ "learning_rate": 1.7613365155131266e-05,
20
+ "loss": 0.1669,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.89,
25
+ "learning_rate": 1.6420047732696897e-05,
26
+ "loss": 0.1556,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 1.0,
31
+ "eval_accuracy": 0.9860524091293322,
32
+ "eval_loss": 0.04897575080394745,
33
+ "eval_runtime": 170.5458,
34
+ "eval_samples_per_second": 13.873,
35
+ "eval_steps_per_second": 1.736,
36
+ "step": 1676
37
+ },
38
+ {
39
+ "epoch": 1.19,
40
+ "learning_rate": 1.5226730310262532e-05,
41
+ "loss": 0.1183,
42
+ "step": 2000
43
+ },
44
+ {
45
+ "epoch": 1.49,
46
+ "learning_rate": 1.4033412887828164e-05,
47
+ "loss": 0.1141,
48
+ "step": 2500
49
+ },
50
+ {
51
+ "epoch": 1.79,
52
+ "learning_rate": 1.2840095465393797e-05,
53
+ "loss": 0.1185,
54
+ "step": 3000
55
+ },
56
+ {
57
+ "epoch": 2.0,
58
+ "eval_accuracy": 0.9894336432797971,
59
+ "eval_loss": 0.03428684547543526,
60
+ "eval_runtime": 161.6386,
61
+ "eval_samples_per_second": 14.638,
62
+ "eval_steps_per_second": 1.831,
63
+ "step": 3352
64
+ },
65
+ {
66
+ "epoch": 2.09,
67
+ "learning_rate": 1.1646778042959427e-05,
68
+ "loss": 0.1127,
69
+ "step": 3500
70
+ },
71
+ {
72
+ "epoch": 2.39,
73
+ "learning_rate": 1.045346062052506e-05,
74
+ "loss": 0.0862,
75
+ "step": 4000
76
+ },
77
+ {
78
+ "epoch": 2.68,
79
+ "learning_rate": 9.260143198090693e-06,
80
+ "loss": 0.1032,
81
+ "step": 4500
82
+ },
83
+ {
84
+ "epoch": 2.98,
85
+ "learning_rate": 8.066825775656326e-06,
86
+ "loss": 0.0815,
87
+ "step": 5000
88
+ },
89
+ {
90
+ "epoch": 3.0,
91
+ "eval_accuracy": 0.9881656804733728,
92
+ "eval_loss": 0.053725942969322205,
93
+ "eval_runtime": 159.6941,
94
+ "eval_samples_per_second": 14.816,
95
+ "eval_steps_per_second": 1.854,
96
+ "step": 5028
97
+ },
98
+ {
99
+ "epoch": 3.28,
100
+ "learning_rate": 6.873508353221957e-06,
101
+ "loss": 0.0744,
102
+ "step": 5500
103
+ },
104
+ {
105
+ "epoch": 3.58,
106
+ "learning_rate": 5.68019093078759e-06,
107
+ "loss": 0.0793,
108
+ "step": 6000
109
+ },
110
+ {
111
+ "epoch": 3.88,
112
+ "learning_rate": 4.486873508353222e-06,
113
+ "loss": 0.0503,
114
+ "step": 6500
115
+ },
116
+ {
117
+ "epoch": 4.0,
118
+ "eval_accuracy": 0.9915469146238377,
119
+ "eval_loss": 0.03739229962229729,
120
+ "eval_runtime": 160.0301,
121
+ "eval_samples_per_second": 14.785,
122
+ "eval_steps_per_second": 1.85,
123
+ "step": 6704
124
+ },
125
+ {
126
+ "epoch": 4.18,
127
+ "learning_rate": 3.293556085918855e-06,
128
+ "loss": 0.0727,
129
+ "step": 7000
130
+ },
131
+ {
132
+ "epoch": 4.47,
133
+ "learning_rate": 2.100238663484487e-06,
134
+ "loss": 0.0466,
135
+ "step": 7500
136
+ },
137
+ {
138
+ "epoch": 4.77,
139
+ "learning_rate": 9.069212410501194e-07,
140
+ "loss": 0.0447,
141
+ "step": 8000
142
+ },
143
+ {
144
+ "epoch": 5.0,
145
+ "eval_accuracy": 0.9915469146238377,
146
+ "eval_loss": 0.03615482896566391,
147
+ "eval_runtime": 162.6104,
148
+ "eval_samples_per_second": 14.55,
149
+ "eval_steps_per_second": 1.82,
150
+ "step": 8380
151
+ },
152
+ {
153
+ "epoch": 5.0,
154
+ "step": 8380,
155
+ "total_flos": 1.8360872742865766e+19,
156
+ "train_loss": 0.10461939445258894,
157
+ "train_runtime": 6074.546,
158
+ "train_samples_per_second": 11.034,
159
+ "train_steps_per_second": 1.38
160
+ }
161
+ ],
162
+ "logging_steps": 500,
163
+ "max_steps": 8380,
164
+ "num_train_epochs": 5,
165
+ "save_steps": 500,
166
+ "total_flos": 1.8360872742865766e+19,
167
+ "trial_name": null,
168
+ "trial_params": null
169
+ }