Hugging Face's logo

Augusto777
/
swinv2-tiny-patch4-window8-256-RH

TensorBoard
Safetensors
swinv2
Generated from Trainer
Eval Results
Model card Files Metrics Community
Augusto777 commited on
Commit
a4b5f2c
·
verified ·
1 Parent(s): c5204c7

End of training

Browse files
Files changed (11) hide show
  1. README.md +117 -0
  2. all_results.json +12 -0
  3. config.json +54 -0
  4. eval_results.json +8 -0
  5. model.safetensors +3 -0
  6. preprocessor_config.json +22 -0
  7. runs/Dec01_10-12-57_DESKTOP-SKBE9FB/events.out.tfevents.1733069578.DESKTOP-SKBE9FB.12596.0 +3 -0
  8. runs/Dec01_10-12-57_DESKTOP-SKBE9FB/events.out.tfevents.1733070161.DESKTOP-SKBE9FB.12596.1 +3 -0
  9. train_results.json +7 -0
  10. trainer_state.json +582 -0
  11. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: microsoft/swinv2-tiny-patch4-window8-256
4
+ tags:
5
+ - generated_from_trainer
6
+ datasets:
7
+ - imagefolder
8
+ metrics:
9
+ - accuracy
10
+ model-index:
11
+ - name: swinv2-tiny-patch4-window8-256-RH
12
+ results:
13
+ - task:
14
+ name: Image Classification
15
+ type: image-classification
16
+ dataset:
17
+ name: imagefolder
18
+ type: imagefolder
19
+ config: default
20
+ split: validation
21
+ args: default
22
+ metrics:
23
+ - name: Accuracy
24
+ type: accuracy
25
+ value: 0.6448598130841121
26
+ ---
27
+
28
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
29
+ should probably proofread and complete it, then remove this comment. -->
30
+
31
+ # swinv2-tiny-patch4-window8-256-RH
32
+
33
+ This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
34
+ It achieves the following results on the evaluation set:
35
+ - Loss: 0.6681
36
+ - Accuracy: 0.6449
37
+
38
+ ## Model description
39
+
40
+ More information needed
41
+
42
+ ## Intended uses & limitations
43
+
44
+ More information needed
45
+
46
+ ## Training and evaluation data
47
+
48
+ More information needed
49
+
50
+ ## Training procedure
51
+
52
+ ### Training hyperparameters
53
+
54
+ The following hyperparameters were used during training:
55
+ - learning_rate: 0.00015
56
+ - train_batch_size: 16
57
+ - eval_batch_size: 16
58
+ - seed: 42
59
+ - gradient_accumulation_steps: 4
60
+ - total_train_batch_size: 64
61
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
62
+ - lr_scheduler_type: linear
63
+ - lr_scheduler_warmup_ratio: 0.1
64
+ - num_epochs: 40
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
69
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
70
+ | No log | 1.0 | 8 | 4.5659 | 0.4112 |
71
+ | 4.5175 | 2.0 | 16 | 3.6362 | 0.4112 |
72
+ | 3.9284 | 3.0 | 24 | 1.6019 | 0.4112 |
73
+ | 1.6086 | 4.0 | 32 | 0.7110 | 0.4112 |
74
+ | 0.7392 | 5.0 | 40 | 0.6825 | 0.5888 |
75
+ | 0.7392 | 6.0 | 48 | 0.6795 | 0.5888 |
76
+ | 0.7073 | 7.0 | 56 | 0.6814 | 0.5888 |
77
+ | 0.6956 | 8.0 | 64 | 0.7061 | 0.5888 |
78
+ | 0.6898 | 9.0 | 72 | 0.7014 | 0.5888 |
79
+ | 0.7026 | 10.0 | 80 | 0.7214 | 0.4112 |
80
+ | 0.7026 | 11.0 | 88 | 0.7186 | 0.5888 |
81
+ | 0.7696 | 12.0 | 96 | 0.6837 | 0.5888 |
82
+ | 0.6909 | 13.0 | 104 | 0.6823 | 0.5888 |
83
+ | 0.6799 | 14.0 | 112 | 0.6781 | 0.5888 |
84
+ | 0.6782 | 15.0 | 120 | 0.6938 | 0.5888 |
85
+ | 0.6782 | 16.0 | 128 | 0.6766 | 0.5888 |
86
+ | 0.6952 | 17.0 | 136 | 0.7123 | 0.5888 |
87
+ | 0.6875 | 18.0 | 144 | 0.6891 | 0.5607 |
88
+ | 0.6919 | 19.0 | 152 | 0.7076 | 0.5888 |
89
+ | 0.6751 | 20.0 | 160 | 0.7011 | 0.4953 |
90
+ | 0.6751 | 21.0 | 168 | 0.6962 | 0.5888 |
91
+ | 0.689 | 22.0 | 176 | 0.6857 | 0.5701 |
92
+ | 0.6826 | 23.0 | 184 | 0.6935 | 0.5888 |
93
+ | 0.6841 | 24.0 | 192 | 0.7219 | 0.5888 |
94
+ | 0.6657 | 25.0 | 200 | 0.6610 | 0.5888 |
95
+ | 0.6657 | 26.0 | 208 | 0.6681 | 0.6449 |
96
+ | 0.6524 | 27.0 | 216 | 0.7225 | 0.5888 |
97
+ | 0.6567 | 28.0 | 224 | 0.7117 | 0.5888 |
98
+ | 0.6402 | 29.0 | 232 | 0.6999 | 0.6262 |
99
+ | 0.66 | 30.0 | 240 | 0.6799 | 0.6075 |
100
+ | 0.66 | 31.0 | 248 | 0.6677 | 0.6075 |
101
+ | 0.6469 | 32.0 | 256 | 0.6735 | 0.5981 |
102
+ | 0.6355 | 33.0 | 264 | 0.6853 | 0.6168 |
103
+ | 0.6245 | 34.0 | 272 | 0.7008 | 0.6262 |
104
+ | 0.6306 | 35.0 | 280 | 0.6990 | 0.5981 |
105
+ | 0.6306 | 36.0 | 288 | 0.6981 | 0.6355 |
106
+ | 0.6208 | 37.0 | 296 | 0.7103 | 0.6262 |
107
+ | 0.6339 | 38.0 | 304 | 0.7050 | 0.6355 |
108
+ | 0.5959 | 39.0 | 312 | 0.6989 | 0.6355 |
109
+ | 0.6059 | 40.0 | 320 | 0.6990 | 0.6355 |
110
+
111
+
112
+ ### Framework versions
113
+
114
+ - Transformers 4.36.2
115
+ - Pytorch 2.1.2+cu118
116
+ - Datasets 2.16.1
117
+ - Tokenizers 0.15.0
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 40.0,
3
+ "eval_accuracy": 0.6448598130841121,
4
+ "eval_loss": 0.6681405305862427,
5
+ "eval_runtime": 2.1165,
6
+ "eval_samples_per_second": 50.556,
7
+ "eval_steps_per_second": 3.307,
8
+ "train_loss": 0.9213116973638534,
9
+ "train_runtime": 581.2908,
10
+ "train_samples_per_second": 34.269,
11
+ "train_steps_per_second": 0.55
12
+ }
config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/swinv2-tiny-patch4-window8-256",
3
+ "architectures": [
4
+ "Swinv2ForImageClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "depths": [
8
+ 2,
9
+ 2,
10
+ 6,
11
+ 2
12
+ ],
13
+ "drop_path_rate": 0.1,
14
+ "embed_dim": 96,
15
+ "encoder_stride": 32,
16
+ "hidden_act": "gelu",
17
+ "hidden_dropout_prob": 0.0,
18
+ "hidden_size": 768,
19
+ "id2label": {
20
+ "0": "HR",
21
+ "1": "No HR"
22
+ },
23
+ "image_size": 256,
24
+ "initializer_range": 0.02,
25
+ "label2id": {
26
+ "HR": 0,
27
+ "No HR": 1
28
+ },
29
+ "layer_norm_eps": 1e-05,
30
+ "mlp_ratio": 4.0,
31
+ "model_type": "swinv2",
32
+ "num_channels": 3,
33
+ "num_heads": [
34
+ 3,
35
+ 6,
36
+ 12,
37
+ 24
38
+ ],
39
+ "num_layers": 4,
40
+ "patch_size": 4,
41
+ "path_norm": true,
42
+ "pretrained_window_sizes": [
43
+ 0,
44
+ 0,
45
+ 0,
46
+ 0
47
+ ],
48
+ "problem_type": "single_label_classification",
49
+ "qkv_bias": true,
50
+ "torch_dtype": "float32",
51
+ "transformers_version": "4.36.2",
52
+ "use_absolute_embeddings": false,
53
+ "window_size": 8
54
+ }
eval_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 40.0,
3
+ "eval_accuracy": 0.6448598130841121,
4
+ "eval_loss": 0.6681405305862427,
5
+ "eval_runtime": 2.1165,
6
+ "eval_samples_per_second": 50.556,
7
+ "eval_steps_per_second": 3.307
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6be82f55800482222ec52265737ecf590a2ed114bb2716e0ce24c4dc4443a322
3
+ size 110350128
preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.485,
7
+ 0.456,
8
+ 0.406
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.229,
13
+ 0.224,
14
+ 0.225
15
+ ],
16
+ "resample": 3,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 256,
20
+ "width": 256
21
+ }
22
+ }
runs/Dec01_10-12-57_DESKTOP-SKBE9FB/events.out.tfevents.1733069578.DESKTOP-SKBE9FB.12596.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c567a4161112737090e29d080245395baea8cb499532618080350ed1859d290
3
+ size 22872
runs/Dec01_10-12-57_DESKTOP-SKBE9FB/events.out.tfevents.1733070161.DESKTOP-SKBE9FB.12596.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7775d3eb0f08c3e70088ebf0b679e920560b22c352def84b668ecbda63cd5f06
3
+ size 411
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 40.0,
3
+ "train_loss": 0.9213116973638534,
4
+ "train_runtime": 581.2908,
5
+ "train_samples_per_second": 34.269,
6
+ "train_steps_per_second": 0.55
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,582 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6448598130841121,
3
+ "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-RH\\checkpoint-208",
4
+ "epoch": 40.0,
5
+ "eval_steps": 500,
6
+ "global_step": 320,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "eval_accuracy": 0.411214953271028,
14
+ "eval_loss": 4.565852165222168,
15
+ "eval_runtime": 1.9881,
16
+ "eval_samples_per_second": 53.821,
17
+ "eval_steps_per_second": 3.521,
18
+ "step": 8
19
+ },
20
+ {
21
+ "epoch": 1.25,
22
+ "learning_rate": 4.6874999999999994e-05,
23
+ "loss": 4.5175,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "eval_accuracy": 0.411214953271028,
29
+ "eval_loss": 3.636173963546753,
30
+ "eval_runtime": 1.951,
31
+ "eval_samples_per_second": 54.845,
32
+ "eval_steps_per_second": 3.588,
33
+ "step": 16
34
+ },
35
+ {
36
+ "epoch": 2.5,
37
+ "learning_rate": 9.374999999999999e-05,
38
+ "loss": 3.9284,
39
+ "step": 20
40
+ },
41
+ {
42
+ "epoch": 3.0,
43
+ "eval_accuracy": 0.411214953271028,
44
+ "eval_loss": 1.6019313335418701,
45
+ "eval_runtime": 1.8634,
46
+ "eval_samples_per_second": 57.421,
47
+ "eval_steps_per_second": 3.757,
48
+ "step": 24
49
+ },
50
+ {
51
+ "epoch": 3.75,
52
+ "learning_rate": 0.000140625,
53
+ "loss": 1.6086,
54
+ "step": 30
55
+ },
56
+ {
57
+ "epoch": 4.0,
58
+ "eval_accuracy": 0.411214953271028,
59
+ "eval_loss": 0.7110145092010498,
60
+ "eval_runtime": 1.8614,
61
+ "eval_samples_per_second": 57.482,
62
+ "eval_steps_per_second": 3.761,
63
+ "step": 32
64
+ },
65
+ {
66
+ "epoch": 5.0,
67
+ "learning_rate": 0.00014583333333333332,
68
+ "loss": 0.7392,
69
+ "step": 40
70
+ },
71
+ {
72
+ "epoch": 5.0,
73
+ "eval_accuracy": 0.5887850467289719,
74
+ "eval_loss": 0.6824622750282288,
75
+ "eval_runtime": 1.8634,
76
+ "eval_samples_per_second": 57.42,
77
+ "eval_steps_per_second": 3.756,
78
+ "step": 40
79
+ },
80
+ {
81
+ "epoch": 6.0,
82
+ "eval_accuracy": 0.5887850467289719,
83
+ "eval_loss": 0.679547905921936,
84
+ "eval_runtime": 1.9335,
85
+ "eval_samples_per_second": 55.341,
86
+ "eval_steps_per_second": 3.62,
87
+ "step": 48
88
+ },
89
+ {
90
+ "epoch": 6.25,
91
+ "learning_rate": 0.000140625,
92
+ "loss": 0.7073,
93
+ "step": 50
94
+ },
95
+ {
96
+ "epoch": 7.0,
97
+ "eval_accuracy": 0.5887850467289719,
98
+ "eval_loss": 0.6814281940460205,
99
+ "eval_runtime": 1.8789,
100
+ "eval_samples_per_second": 56.947,
101
+ "eval_steps_per_second": 3.725,
102
+ "step": 56
103
+ },
104
+ {
105
+ "epoch": 7.5,
106
+ "learning_rate": 0.00013541666666666666,
107
+ "loss": 0.6956,
108
+ "step": 60
109
+ },
110
+ {
111
+ "epoch": 8.0,
112
+ "eval_accuracy": 0.5887850467289719,
113
+ "eval_loss": 0.7060601711273193,
114
+ "eval_runtime": 1.8394,
115
+ "eval_samples_per_second": 58.17,
116
+ "eval_steps_per_second": 3.805,
117
+ "step": 64
118
+ },
119
+ {
120
+ "epoch": 8.75,
121
+ "learning_rate": 0.00013020833333333333,
122
+ "loss": 0.6898,
123
+ "step": 70
124
+ },
125
+ {
126
+ "epoch": 9.0,
127
+ "eval_accuracy": 0.5887850467289719,
128
+ "eval_loss": 0.7014247179031372,
129
+ "eval_runtime": 1.9775,
130
+ "eval_samples_per_second": 54.11,
131
+ "eval_steps_per_second": 3.54,
132
+ "step": 72
133
+ },
134
+ {
135
+ "epoch": 10.0,
136
+ "learning_rate": 0.000125,
137
+ "loss": 0.7026,
138
+ "step": 80
139
+ },
140
+ {
141
+ "epoch": 10.0,
142
+ "eval_accuracy": 0.411214953271028,
143
+ "eval_loss": 0.7213616967201233,
144
+ "eval_runtime": 1.881,
145
+ "eval_samples_per_second": 56.886,
146
+ "eval_steps_per_second": 3.721,
147
+ "step": 80
148
+ },
149
+ {
150
+ "epoch": 11.0,
151
+ "eval_accuracy": 0.5887850467289719,
152
+ "eval_loss": 0.7185712456703186,
153
+ "eval_runtime": 1.9299,
154
+ "eval_samples_per_second": 55.442,
155
+ "eval_steps_per_second": 3.627,
156
+ "step": 88
157
+ },
158
+ {
159
+ "epoch": 11.25,
160
+ "learning_rate": 0.00011979166666666666,
161
+ "loss": 0.7696,
162
+ "step": 90
163
+ },
164
+ {
165
+ "epoch": 12.0,
166
+ "eval_accuracy": 0.5887850467289719,
167
+ "eval_loss": 0.6837121844291687,
168
+ "eval_runtime": 1.8429,
169
+ "eval_samples_per_second": 58.06,
170
+ "eval_steps_per_second": 3.798,
171
+ "step": 96
172
+ },
173
+ {
174
+ "epoch": 12.5,
175
+ "learning_rate": 0.00011458333333333332,
176
+ "loss": 0.6909,
177
+ "step": 100
178
+ },
179
+ {
180
+ "epoch": 13.0,
181
+ "eval_accuracy": 0.5887850467289719,
182
+ "eval_loss": 0.6822527647018433,
183
+ "eval_runtime": 1.8714,
184
+ "eval_samples_per_second": 57.175,
185
+ "eval_steps_per_second": 3.74,
186
+ "step": 104
187
+ },
188
+ {
189
+ "epoch": 13.75,
190
+ "learning_rate": 0.00010937499999999999,
191
+ "loss": 0.6799,
192
+ "step": 110
193
+ },
194
+ {
195
+ "epoch": 14.0,
196
+ "eval_accuracy": 0.5887850467289719,
197
+ "eval_loss": 0.6780915260314941,
198
+ "eval_runtime": 1.8835,
199
+ "eval_samples_per_second": 56.811,
200
+ "eval_steps_per_second": 3.717,
201
+ "step": 112
202
+ },
203
+ {
204
+ "epoch": 15.0,
205
+ "learning_rate": 0.00010416666666666666,
206
+ "loss": 0.6782,
207
+ "step": 120
208
+ },
209
+ {
210
+ "epoch": 15.0,
211
+ "eval_accuracy": 0.5887850467289719,
212
+ "eval_loss": 0.6938412189483643,
213
+ "eval_runtime": 2.0385,
214
+ "eval_samples_per_second": 52.489,
215
+ "eval_steps_per_second": 3.434,
216
+ "step": 120
217
+ },
218
+ {
219
+ "epoch": 16.0,
220
+ "eval_accuracy": 0.5887850467289719,
221
+ "eval_loss": 0.6766122579574585,
222
+ "eval_runtime": 2.019,
223
+ "eval_samples_per_second": 52.997,
224
+ "eval_steps_per_second": 3.467,
225
+ "step": 128
226
+ },
227
+ {
228
+ "epoch": 16.25,
229
+ "learning_rate": 9.895833333333332e-05,
230
+ "loss": 0.6952,
231
+ "step": 130
232
+ },
233
+ {
234
+ "epoch": 17.0,
235
+ "eval_accuracy": 0.5887850467289719,
236
+ "eval_loss": 0.7123367190361023,
237
+ "eval_runtime": 1.8724,
238
+ "eval_samples_per_second": 57.145,
239
+ "eval_steps_per_second": 3.738,
240
+ "step": 136
241
+ },
242
+ {
243
+ "epoch": 17.5,
244
+ "learning_rate": 9.374999999999999e-05,
245
+ "loss": 0.6875,
246
+ "step": 140
247
+ },
248
+ {
249
+ "epoch": 18.0,
250
+ "eval_accuracy": 0.5607476635514018,
251
+ "eval_loss": 0.6890613436698914,
252
+ "eval_runtime": 1.9095,
253
+ "eval_samples_per_second": 56.037,
254
+ "eval_steps_per_second": 3.666,
255
+ "step": 144
256
+ },
257
+ {
258
+ "epoch": 18.75,
259
+ "learning_rate": 8.854166666666666e-05,
260
+ "loss": 0.6919,
261
+ "step": 150
262
+ },
263
+ {
264
+ "epoch": 19.0,
265
+ "eval_accuracy": 0.5887850467289719,
266
+ "eval_loss": 0.7076395153999329,
267
+ "eval_runtime": 1.941,
268
+ "eval_samples_per_second": 55.127,
269
+ "eval_steps_per_second": 3.606,
270
+ "step": 152
271
+ },
272
+ {
273
+ "epoch": 20.0,
274
+ "learning_rate": 8.333333333333333e-05,
275
+ "loss": 0.6751,
276
+ "step": 160
277
+ },
278
+ {
279
+ "epoch": 20.0,
280
+ "eval_accuracy": 0.4953271028037383,
281
+ "eval_loss": 0.7010804414749146,
282
+ "eval_runtime": 1.8779,
283
+ "eval_samples_per_second": 56.977,
284
+ "eval_steps_per_second": 3.727,
285
+ "step": 160
286
+ },
287
+ {
288
+ "epoch": 21.0,
289
+ "eval_accuracy": 0.5887850467289719,
290
+ "eval_loss": 0.6961632966995239,
291
+ "eval_runtime": 1.8549,
292
+ "eval_samples_per_second": 57.684,
293
+ "eval_steps_per_second": 3.774,
294
+ "step": 168
295
+ },
296
+ {
297
+ "epoch": 21.25,
298
+ "learning_rate": 7.8125e-05,
299
+ "loss": 0.689,
300
+ "step": 170
301
+ },
302
+ {
303
+ "epoch": 22.0,
304
+ "eval_accuracy": 0.5700934579439252,
305
+ "eval_loss": 0.6856827139854431,
306
+ "eval_runtime": 1.8844,
307
+ "eval_samples_per_second": 56.781,
308
+ "eval_steps_per_second": 3.715,
309
+ "step": 176
310
+ },
311
+ {
312
+ "epoch": 22.5,
313
+ "learning_rate": 7.291666666666666e-05,
314
+ "loss": 0.6826,
315
+ "step": 180
316
+ },
317
+ {
318
+ "epoch": 23.0,
319
+ "eval_accuracy": 0.5887850467289719,
320
+ "eval_loss": 0.6935387253761292,
321
+ "eval_runtime": 1.9324,
322
+ "eval_samples_per_second": 55.37,
323
+ "eval_steps_per_second": 3.622,
324
+ "step": 184
325
+ },
326
+ {
327
+ "epoch": 23.75,
328
+ "learning_rate": 6.770833333333333e-05,
329
+ "loss": 0.6841,
330
+ "step": 190
331
+ },
332
+ {
333
+ "epoch": 24.0,
334
+ "eval_accuracy": 0.5887850467289719,
335
+ "eval_loss": 0.7218518853187561,
336
+ "eval_runtime": 1.9105,
337
+ "eval_samples_per_second": 56.008,
338
+ "eval_steps_per_second": 3.664,
339
+ "step": 192
340
+ },
341
+ {
342
+ "epoch": 25.0,
343
+ "learning_rate": 6.25e-05,
344
+ "loss": 0.6657,
345
+ "step": 200
346
+ },
347
+ {
348
+ "epoch": 25.0,
349
+ "eval_accuracy": 0.5887850467289719,
350
+ "eval_loss": 0.6609508395195007,
351
+ "eval_runtime": 1.8645,
352
+ "eval_samples_per_second": 57.39,
353
+ "eval_steps_per_second": 3.754,
354
+ "step": 200
355
+ },
356
+ {
357
+ "epoch": 26.0,
358
+ "eval_accuracy": 0.6448598130841121,
359
+ "eval_loss": 0.6681405305862427,
360
+ "eval_runtime": 1.8574,
361
+ "eval_samples_per_second": 57.607,
362
+ "eval_steps_per_second": 3.769,
363
+ "step": 208
364
+ },
365
+ {
366
+ "epoch": 26.25,
367
+ "learning_rate": 5.729166666666666e-05,
368
+ "loss": 0.6524,
369
+ "step": 210
370
+ },
371
+ {
372
+ "epoch": 27.0,
373
+ "eval_accuracy": 0.5887850467289719,
374
+ "eval_loss": 0.7224901914596558,
375
+ "eval_runtime": 1.881,
376
+ "eval_samples_per_second": 56.886,
377
+ "eval_steps_per_second": 3.722,
378
+ "step": 216
379
+ },
380
+ {
381
+ "epoch": 27.5,
382
+ "learning_rate": 5.208333333333333e-05,
383
+ "loss": 0.6567,
384
+ "step": 220
385
+ },
386
+ {
387
+ "epoch": 28.0,
388
+ "eval_accuracy": 0.5887850467289719,
389
+ "eval_loss": 0.7117226719856262,
390
+ "eval_runtime": 1.9114,
391
+ "eval_samples_per_second": 55.979,
392
+ "eval_steps_per_second": 3.662,
393
+ "step": 224
394
+ },
395
+ {
396
+ "epoch": 28.75,
397
+ "learning_rate": 4.6874999999999994e-05,
398
+ "loss": 0.6402,
399
+ "step": 230
400
+ },
401
+ {
402
+ "epoch": 29.0,
403
+ "eval_accuracy": 0.6261682242990654,
404
+ "eval_loss": 0.6999447345733643,
405
+ "eval_runtime": 1.913,
406
+ "eval_samples_per_second": 55.934,
407
+ "eval_steps_per_second": 3.659,
408
+ "step": 232
409
+ },
410
+ {
411
+ "epoch": 30.0,
412
+ "learning_rate": 4.1666666666666665e-05,
413
+ "loss": 0.66,
414
+ "step": 240
415
+ },
416
+ {
417
+ "epoch": 30.0,
418
+ "eval_accuracy": 0.6074766355140186,
419
+ "eval_loss": 0.6799367070198059,
420
+ "eval_runtime": 1.8619,
421
+ "eval_samples_per_second": 57.467,
422
+ "eval_steps_per_second": 3.76,
423
+ "step": 240
424
+ },
425
+ {
426
+ "epoch": 31.0,
427
+ "eval_accuracy": 0.6074766355140186,
428
+ "eval_loss": 0.6676725745201111,
429
+ "eval_runtime": 1.8574,
430
+ "eval_samples_per_second": 57.606,
431
+ "eval_steps_per_second": 3.769,
432
+ "step": 248
433
+ },
434
+ {
435
+ "epoch": 31.25,
436
+ "learning_rate": 3.645833333333333e-05,
437
+ "loss": 0.6469,
438
+ "step": 250
439
+ },
440
+ {
441
+ "epoch": 32.0,
442
+ "eval_accuracy": 0.5981308411214953,
443
+ "eval_loss": 0.6734670996665955,
444
+ "eval_runtime": 1.8664,
445
+ "eval_samples_per_second": 57.329,
446
+ "eval_steps_per_second": 3.75,
447
+ "step": 256
448
+ },
449
+ {
450
+ "epoch": 32.5,
451
+ "learning_rate": 3.125e-05,
452
+ "loss": 0.6355,
453
+ "step": 260
454
+ },
455
+ {
456
+ "epoch": 33.0,
457
+ "eval_accuracy": 0.616822429906542,
458
+ "eval_loss": 0.6852976083755493,
459
+ "eval_runtime": 2.0311,
460
+ "eval_samples_per_second": 52.681,
461
+ "eval_steps_per_second": 3.446,
462
+ "step": 264
463
+ },
464
+ {
465
+ "epoch": 33.75,
466
+ "learning_rate": 2.6041666666666665e-05,
467
+ "loss": 0.6245,
468
+ "step": 270
469
+ },
470
+ {
471
+ "epoch": 34.0,
472
+ "eval_accuracy": 0.6261682242990654,
473
+ "eval_loss": 0.7008370161056519,
474
+ "eval_runtime": 1.8744,
475
+ "eval_samples_per_second": 57.084,
476
+ "eval_steps_per_second": 3.734,
477
+ "step": 272
478
+ },
479
+ {
480
+ "epoch": 35.0,
481
+ "learning_rate": 2.0833333333333333e-05,
482
+ "loss": 0.6306,
483
+ "step": 280
484
+ },
485
+ {
486
+ "epoch": 35.0,
487
+ "eval_accuracy": 0.5981308411214953,
488
+ "eval_loss": 0.6990097761154175,
489
+ "eval_runtime": 1.9194,
490
+ "eval_samples_per_second": 55.745,
491
+ "eval_steps_per_second": 3.647,
492
+ "step": 280
493
+ },
494
+ {
495
+ "epoch": 36.0,
496
+ "eval_accuracy": 0.6355140186915887,
497
+ "eval_loss": 0.6980569958686829,
498
+ "eval_runtime": 1.9115,
499
+ "eval_samples_per_second": 55.978,
500
+ "eval_steps_per_second": 3.662,
501
+ "step": 288
502
+ },
503
+ {
504
+ "epoch": 36.25,
505
+ "learning_rate": 1.5625e-05,
506
+ "loss": 0.6208,
507
+ "step": 290
508
+ },
509
+ {
510
+ "epoch": 37.0,
511
+ "eval_accuracy": 0.6261682242990654,
512
+ "eval_loss": 0.7103211879730225,
513
+ "eval_runtime": 1.9169,
514
+ "eval_samples_per_second": 55.818,
515
+ "eval_steps_per_second": 3.652,
516
+ "step": 296
517
+ },
518
+ {
519
+ "epoch": 37.5,
520
+ "learning_rate": 1.0416666666666666e-05,
521
+ "loss": 0.6339,
522
+ "step": 300
523
+ },
524
+ {
525
+ "epoch": 38.0,
526
+ "eval_accuracy": 0.6355140186915887,
527
+ "eval_loss": 0.7050479054450989,
528
+ "eval_runtime": 1.9345,
529
+ "eval_samples_per_second": 55.312,
530
+ "eval_steps_per_second": 3.619,
531
+ "step": 304
532
+ },
533
+ {
534
+ "epoch": 38.75,
535
+ "learning_rate": 5.208333333333333e-06,
536
+ "loss": 0.5959,
537
+ "step": 310
538
+ },
539
+ {
540
+ "epoch": 39.0,
541
+ "eval_accuracy": 0.6355140186915887,
542
+ "eval_loss": 0.698934018611908,
543
+ "eval_runtime": 1.8634,
544
+ "eval_samples_per_second": 57.421,
545
+ "eval_steps_per_second": 3.757,
546
+ "step": 312
547
+ },
548
+ {
549
+ "epoch": 40.0,
550
+ "learning_rate": 0.0,
551
+ "loss": 0.6059,
552
+ "step": 320
553
+ },
554
+ {
555
+ "epoch": 40.0,
556
+ "eval_accuracy": 0.6355140186915887,
557
+ "eval_loss": 0.6989688277244568,
558
+ "eval_runtime": 1.8409,
559
+ "eval_samples_per_second": 58.123,
560
+ "eval_steps_per_second": 3.802,
561
+ "step": 320
562
+ },
563
+ {
564
+ "epoch": 40.0,
565
+ "step": 320,
566
+ "total_flos": 6.480838238876467e+17,
567
+ "train_loss": 0.9213116973638534,
568
+ "train_runtime": 581.2908,
569
+ "train_samples_per_second": 34.269,
570
+ "train_steps_per_second": 0.55
571
+ }
572
+ ],
573
+ "logging_steps": 10,
574
+ "max_steps": 320,
575
+ "num_input_tokens_seen": 0,
576
+ "num_train_epochs": 40,
577
+ "save_steps": 500,
578
+ "total_flos": 6.480838238876467e+17,
579
+ "train_batch_size": 16,
580
+ "trial_name": null,
581
+ "trial_params": null
582
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1eb3ae6da4bd4826a72ebe0f4e937ad3b5d737b9c90b89682ccd8f7234d78ea3
3
+ size 4728