Hugging Face's logo

Spaces:
kunwarsaaim
/
Garbage_Classification_ViT
Sleeping

App Files Community
kunwarsaaim commited on
Commit
05d6a45
·
1 Parent(s): a610f96

intial commit

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. app.py +38 -0
  2. cardboard373.jpg +0 -0
  3. glass456.jpg +0 -0
  4. metal382.jpg +0 -0
  5. models/.DS_Store +0 -0
  6. models/apple/.DS_Store +0 -0
  7. models/apple/mobilevit-small-garbage/all_results.json +16 -0
  8. models/apple/mobilevit-small-garbage/config.json +63 -0
  9. models/apple/mobilevit-small-garbage/preprocessor_config.json +10 -0
  10. models/apple/mobilevit-small-garbage/pytorch_model.bin +3 -0
  11. models/apple/mobilevit-small-garbage/runs/Dec01_01-56-59_48dca185d887/1669859844.908253/events.out.tfevents.1669859844.48dca185d887.5019.7 +3 -0
  12. models/apple/mobilevit-small-garbage/runs/Dec01_01-56-59_48dca185d887/events.out.tfevents.1669859844.48dca185d887.5019.6 +3 -0
  13. models/apple/mobilevit-small-garbage/runs/Dec01_01-58-12_48dca185d887/1669859892.1544209/events.out.tfevents.1669859892.48dca185d887.5019.9 +3 -0
  14. models/apple/mobilevit-small-garbage/runs/Dec01_01-58-12_48dca185d887/events.out.tfevents.1669859892.48dca185d887.5019.8 +3 -0
  15. models/apple/mobilevit-small-garbage/runs/Dec01_02-03-03_48dca185d887/1669860183.7440953/events.out.tfevents.1669860183.48dca185d887.5019.11 +3 -0
  16. models/apple/mobilevit-small-garbage/runs/Dec01_02-03-03_48dca185d887/events.out.tfevents.1669860183.48dca185d887.5019.10 +3 -0
  17. models/apple/mobilevit-small-garbage/runs/Dec12_21-56-05_30c96569bd60/events.out.tfevents.1670882184.30c96569bd60.71.2 +3 -0
  18. models/apple/mobilevit-small-garbage/test_results.json +10 -0
  19. models/apple/mobilevit-small-garbage/train_results.json +8 -0
  20. models/apple/mobilevit-small-garbage/trainer_state.json +400 -0
  21. models/apple/mobilevit-small-garbage/training_args.bin +3 -0
  22. models/vit-base-garbage/.DS_Store +0 -0
  23. models/vit-base-garbage/all_results.json +16 -0
  24. models/vit-base-garbage/checkpoint-300/config.json +40 -0
  25. models/vit-base-garbage/checkpoint-300/optimizer.pt +3 -0
  26. models/vit-base-garbage/checkpoint-300/preprocessor_config.json +17 -0
  27. models/vit-base-garbage/checkpoint-300/pytorch_model.bin +3 -0
  28. models/vit-base-garbage/checkpoint-300/rng_state.pth +3 -0
  29. models/vit-base-garbage/checkpoint-300/scaler.pt +3 -0
  30. models/vit-base-garbage/checkpoint-300/scheduler.pt +3 -0
  31. models/vit-base-garbage/checkpoint-300/trainer_state.json +223 -0
  32. models/vit-base-garbage/checkpoint-300/training_args.bin +3 -0
  33. models/vit-base-garbage/checkpoint-400/config.json +40 -0
  34. models/vit-base-garbage/checkpoint-400/optimizer.pt +3 -0
  35. models/vit-base-garbage/checkpoint-400/preprocessor_config.json +17 -0
  36. models/vit-base-garbage/checkpoint-400/pytorch_model.bin +3 -0
  37. models/vit-base-garbage/checkpoint-400/rng_state.pth +3 -0
  38. models/vit-base-garbage/checkpoint-400/scaler.pt +3 -0
  39. models/vit-base-garbage/checkpoint-400/scheduler.pt +3 -0
  40. models/vit-base-garbage/checkpoint-400/trainer_state.json +292 -0
  41. models/vit-base-garbage/checkpoint-400/training_args.bin +3 -0
  42. models/vit-base-garbage/config.json +40 -0
  43. models/vit-base-garbage/preprocessor_config.json +17 -0
  44. models/vit-base-garbage/pytorch_model.bin +3 -0
  45. models/vit-base-garbage/runs/Dec12_19-10-35_ed23ba8f6546/1670872238.0395455/events.out.tfevents.1670872238.ed23ba8f6546.76.2 +3 -0
  46. models/vit-base-garbage/runs/Dec12_19-10-35_ed23ba8f6546/events.out.tfevents.1670872238.ed23ba8f6546.76.1 +3 -0
  47. models/vit-base-garbage/runs/Dec12_19-10-35_ed23ba8f6546/events.out.tfevents.1670872287.ed23ba8f6546.76.3 +3 -0
  48. models/vit-base-garbage/runs/Dec12_21-11-38_30c96569bd60/events.out.tfevents.1670879785.30c96569bd60.71.0 +3 -0
  49. models/vit-base-garbage/runs/Dec12_21-51-08_30c96569bd60/events.out.tfevents.1670881883.30c96569bd60.71.1 +3 -0
  50. models/vit-base-garbage/runs/Nov28_22-29-34_09524550a2a1/1669674600.7132652/events.out.tfevents.1669674600.09524550a2a1.78.1 +3 -0
app.py ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import gradio as gr
2
+ from torch.nn.functional import softmax
3
+ import torch
4
+ from transformers import ViTFeatureExtractor
5
+ from transformers import MobileViTFeatureExtractor
6
+ from transformers import MobileViTForImageClassification
7
+ from transformers import ViTForImageClassification
8
+
9
+
10
+ def predict(model_type, inp):
11
+
12
+ if model_type == "ViT":
13
+ model_name_or_path = './models/vit-base-garbage/'
14
+ feature_extractor = ViTFeatureExtractor.from_pretrained(model_name_or_path)
15
+ model = ViTForImageClassification.from_pretrained(model_name_or_path)
16
+ elif model_type == "MobileViT":
17
+ model_name_or_path = './models/apple/mobilevit-small-garbage/'
18
+ feature_extractor = MobileViTFeatureExtractor.from_pretrained(model_name_or_path)
19
+ model = MobileViTForImageClassification.from_pretrained(model_name_or_path)
20
+ inputs = feature_extractor(inp, return_tensors="pt")
21
+ LABELS = list(model.config.label2id.keys())
22
+ with torch.no_grad():
23
+ logits = model(**inputs)
24
+ print(logits[0])
25
+ probability = torch.nn.functional.softmax(logits[0], dim=-1)
26
+
27
+ confidences = {LABELS[i]:(float(probability[0][i])) for i in range(6)}
28
+ # print(confidences)
29
+ return confidences
30
+
31
+
32
+ demo = gr.Interface(fn=predict,
33
+ inputs=[gr.Dropdown(["ViT", "MobileViT"], label="Model Name", value='ViT'),gr.inputs.Image(type="pil")],
34
+ outputs=gr.outputs.Label(num_top_classes=3),
35
+ examples=[["ViT","paper567.jpg"],["ViT","trash105.jpg"],["ViT","plastic202.jpg"],["MobileViT","metal382.jpg"]],
36
+ )
37
+
38
+ demo.launch()
cardboard373.jpg ADDED
glass456.jpg ADDED
metal382.jpg ADDED
models/.DS_Store ADDED
Binary file (6.15 kB). View file
 
models/apple/.DS_Store ADDED
Binary file (6.15 kB). View file
 
models/apple/mobilevit-small-garbage/all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_accuracy": 0.9396751740139211,
4
+ "eval_f1": 0.9320563646123423,
5
+ "eval_loss": 0.17442849278450012,
6
+ "eval_precision": 0.9422527093650638,
7
+ "eval_recall": 0.9244185656020448,
8
+ "eval_runtime": 7.4671,
9
+ "eval_samples_per_second": 57.72,
10
+ "eval_steps_per_second": 7.232,
11
+ "total_flos": 5.153016901533696e+16,
12
+ "train_loss": 0.5183452816696854,
13
+ "train_runtime": 159.3173,
14
+ "train_samples_per_second": 55.487,
15
+ "train_steps_per_second": 3.484
16
+ }
models/apple/mobilevit-small-garbage/config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "apple/mobilevit-small",
3
+ "architectures": [
4
+ "MobileViTForImageClassification"
5
+ ],
6
+ "aspp_dropout_prob": 0.1,
7
+ "aspp_out_channels": 256,
8
+ "atrous_rates": [
9
+ 6,
10
+ 12,
11
+ 18
12
+ ],
13
+ "attention_probs_dropout_prob": 0.0,
14
+ "classifier_dropout_prob": 0.1,
15
+ "conv_kernel_size": 3,
16
+ "expand_ratio": 4.0,
17
+ "hidden_act": "silu",
18
+ "hidden_dropout_prob": 0.1,
19
+ "hidden_sizes": [
20
+ 144,
21
+ 192,
22
+ 240
23
+ ],
24
+ "id2label": {
25
+ "0": "cardboard",
26
+ "1": "glass",
27
+ "2": "metal",
28
+ "3": "paper",
29
+ "4": "plastic",
30
+ "5": "trash"
31
+ },
32
+ "image_size": 256,
33
+ "initializer_range": 0.02,
34
+ "label2id": {
35
+ "cardboard": "0",
36
+ "glass": "1",
37
+ "metal": "2",
38
+ "paper": "3",
39
+ "plastic": "4",
40
+ "trash": "5"
41
+ },
42
+ "layer_norm_eps": 1e-05,
43
+ "mlp_ratio": 2.0,
44
+ "model_type": "mobilevit",
45
+ "neck_hidden_sizes": [
46
+ 16,
47
+ 32,
48
+ 64,
49
+ 96,
50
+ 128,
51
+ 160,
52
+ 640
53
+ ],
54
+ "num_attention_heads": 4,
55
+ "num_channels": 3,
56
+ "output_stride": 32,
57
+ "patch_size": 2,
58
+ "problem_type": "single_label_classification",
59
+ "qkv_bias": true,
60
+ "semantic_loss_ignore_index": 255,
61
+ "torch_dtype": "float32",
62
+ "transformers_version": "4.24.0"
63
+ }
models/apple/mobilevit-small-garbage/preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": 256,
3
+ "do_center_crop": true,
4
+ "do_flip_channel_order": true,
5
+ "do_flip_channels": true,
6
+ "do_resize": true,
7
+ "feature_extractor_type": "MobileViTFeatureExtractor",
8
+ "resample": 2,
9
+ "size": 288
10
+ }
models/apple/mobilevit-small-garbage/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3de3d2c2a6114c79a2a1814b21ee1b602f130851a3f70611e55b52d652d5dd31
3
+ size 19937419
models/apple/mobilevit-small-garbage/runs/Dec01_01-56-59_48dca185d887/1669859844.908253/events.out.tfevents.1669859844.48dca185d887.5019.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9029ef393ea8d118116c66afed5937f1749a702460878cb2cb78ad3c8232e568
3
+ size 5520
models/apple/mobilevit-small-garbage/runs/Dec01_01-56-59_48dca185d887/events.out.tfevents.1669859844.48dca185d887.5019.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1ef8ff7f11f7bdbec4874fbf7309af69251cfad89d4db2f9b936ef27ab03e82
3
+ size 5775
models/apple/mobilevit-small-garbage/runs/Dec01_01-58-12_48dca185d887/1669859892.1544209/events.out.tfevents.1669859892.48dca185d887.5019.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb6cd0a21c8eedb632c85fcff05dacc7bf3428542263c618958d97749d88e625
3
+ size 5520
models/apple/mobilevit-small-garbage/runs/Dec01_01-58-12_48dca185d887/events.out.tfevents.1669859892.48dca185d887.5019.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3226a08a0d3739a8da4b6ef5e37f81701bf60a5f4c4e7bbe2678e7b4bc887072
3
+ size 10697
models/apple/mobilevit-small-garbage/runs/Dec01_02-03-03_48dca185d887/1669860183.7440953/events.out.tfevents.1669860183.48dca185d887.5019.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b6dfd2faa33c11bbb5f9638c7ed47fcce4499450a2912573286a751c1fdd878
3
+ size 5520
models/apple/mobilevit-small-garbage/runs/Dec01_02-03-03_48dca185d887/events.out.tfevents.1669860183.48dca185d887.5019.10 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a69e9734c9a1d66df23d6d2326cf01248e11e73a5f757604b12fd305e896e3bf
3
+ size 14797
models/apple/mobilevit-small-garbage/runs/Dec12_21-56-05_30c96569bd60/events.out.tfevents.1670882184.30c96569bd60.71.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc7b9c1ba45264fb14f86d79d0ea4d1ef43bbf7b7de5386e19de02b293293250
3
+ size 438
models/apple/mobilevit-small-garbage/test_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "eval_accuracy": 0.9396751740139211,
3
+ "eval_f1": 0.9320563646123423,
4
+ "eval_loss": 0.17442849278450012,
5
+ "eval_precision": 0.9422527093650638,
6
+ "eval_recall": 0.9244185656020448,
7
+ "eval_runtime": 7.4671,
8
+ "eval_samples_per_second": 57.72,
9
+ "eval_steps_per_second": 7.232
10
+ }
models/apple/mobilevit-small-garbage/train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "total_flos": 5.153016901533696e+16,
4
+ "train_loss": 0.5183452816696854,
5
+ "train_runtime": 159.3173,
6
+ "train_samples_per_second": 55.487,
7
+ "train_steps_per_second": 3.484
8
+ }
models/apple/mobilevit-small-garbage/trainer_state.json ADDED
@@ -0,0 +1,400 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.26617246866226196,
3
+ "best_model_checkpoint": "./apple/mobilevit-small-garbage/checkpoint-500",
4
+ "epoch": 5.0,
5
+ "global_step": 555,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.09,
12
+ "learning_rate": 0.0001963963963963964,
13
+ "loss": 1.7495,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.18,
18
+ "learning_rate": 0.00019279279279279282,
19
+ "loss": 1.6591,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.27,
24
+ "learning_rate": 0.0001891891891891892,
25
+ "loss": 1.5418,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.36,
30
+ "learning_rate": 0.00018594594594594597,
31
+ "loss": 1.3934,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.45,
36
+ "learning_rate": 0.00018234234234234235,
37
+ "loss": 1.2802,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.54,
42
+ "learning_rate": 0.00017873873873873876,
43
+ "loss": 1.18,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.63,
48
+ "learning_rate": 0.00017513513513513516,
49
+ "loss": 0.9854,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.72,
54
+ "learning_rate": 0.00017153153153153154,
55
+ "loss": 0.9527,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.81,
60
+ "learning_rate": 0.00016792792792792794,
61
+ "loss": 0.8574,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.9,
66
+ "learning_rate": 0.00016432432432432435,
67
+ "loss": 0.8589,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.9,
72
+ "eval_accuracy": 0.8079268292682927,
73
+ "eval_loss": 0.661000669002533,
74
+ "eval_runtime": 4.4994,
75
+ "eval_samples_per_second": 72.898,
76
+ "eval_steps_per_second": 9.112,
77
+ "step": 100
78
+ },
79
+ {
80
+ "epoch": 0.99,
81
+ "learning_rate": 0.0001610810810810811,
82
+ "loss": 0.7726,
83
+ "step": 110
84
+ },
85
+ {
86
+ "epoch": 1.08,
87
+ "learning_rate": 0.00015747747747747747,
88
+ "loss": 0.74,
89
+ "step": 120
90
+ },
91
+ {
92
+ "epoch": 1.17,
93
+ "learning_rate": 0.00015387387387387388,
94
+ "loss": 0.6325,
95
+ "step": 130
96
+ },
97
+ {
98
+ "epoch": 1.26,
99
+ "learning_rate": 0.00015063063063063063,
100
+ "loss": 0.5792,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 1.35,
105
+ "learning_rate": 0.00014702702702702703,
106
+ "loss": 0.5582,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 1.44,
111
+ "learning_rate": 0.00014342342342342344,
112
+ "loss": 0.5644,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 1.53,
117
+ "learning_rate": 0.00013981981981981982,
118
+ "loss": 0.4776,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 1.62,
123
+ "learning_rate": 0.00013621621621621622,
124
+ "loss": 0.4537,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 1.71,
129
+ "learning_rate": 0.00013261261261261263,
130
+ "loss": 0.4813,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 1.8,
135
+ "learning_rate": 0.000129009009009009,
136
+ "loss": 0.4676,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 1.8,
141
+ "eval_accuracy": 0.8597560975609756,
142
+ "eval_loss": 0.4162694811820984,
143
+ "eval_runtime": 4.3992,
144
+ "eval_samples_per_second": 74.559,
145
+ "eval_steps_per_second": 9.32,
146
+ "step": 200
147
+ },
148
+ {
149
+ "epoch": 1.89,
150
+ "learning_rate": 0.0001254054054054054,
151
+ "loss": 0.5374,
152
+ "step": 210
153
+ },
154
+ {
155
+ "epoch": 1.98,
156
+ "learning_rate": 0.0001218018018018018,
157
+ "loss": 0.4006,
158
+ "step": 220
159
+ },
160
+ {
161
+ "epoch": 2.07,
162
+ "learning_rate": 0.00011855855855855858,
163
+ "loss": 0.4236,
164
+ "step": 230
165
+ },
166
+ {
167
+ "epoch": 2.16,
168
+ "learning_rate": 0.00011495495495495497,
169
+ "loss": 0.4494,
170
+ "step": 240
171
+ },
172
+ {
173
+ "epoch": 2.25,
174
+ "learning_rate": 0.00011135135135135135,
175
+ "loss": 0.3139,
176
+ "step": 250
177
+ },
178
+ {
179
+ "epoch": 2.34,
180
+ "learning_rate": 0.00010774774774774776,
181
+ "loss": 0.3151,
182
+ "step": 260
183
+ },
184
+ {
185
+ "epoch": 2.43,
186
+ "learning_rate": 0.00010414414414414416,
187
+ "loss": 0.2666,
188
+ "step": 270
189
+ },
190
+ {
191
+ "epoch": 2.52,
192
+ "learning_rate": 0.00010054054054054053,
193
+ "loss": 0.404,
194
+ "step": 280
195
+ },
196
+ {
197
+ "epoch": 2.61,
198
+ "learning_rate": 9.693693693693694e-05,
199
+ "loss": 0.3745,
200
+ "step": 290
201
+ },
202
+ {
203
+ "epoch": 2.7,
204
+ "learning_rate": 9.333333333333334e-05,
205
+ "loss": 0.3111,
206
+ "step": 300
207
+ },
208
+ {
209
+ "epoch": 2.7,
210
+ "eval_accuracy": 0.8932926829268293,
211
+ "eval_loss": 0.3442399501800537,
212
+ "eval_runtime": 4.2748,
213
+ "eval_samples_per_second": 76.729,
214
+ "eval_steps_per_second": 9.591,
215
+ "step": 300
216
+ },
217
+ {
218
+ "epoch": 2.79,
219
+ "learning_rate": 8.972972972972973e-05,
220
+ "loss": 0.3372,
221
+ "step": 310
222
+ },
223
+ {
224
+ "epoch": 2.88,
225
+ "learning_rate": 8.612612612612613e-05,
226
+ "loss": 0.4432,
227
+ "step": 320
228
+ },
229
+ {
230
+ "epoch": 2.97,
231
+ "learning_rate": 8.252252252252253e-05,
232
+ "loss": 0.3007,
233
+ "step": 330
234
+ },
235
+ {
236
+ "epoch": 3.06,
237
+ "learning_rate": 7.891891891891892e-05,
238
+ "loss": 0.2942,
239
+ "step": 340
240
+ },
241
+ {
242
+ "epoch": 3.15,
243
+ "learning_rate": 7.531531531531531e-05,
244
+ "loss": 0.2356,
245
+ "step": 350
246
+ },
247
+ {
248
+ "epoch": 3.24,
249
+ "learning_rate": 7.171171171171172e-05,
250
+ "loss": 0.2971,
251
+ "step": 360
252
+ },
253
+ {
254
+ "epoch": 3.33,
255
+ "learning_rate": 6.810810810810811e-05,
256
+ "loss": 0.2004,
257
+ "step": 370
258
+ },
259
+ {
260
+ "epoch": 3.42,
261
+ "learning_rate": 6.45045045045045e-05,
262
+ "loss": 0.3405,
263
+ "step": 380
264
+ },
265
+ {
266
+ "epoch": 3.51,
267
+ "learning_rate": 6.09009009009009e-05,
268
+ "loss": 0.3609,
269
+ "step": 390
270
+ },
271
+ {
272
+ "epoch": 3.6,
273
+ "learning_rate": 5.7297297297297305e-05,
274
+ "loss": 0.3709,
275
+ "step": 400
276
+ },
277
+ {
278
+ "epoch": 3.6,
279
+ "eval_accuracy": 0.9146341463414634,
280
+ "eval_loss": 0.3073720932006836,
281
+ "eval_runtime": 4.3122,
282
+ "eval_samples_per_second": 76.064,
283
+ "eval_steps_per_second": 9.508,
284
+ "step": 400
285
+ },
286
+ {
287
+ "epoch": 3.69,
288
+ "learning_rate": 5.369369369369369e-05,
289
+ "loss": 0.3094,
290
+ "step": 410
291
+ },
292
+ {
293
+ "epoch": 3.78,
294
+ "learning_rate": 5.009009009009009e-05,
295
+ "loss": 0.2294,
296
+ "step": 420
297
+ },
298
+ {
299
+ "epoch": 3.87,
300
+ "learning_rate": 4.648648648648649e-05,
301
+ "loss": 0.3965,
302
+ "step": 430
303
+ },
304
+ {
305
+ "epoch": 3.96,
306
+ "learning_rate": 4.2882882882882885e-05,
307
+ "loss": 0.2148,
308
+ "step": 440
309
+ },
310
+ {
311
+ "epoch": 4.05,
312
+ "learning_rate": 3.927927927927928e-05,
313
+ "loss": 0.3167,
314
+ "step": 450
315
+ },
316
+ {
317
+ "epoch": 4.14,
318
+ "learning_rate": 3.567567567567568e-05,
319
+ "loss": 0.1351,
320
+ "step": 460
321
+ },
322
+ {
323
+ "epoch": 4.23,
324
+ "learning_rate": 3.207207207207207e-05,
325
+ "loss": 0.275,
326
+ "step": 470
327
+ },
328
+ {
329
+ "epoch": 4.32,
330
+ "learning_rate": 2.8468468468468467e-05,
331
+ "loss": 0.1743,
332
+ "step": 480
333
+ },
334
+ {
335
+ "epoch": 4.41,
336
+ "learning_rate": 2.486486486486487e-05,
337
+ "loss": 0.1459,
338
+ "step": 490
339
+ },
340
+ {
341
+ "epoch": 4.5,
342
+ "learning_rate": 2.126126126126126e-05,
343
+ "loss": 0.1942,
344
+ "step": 500
345
+ },
346
+ {
347
+ "epoch": 4.5,
348
+ "eval_accuracy": 0.9115853658536586,
349
+ "eval_loss": 0.26617246866226196,
350
+ "eval_runtime": 4.2639,
351
+ "eval_samples_per_second": 76.925,
352
+ "eval_steps_per_second": 9.616,
353
+ "step": 500
354
+ },
355
+ {
356
+ "epoch": 4.59,
357
+ "learning_rate": 1.765765765765766e-05,
358
+ "loss": 0.1843,
359
+ "step": 510
360
+ },
361
+ {
362
+ "epoch": 4.68,
363
+ "learning_rate": 1.4054054054054055e-05,
364
+ "loss": 0.3644,
365
+ "step": 520
366
+ },
367
+ {
368
+ "epoch": 4.77,
369
+ "learning_rate": 1.0450450450450452e-05,
370
+ "loss": 0.1587,
371
+ "step": 530
372
+ },
373
+ {
374
+ "epoch": 4.86,
375
+ "learning_rate": 6.846846846846847e-06,
376
+ "loss": 0.1534,
377
+ "step": 540
378
+ },
379
+ {
380
+ "epoch": 4.95,
381
+ "learning_rate": 3.2432432432432437e-06,
382
+ "loss": 0.1954,
383
+ "step": 550
384
+ },
385
+ {
386
+ "epoch": 5.0,
387
+ "step": 555,
388
+ "total_flos": 5.153016901533696e+16,
389
+ "train_loss": 0.5183452816696854,
390
+ "train_runtime": 159.3173,
391
+ "train_samples_per_second": 55.487,
392
+ "train_steps_per_second": 3.484
393
+ }
394
+ ],
395
+ "max_steps": 555,
396
+ "num_train_epochs": 5,
397
+ "total_flos": 5.153016901533696e+16,
398
+ "trial_name": null,
399
+ "trial_params": null
400
+ }
models/apple/mobilevit-small-garbage/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:caccd1c746faf29e8da18b5a35181f77e8b3e0ebbfc29f59627f281fbd6e21d0
3
+ size 3439
models/vit-base-garbage/.DS_Store ADDED
Binary file (6.15 kB). View file
 
models/vit-base-garbage/all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_accuracy": 0.9651972157772621,
4
+ "eval_f1": 0.9598970898619156,
5
+ "eval_loss": 0.15073566138744354,
6
+ "eval_precision": 0.9580334762339824,
7
+ "eval_recall": 0.9624065831267469,
8
+ "eval_runtime": 8.4033,
9
+ "eval_samples_per_second": 51.289,
10
+ "eval_steps_per_second": 6.426,
11
+ "total_flos": 5.480429975941939e+17,
12
+ "train_loss": 0.2510145430092339,
13
+ "train_runtime": 644.0352,
14
+ "train_samples_per_second": 10.981,
15
+ "train_steps_per_second": 0.689
16
+ }
models/vit-base-garbage/checkpoint-300/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/vit-base-patch16-224-in21k",
3
+ "architectures": [
4
+ "ViTForImageClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "cardboard",
13
+ "1": "glass",
14
+ "2": "metal",
15
+ "3": "paper",
16
+ "4": "plastic",
17
+ "5": "trash"
18
+ },
19
+ "image_size": 224,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 3072,
22
+ "label2id": {
23
+ "cardboard": "0",
24
+ "glass": "1",
25
+ "metal": "2",
26
+ "paper": "3",
27
+ "plastic": "4",
28
+ "trash": "5"
29
+ },
30
+ "layer_norm_eps": 1e-12,
31
+ "model_type": "vit",
32
+ "num_attention_heads": 12,
33
+ "num_channels": 3,
34
+ "num_hidden_layers": 12,
35
+ "patch_size": 16,
36
+ "problem_type": "single_label_classification",
37
+ "qkv_bias": true,
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.24.0"
40
+ }
models/vit-base-garbage/checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45d91074a9863dc2fcba115a7d96f102c35d41ab1052c1055d4c491aa3157eb5
3
+ size 686542177
models/vit-base-garbage/checkpoint-300/preprocessor_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_resize": true,
4
+ "feature_extractor_type": "ViTFeatureExtractor",
5
+ "image_mean": [
6
+ 0.5,
7
+ 0.5,
8
+ 0.5
9
+ ],
10
+ "image_std": [
11
+ 0.5,
12
+ 0.5,
13
+ 0.5
14
+ ],
15
+ "resample": 2,
16
+ "size": 224
17
+ }
models/vit-base-garbage/checkpoint-300/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2bd1080c7a3858edda617a6b2e4814025d90da3f24591c916c4ca3618c247a6
3
+ size 343279281
models/vit-base-garbage/checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2ab9b52e364e49a6593a765f851264b1dd5e9b0207961295e70ed4788ef648c
3
+ size 14503
models/vit-base-garbage/checkpoint-300/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0967b9f865f16344c55f5ccc3cf7d6e8e97ca61dda304e931ca6bad130f48dd1
3
+ size 559
models/vit-base-garbage/checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d806fe60265194e6eba06de83a8b30480bcb818797ce556e7298374edc1993a
3
+ size 623
models/vit-base-garbage/checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,223 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.2087339609861374,
3
+ "best_model_checkpoint": "./vit-base-beans/checkpoint-300",
4
+ "epoch": 2.7027027027027026,
5
+ "global_step": 300,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.09,
12
+ "learning_rate": 0.0001954954954954955,
13
+ "loss": 1.552,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.18,
18
+ "learning_rate": 0.000190990990990991,
19
+ "loss": 1.0968,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.27,
24
+ "learning_rate": 0.0001864864864864865,
25
+ "loss": 0.8214,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.36,
30
+ "learning_rate": 0.000181981981981982,
31
+ "loss": 0.655,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.45,
36
+ "learning_rate": 0.0001774774774774775,
37
+ "loss": 0.5131,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.54,
42
+ "learning_rate": 0.000172972972972973,
43
+ "loss": 0.5863,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.63,
48
+ "learning_rate": 0.00016846846846846846,
49
+ "loss": 0.4228,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.72,
54
+ "learning_rate": 0.00016396396396396395,
55
+ "loss": 0.4562,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.81,
60
+ "learning_rate": 0.00015945945945945947,
61
+ "loss": 0.4817,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.9,
66
+ "learning_rate": 0.00015495495495495496,
67
+ "loss": 0.5217,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.9,
72
+ "eval_accuracy": 0.8871951219512195,
73
+ "eval_loss": 0.37025415897369385,
74
+ "eval_runtime": 82.1306,
75
+ "eval_samples_per_second": 3.994,
76
+ "eval_steps_per_second": 0.499,
77
+ "step": 100
78
+ },
79
+ {
80
+ "epoch": 0.99,
81
+ "learning_rate": 0.00015045045045045046,
82
+ "loss": 0.4543,
83
+ "step": 110
84
+ },
85
+ {
86
+ "epoch": 1.08,
87
+ "learning_rate": 0.00014594594594594595,
88
+ "loss": 0.2195,
89
+ "step": 120
90
+ },
91
+ {
92
+ "epoch": 1.17,
93
+ "learning_rate": 0.00014144144144144144,
94
+ "loss": 0.2413,
95
+ "step": 130
96
+ },
97
+ {
98
+ "epoch": 1.26,
99
+ "learning_rate": 0.00013693693693693693,
100
+ "loss": 0.2153,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 1.35,
105
+ "learning_rate": 0.00013243243243243243,
106
+ "loss": 0.2376,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 1.44,
111
+ "learning_rate": 0.00012792792792792795,
112
+ "loss": 0.215,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 1.53,
117
+ "learning_rate": 0.00012342342342342344,
118
+ "loss": 0.1317,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 1.62,
123
+ "learning_rate": 0.00011891891891891893,
124
+ "loss": 0.143,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 1.71,
129
+ "learning_rate": 0.00011441441441441443,
130
+ "loss": 0.2173,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 1.8,
135
+ "learning_rate": 0.00010990990990990993,
136
+ "loss": 0.1637,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 1.8,
141
+ "eval_accuracy": 0.8597560975609756,
142
+ "eval_loss": 0.5076180696487427,
143
+ "eval_runtime": 4.074,
144
+ "eval_samples_per_second": 80.511,
145
+ "eval_steps_per_second": 10.064,
146
+ "step": 200
147
+ },
148
+ {
149
+ "epoch": 1.89,
150
+ "learning_rate": 0.0001054054054054054,
151
+ "loss": 0.3325,
152
+ "step": 210
153
+ },
154
+ {
155
+ "epoch": 1.98,
156
+ "learning_rate": 0.00010090090090090089,
157
+ "loss": 0.2099,
158
+ "step": 220
159
+ },
160
+ {
161
+ "epoch": 2.07,
162
+ "learning_rate": 9.639639639639641e-05,
163
+ "loss": 0.1132,
164
+ "step": 230
165
+ },
166
+ {
167
+ "epoch": 2.16,
168
+ "learning_rate": 9.18918918918919e-05,
169
+ "loss": 0.1517,
170
+ "step": 240
171
+ },
172
+ {
173
+ "epoch": 2.25,
174
+ "learning_rate": 8.738738738738738e-05,
175
+ "loss": 0.0624,
176
+ "step": 250
177
+ },
178
+ {
179
+ "epoch": 2.34,
180
+ "learning_rate": 8.288288288288289e-05,
181
+ "loss": 0.0905,
182
+ "step": 260
183
+ },
184
+ {
185
+ "epoch": 2.43,
186
+ "learning_rate": 7.837837837837838e-05,
187
+ "loss": 0.0586,
188
+ "step": 270
189
+ },
190
+ {
191
+ "epoch": 2.52,
192
+ "learning_rate": 7.387387387387387e-05,
193
+ "loss": 0.071,
194
+ "step": 280
195
+ },
196
+ {
197
+ "epoch": 2.61,
198
+ "learning_rate": 6.936936936936938e-05,
199
+ "loss": 0.0393,
200
+ "step": 290
201
+ },
202
+ {
203
+ "epoch": 2.7,
204
+ "learning_rate": 6.486486486486487e-05,
205
+ "loss": 0.0291,
206
+ "step": 300
207
+ },
208
+ {
209
+ "epoch": 2.7,
210
+ "eval_accuracy": 0.9420731707317073,
211
+ "eval_loss": 0.2087339609861374,
212
+ "eval_runtime": 4.0634,
213
+ "eval_samples_per_second": 80.72,
214
+ "eval_steps_per_second": 10.09,
215
+ "step": 300
216
+ }
217
+ ],
218
+ "max_steps": 444,
219
+ "num_train_epochs": 4,
220
+ "total_flos": 3.7073496896077824e+17,
221
+ "trial_name": null,
222
+ "trial_params": null
223
+ }
models/vit-base-garbage/checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b388b03a0bc352599c3f1a7b29932b4fb98d8f608da2c3b16a52bf3c7baf4496
3
+ size 3375
models/vit-base-garbage/checkpoint-400/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/vit-base-patch16-224-in21k",
3
+ "architectures": [
4
+ "ViTForImageClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "cardboard",
13
+ "1": "glass",
14
+ "2": "metal",
15
+ "3": "paper",
16
+ "4": "plastic",
17
+ "5": "trash"
18
+ },
19
+ "image_size": 224,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 3072,
22
+ "label2id": {
23
+ "cardboard": "0",
24
+ "glass": "1",
25
+ "metal": "2",
26
+ "paper": "3",
27
+ "plastic": "4",
28
+ "trash": "5"
29
+ },
30
+ "layer_norm_eps": 1e-12,
31
+ "model_type": "vit",
32
+ "num_attention_heads": 12,
33
+ "num_channels": 3,
34
+ "num_hidden_layers": 12,
35
+ "patch_size": 16,
36
+ "problem_type": "single_label_classification",
37
+ "qkv_bias": true,
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.24.0"
40
+ }
models/vit-base-garbage/checkpoint-400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd9df6a4f363cff65999f1dd47fad0cee5d4a5a09ffe3efb82953939a0f36237
3
+ size 686542177
models/vit-base-garbage/checkpoint-400/preprocessor_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_resize": true,
4
+ "feature_extractor_type": "ViTFeatureExtractor",
5
+ "image_mean": [
6
+ 0.5,
7
+ 0.5,
8
+ 0.5
9
+ ],
10
+ "image_std": [
11
+ 0.5,
12
+ 0.5,
13
+ 0.5
14
+ ],
15
+ "resample": 2,
16
+ "size": 224
17
+ }
models/vit-base-garbage/checkpoint-400/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b61389241626facd22ef1a4236cf35bfbdcd4c59cf1e2a55cd438f5472ee1ab0
3
+ size 343279281
models/vit-base-garbage/checkpoint-400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc81a4ed07eaae97e601e6dad9894c368270413e9040807580143d45f03df338
3
+ size 14503
models/vit-base-garbage/checkpoint-400/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:476e510c8ea7edbd2b51d1e76a4e037820a5639381c0d8b5d32dafa492795a1e
3
+ size 559
models/vit-base-garbage/checkpoint-400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5a43e941fea784321070831dd2f6b1bc4f69813f39be747b387fb6a717115bf
3
+ size 623
models/vit-base-garbage/checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,292 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.172527477145195,
3
+ "best_model_checkpoint": "./vit-base-beans/checkpoint-400",
4
+ "epoch": 3.6036036036036037,
5
+ "global_step": 400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.09,
12
+ "learning_rate": 0.0001954954954954955,
13
+ "loss": 1.552,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.18,
18
+ "learning_rate": 0.000190990990990991,
19
+ "loss": 1.0968,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.27,
24
+ "learning_rate": 0.0001864864864864865,
25
+ "loss": 0.8214,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.36,
30
+ "learning_rate": 0.000181981981981982,
31
+ "loss": 0.655,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.45,
36
+ "learning_rate": 0.0001774774774774775,
37
+ "loss": 0.5131,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.54,
42
+ "learning_rate": 0.000172972972972973,
43
+ "loss": 0.5863,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.63,
48
+ "learning_rate": 0.00016846846846846846,
49
+ "loss": 0.4228,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.72,
54
+ "learning_rate": 0.00016396396396396395,
55
+ "loss": 0.4562,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.81,
60
+ "learning_rate": 0.00015945945945945947,
61
+ "loss": 0.4817,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.9,
66
+ "learning_rate": 0.00015495495495495496,
67
+ "loss": 0.5217,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.9,
72
+ "eval_accuracy": 0.8871951219512195,
73
+ "eval_loss": 0.37025415897369385,
74
+ "eval_runtime": 82.1306,
75
+ "eval_samples_per_second": 3.994,
76
+ "eval_steps_per_second": 0.499,
77
+ "step": 100
78
+ },
79
+ {
80
+ "epoch": 0.99,
81
+ "learning_rate": 0.00015045045045045046,
82
+ "loss": 0.4543,
83
+ "step": 110
84
+ },
85
+ {
86
+ "epoch": 1.08,
87
+ "learning_rate": 0.00014594594594594595,
88
+ "loss": 0.2195,
89
+ "step": 120
90
+ },
91
+ {
92
+ "epoch": 1.17,
93
+ "learning_rate": 0.00014144144144144144,
94
+ "loss": 0.2413,
95
+ "step": 130
96
+ },
97
+ {
98
+ "epoch": 1.26,
99
+ "learning_rate": 0.00013693693693693693,
100
+ "loss": 0.2153,
101
+ "step": 140
102
+ },
103
+ {
104
+ "epoch": 1.35,
105
+ "learning_rate": 0.00013243243243243243,
106
+ "loss": 0.2376,
107
+ "step": 150
108
+ },
109
+ {
110
+ "epoch": 1.44,
111
+ "learning_rate": 0.00012792792792792795,
112
+ "loss": 0.215,
113
+ "step": 160
114
+ },
115
+ {
116
+ "epoch": 1.53,
117
+ "learning_rate": 0.00012342342342342344,
118
+ "loss": 0.1317,
119
+ "step": 170
120
+ },
121
+ {
122
+ "epoch": 1.62,
123
+ "learning_rate": 0.00011891891891891893,
124
+ "loss": 0.143,
125
+ "step": 180
126
+ },
127
+ {
128
+ "epoch": 1.71,
129
+ "learning_rate": 0.00011441441441441443,
130
+ "loss": 0.2173,
131
+ "step": 190
132
+ },
133
+ {
134
+ "epoch": 1.8,
135
+ "learning_rate": 0.00010990990990990993,
136
+ "loss": 0.1637,
137
+ "step": 200
138
+ },
139
+ {
140
+ "epoch": 1.8,
141
+ "eval_accuracy": 0.8597560975609756,
142
+ "eval_loss": 0.5076180696487427,
143
+ "eval_runtime": 4.074,
144
+ "eval_samples_per_second": 80.511,
145
+ "eval_steps_per_second": 10.064,
146
+ "step": 200
147
+ },
148
+ {
149
+ "epoch": 1.89,
150
+ "learning_rate": 0.0001054054054054054,
151
+ "loss": 0.3325,
152
+ "step": 210
153
+ },
154
+ {
155
+ "epoch": 1.98,
156
+ "learning_rate": 0.00010090090090090089,
157
+ "loss": 0.2099,
158
+ "step": 220
159
+ },
160
+ {
161
+ "epoch": 2.07,
162
+ "learning_rate": 9.639639639639641e-05,
163
+ "loss": 0.1132,
164
+ "step": 230
165
+ },
166
+ {
167
+ "epoch": 2.16,
168
+ "learning_rate": 9.18918918918919e-05,
169
+ "loss": 0.1517,
170
+ "step": 240
171
+ },
172
+ {
173
+ "epoch": 2.25,
174
+ "learning_rate": 8.738738738738738e-05,
175
+ "loss": 0.0624,
176
+ "step": 250
177
+ },
178
+ {
179
+ "epoch": 2.34,
180
+ "learning_rate": 8.288288288288289e-05,
181
+ "loss": 0.0905,
182
+ "step": 260
183
+ },
184
+ {
185
+ "epoch": 2.43,
186
+ "learning_rate": 7.837837837837838e-05,
187
+ "loss": 0.0586,
188
+ "step": 270
189
+ },
190
+ {
191
+ "epoch": 2.52,
192
+ "learning_rate": 7.387387387387387e-05,
193
+ "loss": 0.071,
194
+ "step": 280
195
+ },
196
+ {
197
+ "epoch": 2.61,
198
+ "learning_rate": 6.936936936936938e-05,
199
+ "loss": 0.0393,
200
+ "step": 290
201
+ },
202
+ {
203
+ "epoch": 2.7,
204
+ "learning_rate": 6.486486486486487e-05,
205
+ "loss": 0.0291,
206
+ "step": 300
207
+ },
208
+ {
209
+ "epoch": 2.7,
210
+ "eval_accuracy": 0.9420731707317073,
211
+ "eval_loss": 0.2087339609861374,
212
+ "eval_runtime": 4.0634,
213
+ "eval_samples_per_second": 80.72,
214
+ "eval_steps_per_second": 10.09,
215
+ "step": 300
216
+ },
217
+ {
218
+ "epoch": 2.79,
219
+ "learning_rate": 6.0360360360360365e-05,
220
+ "loss": 0.075,
221
+ "step": 310
222
+ },
223
+ {
224
+ "epoch": 2.88,
225
+ "learning_rate": 5.585585585585585e-05,
226
+ "loss": 0.1357,
227
+ "step": 320
228
+ },
229
+ {
230
+ "epoch": 2.97,
231
+ "learning_rate": 5.135135135135135e-05,
232
+ "loss": 0.0312,
233
+ "step": 330
234
+ },
235
+ {
236
+ "epoch": 3.06,
237
+ "learning_rate": 4.684684684684685e-05,
238
+ "loss": 0.0753,
239
+ "step": 340
240
+ },
241
+ {
242
+ "epoch": 3.15,
243
+ "learning_rate": 4.234234234234234e-05,
244
+ "loss": 0.0231,
245
+ "step": 350
246
+ },
247
+ {
248
+ "epoch": 3.24,
249
+ "learning_rate": 3.783783783783784e-05,
250
+ "loss": 0.0266,
251
+ "step": 360
252
+ },
253
+ {
254
+ "epoch": 3.33,
255
+ "learning_rate": 3.3333333333333335e-05,
256
+ "loss": 0.023,
257
+ "step": 370
258
+ },
259
+ {
260
+ "epoch": 3.42,
261
+ "learning_rate": 2.882882882882883e-05,
262
+ "loss": 0.0224,
263
+ "step": 380
264
+ },
265
+ {
266
+ "epoch": 3.51,
267
+ "learning_rate": 2.4324324324324327e-05,
268
+ "loss": 0.0251,
269
+ "step": 390
270
+ },
271
+ {
272
+ "epoch": 3.6,
273
+ "learning_rate": 1.981981981981982e-05,
274
+ "loss": 0.025,
275
+ "step": 400
276
+ },
277
+ {
278
+ "epoch": 3.6,
279
+ "eval_accuracy": 0.9573170731707317,
280
+ "eval_loss": 0.172527477145195,
281
+ "eval_runtime": 4.0556,
282
+ "eval_samples_per_second": 80.877,
283
+ "eval_steps_per_second": 10.11,
284
+ "step": 400
285
+ }
286
+ ],
287
+ "max_steps": 444,
288
+ "num_train_epochs": 4,
289
+ "total_flos": 4.9410663923367936e+17,
290
+ "trial_name": null,
291
+ "trial_params": null
292
+ }
models/vit-base-garbage/checkpoint-400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b388b03a0bc352599c3f1a7b29932b4fb98d8f608da2c3b16a52bf3c7baf4496
3
+ size 3375
models/vit-base-garbage/config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/vit-base-patch16-224-in21k",
3
+ "architectures": [
4
+ "ViTForImageClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "cardboard",
13
+ "1": "glass",
14
+ "2": "metal",
15
+ "3": "paper",
16
+ "4": "plastic",
17
+ "5": "trash"
18
+ },
19
+ "image_size": 224,
20
+ "initializer_range": 0.02,
21
+ "intermediate_size": 3072,
22
+ "label2id": {
23
+ "cardboard": "0",
24
+ "glass": "1",
25
+ "metal": "2",
26
+ "paper": "3",
27
+ "plastic": "4",
28
+ "trash": "5"
29
+ },
30
+ "layer_norm_eps": 1e-12,
31
+ "model_type": "vit",
32
+ "num_attention_heads": 12,
33
+ "num_channels": 3,
34
+ "num_hidden_layers": 12,
35
+ "patch_size": 16,
36
+ "problem_type": "single_label_classification",
37
+ "qkv_bias": true,
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.24.0"
40
+ }
models/vit-base-garbage/preprocessor_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_resize": true,
4
+ "feature_extractor_type": "ViTFeatureExtractor",
5
+ "image_mean": [
6
+ 0.5,
7
+ 0.5,
8
+ 0.5
9
+ ],
10
+ "image_std": [
11
+ 0.5,
12
+ 0.5,
13
+ 0.5
14
+ ],
15
+ "resample": 2,
16
+ "size": 224
17
+ }
models/vit-base-garbage/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b61389241626facd22ef1a4236cf35bfbdcd4c59cf1e2a55cd438f5472ee1ab0
3
+ size 343279281
models/vit-base-garbage/runs/Dec12_19-10-35_ed23ba8f6546/1670872238.0395455/events.out.tfevents.1670872238.ed23ba8f6546.76.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6fad8c207a1a9f50421bd83fe5c5536f74f143ef54da571f20c56126e9e0654
3
+ size 5513
models/vit-base-garbage/runs/Dec12_19-10-35_ed23ba8f6546/events.out.tfevents.1670872238.ed23ba8f6546.76.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e69af10d73925b30f110cdad2f552fe6c74f9e26c2b71eea25bd103822c32c6c
3
+ size 3944
models/vit-base-garbage/runs/Dec12_19-10-35_ed23ba8f6546/events.out.tfevents.1670872287.ed23ba8f6546.76.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83f3175ad6971770c09cab115bd878851c1b6955a575e58a2f5ecd8f401aba66
3
+ size 298
models/vit-base-garbage/runs/Dec12_21-11-38_30c96569bd60/events.out.tfevents.1670879785.30c96569bd60.71.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea778ed9b6506059fdb91fa8755db5f01a45726eb172139fc2a361675d576579
3
+ size 298
models/vit-base-garbage/runs/Dec12_21-51-08_30c96569bd60/events.out.tfevents.1670881883.30c96569bd60.71.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a10eb065b3cacdcdb9c4be462f2ee61e3b114891cc8d09d374632916c531c688
3
+ size 438
models/vit-base-garbage/runs/Nov28_22-29-34_09524550a2a1/1669674600.7132652/events.out.tfevents.1669674600.09524550a2a1.78.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3071e168dc29887ffce431e759528cdfe8d5e8eabaeb83255bdb68d496698dd
3
+ size 5475