End of training
Browse files- README.md +30 -20
- all_results.json +10 -10
- eval_results.json +6 -6
- model.safetensors +1 -1
- runs/Dec28_00-52-56_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703721182.MacBook-Pro-de-Max-2.local.31343.9 +3 -0
- runs/Dec28_00-52-56_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703723525.MacBook-Pro-de-Max-2.local.31343.10 +3 -0
- train_results.json +5 -5
- trainer_state.json +303 -129
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -22,7 +22,7 @@ model-index:
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
-
value: 0.
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -32,12 +32,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 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.
|
| 36 |
-
- Accuracy: 0.
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
| 40 |
-
|
| 41 |
|
| 42 |
## Intended uses & limitations
|
| 43 |
|
|
@@ -61,27 +61,37 @@ The following hyperparameters were used during training:
|
|
| 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:
|
| 65 |
|
| 66 |
### Training results
|
| 67 |
|
| 68 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
-
| 1.
|
| 71 |
-
| 1.
|
| 72 |
-
| 1.
|
| 73 |
-
| 1.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
|
| 86 |
|
| 87 |
### Framework versions
|
|
|
|
| 22 |
metrics:
|
| 23 |
- name: Accuracy
|
| 24 |
type: accuracy
|
| 25 |
+
value: 0.8389261744966443
|
| 26 |
---
|
| 27 |
|
| 28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 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: 0.5312
|
| 36 |
+
- Accuracy: 0.8389
|
| 37 |
|
| 38 |
## Model description
|
| 39 |
|
| 40 |
+
More information needed
|
| 41 |
|
| 42 |
## Intended uses & limitations
|
| 43 |
|
|
|
|
| 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: 25
|
| 65 |
|
| 66 |
### Training results
|
| 67 |
|
| 68 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 69 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 70 |
+
| 1.6068 | 0.97 | 14 | 1.5809 | 0.5415 |
|
| 71 |
+
| 1.56 | 2.0 | 29 | 1.2830 | 0.5415 |
|
| 72 |
+
| 1.1852 | 2.97 | 43 | 1.0794 | 0.5415 |
|
| 73 |
+
| 1.1132 | 4.0 | 58 | 0.9314 | 0.6488 |
|
| 74 |
+
| 0.9416 | 4.97 | 72 | 0.8935 | 0.6341 |
|
| 75 |
+
| 0.9143 | 6.0 | 87 | 0.8009 | 0.6829 |
|
| 76 |
+
| 0.8243 | 6.97 | 101 | 0.8067 | 0.6634 |
|
| 77 |
+
| 0.8171 | 8.0 | 116 | 0.7783 | 0.6780 |
|
| 78 |
+
| 0.7901 | 8.97 | 130 | 0.7871 | 0.6585 |
|
| 79 |
+
| 0.7944 | 10.0 | 145 | 0.7414 | 0.6976 |
|
| 80 |
+
| 0.7669 | 10.97 | 159 | 0.6977 | 0.7122 |
|
| 81 |
+
| 0.7478 | 12.0 | 174 | 0.7043 | 0.7122 |
|
| 82 |
+
| 0.766 | 12.97 | 188 | 0.7778 | 0.6585 |
|
| 83 |
+
| 0.7322 | 14.0 | 203 | 0.7504 | 0.6780 |
|
| 84 |
+
| 0.7242 | 14.97 | 217 | 0.7291 | 0.6829 |
|
| 85 |
+
| 0.7554 | 16.0 | 232 | 0.7694 | 0.6634 |
|
| 86 |
+
| 0.7422 | 16.97 | 246 | 0.7569 | 0.6829 |
|
| 87 |
+
| 0.7292 | 18.0 | 261 | 0.7389 | 0.6780 |
|
| 88 |
+
| 0.7354 | 18.97 | 275 | 0.6684 | 0.7122 |
|
| 89 |
+
| 0.6847 | 20.0 | 290 | 0.6821 | 0.7122 |
|
| 90 |
+
| 0.7231 | 20.97 | 304 | 0.6839 | 0.7024 |
|
| 91 |
+
| 0.6962 | 22.0 | 319 | 0.6958 | 0.6878 |
|
| 92 |
+
| 0.7079 | 22.97 | 333 | 0.7039 | 0.6878 |
|
| 93 |
+
| 0.7088 | 24.0 | 348 | 0.6974 | 0.6878 |
|
| 94 |
+
| 0.7106 | 24.14 | 350 | 0.6975 | 0.6878 |
|
| 95 |
|
| 96 |
|
| 97 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 24.14,
|
| 3 |
+
"eval_accuracy": 0.8389261744966443,
|
| 4 |
+
"eval_loss": 0.5312080979347229,
|
| 5 |
+
"eval_runtime": 6.5357,
|
| 6 |
+
"eval_samples_per_second": 45.596,
|
| 7 |
+
"eval_steps_per_second": 1.53,
|
| 8 |
+
"train_loss": 0.8573012270246233,
|
| 9 |
+
"train_runtime": 2334.0273,
|
| 10 |
+
"train_samples_per_second": 19.698,
|
| 11 |
+
"train_steps_per_second": 0.15
|
| 12 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 24.14,
|
| 3 |
+
"eval_accuracy": 0.8389261744966443,
|
| 4 |
+
"eval_loss": 0.5312080979347229,
|
| 5 |
+
"eval_runtime": 6.5357,
|
| 6 |
+
"eval_samples_per_second": 45.596,
|
| 7 |
+
"eval_steps_per_second": 1.53
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 110359372
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3efa52dfd3f432fcd7e64448044d880f0ab3510db098be6c7896a1dadb4c6fd3
|
| 3 |
size 110359372
|
runs/Dec28_00-52-56_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703721182.MacBook-Pro-de-Max-2.local.31343.9
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ec33d0976e9fa7bb349d40c55b93cb4f7052e406b8a42eb4765391a2702541e
|
| 3 |
+
size 18671
|
runs/Dec28_00-52-56_MacBook-Pro-de-Max-2.local/events.out.tfevents.1703723525.MacBook-Pro-de-Max-2.local.31343.10
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29da514cabfbb21c5dcf1d41eef3eb7d18dfd601dbc797059b7dcc153dc93f68
|
| 3 |
+
size 734
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 24.14,
|
| 3 |
+
"train_loss": 0.8573012270246233,
|
| 4 |
+
"train_runtime": 2334.0273,
|
| 5 |
+
"train_samples_per_second": 19.698,
|
| 6 |
+
"train_steps_per_second": 0.15
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,290 +1,464 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-exp-max/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.69,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 1.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.97,
|
| 19 |
"eval_accuracy": 0.5414634146341464,
|
| 20 |
-
"eval_loss": 1.
|
| 21 |
-
"eval_runtime": 3.
|
| 22 |
-
"eval_samples_per_second": 58.
|
| 23 |
-
"eval_steps_per_second":
|
| 24 |
"step": 14
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 1.38,
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 20
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 2.0,
|
| 34 |
"eval_accuracy": 0.5414634146341464,
|
| 35 |
-
"eval_loss": 1.
|
| 36 |
-
"eval_runtime": 3.
|
| 37 |
-
"eval_samples_per_second": 58.
|
| 38 |
-
"eval_steps_per_second":
|
| 39 |
"step": 29
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.07,
|
| 43 |
-
"learning_rate": 4.
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.76,
|
| 49 |
-
"learning_rate": 4.
|
| 50 |
-
"loss": 1.
|
| 51 |
"step": 40
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 2.97,
|
| 55 |
-
"eval_accuracy": 0.
|
| 56 |
-
"eval_loss": 1.
|
| 57 |
-
"eval_runtime": 3.
|
| 58 |
-
"eval_samples_per_second":
|
| 59 |
-
"eval_steps_per_second":
|
| 60 |
"step": 43
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 3.45,
|
| 64 |
-
"learning_rate": 4.
|
| 65 |
-
"loss": 1.
|
| 66 |
"step": 50
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 4.0,
|
| 70 |
-
"eval_accuracy": 0.
|
| 71 |
-
"eval_loss": 0.
|
| 72 |
-
"eval_runtime": 3.
|
| 73 |
-
"eval_samples_per_second":
|
| 74 |
-
"eval_steps_per_second":
|
| 75 |
"step": 58
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 4.14,
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 1.
|
| 81 |
"step": 60
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 4.83,
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 70
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 4.97,
|
| 91 |
-
"eval_accuracy": 0.
|
| 92 |
-
"eval_loss": 0.
|
| 93 |
-
"eval_runtime": 3.
|
| 94 |
-
"eval_samples_per_second": 58.
|
| 95 |
-
"eval_steps_per_second":
|
| 96 |
"step": 72
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 5.52,
|
| 100 |
-
"learning_rate":
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 80
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 6.0,
|
| 106 |
-
"eval_accuracy": 0.
|
| 107 |
-
"eval_loss": 0.
|
| 108 |
-
"eval_runtime": 3.
|
| 109 |
-
"eval_samples_per_second":
|
| 110 |
-
"eval_steps_per_second":
|
| 111 |
"step": 87
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 6.21,
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 0.
|
| 117 |
"step": 90
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 6.9,
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 100
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 6.97,
|
| 127 |
-
"eval_accuracy": 0.
|
| 128 |
-
"eval_loss": 0.
|
| 129 |
-
"eval_runtime": 3.
|
| 130 |
-
"eval_samples_per_second": 58.
|
| 131 |
-
"eval_steps_per_second":
|
| 132 |
"step": 101
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 7.59,
|
| 136 |
-
"learning_rate":
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 110
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 8.0,
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime": 3.
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second":
|
| 147 |
"step": 116
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 8.28,
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 120
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 8.97,
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 130
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 8.97,
|
| 163 |
-
"eval_accuracy": 0.
|
| 164 |
-
"eval_loss": 0.
|
| 165 |
-
"eval_runtime": 3.
|
| 166 |
-
"eval_samples_per_second":
|
| 167 |
-
"eval_steps_per_second": 2.
|
| 168 |
"step": 130
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 9.66,
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 140
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 10.0,
|
| 178 |
-
"eval_accuracy": 0.
|
| 179 |
-
"eval_loss": 0.
|
| 180 |
-
"eval_runtime": 3.
|
| 181 |
-
"eval_samples_per_second":
|
| 182 |
-
"eval_steps_per_second":
|
| 183 |
"step": 145
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 10.34,
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 150
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 10.97,
|
| 193 |
"eval_accuracy": 0.7121951219512195,
|
| 194 |
-
"eval_loss": 0.
|
| 195 |
-
"eval_runtime": 3.
|
| 196 |
-
"eval_samples_per_second":
|
| 197 |
-
"eval_steps_per_second": 2.
|
| 198 |
"step": 159
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 11.03,
|
| 202 |
-
"learning_rate":
|
| 203 |
-
"loss": 0.
|
| 204 |
"step": 160
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 11.72,
|
| 208 |
-
"learning_rate":
|
| 209 |
-
"loss": 0.
|
| 210 |
"step": 170
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 12.0,
|
| 214 |
-
"eval_accuracy": 0.
|
| 215 |
-
"eval_loss": 0.
|
| 216 |
-
"eval_runtime": 3.
|
| 217 |
-
"eval_samples_per_second":
|
| 218 |
-
"eval_steps_per_second":
|
| 219 |
"step": 174
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 12.41,
|
| 223 |
-
"learning_rate":
|
| 224 |
-
"loss": 0.
|
| 225 |
"step": 180
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 12.97,
|
| 229 |
-
"eval_accuracy": 0.
|
| 230 |
-
"eval_loss": 0.
|
| 231 |
-
"eval_runtime": 3.
|
| 232 |
-
"eval_samples_per_second":
|
| 233 |
-
"eval_steps_per_second":
|
| 234 |
"step": 188
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 13.1,
|
| 238 |
-
"learning_rate":
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 190
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 13.79,
|
| 244 |
-
"learning_rate": 2.
|
| 245 |
-
"loss": 0.
|
| 246 |
"step": 200
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 14.0,
|
| 250 |
-
"eval_accuracy": 0.
|
| 251 |
-
"eval_loss": 0.
|
| 252 |
-
"eval_runtime": 3.
|
| 253 |
-
"eval_samples_per_second":
|
| 254 |
-
"eval_steps_per_second": 2.
|
| 255 |
"step": 203
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 14.48,
|
| 259 |
-
"learning_rate":
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 210
|
| 262 |
},
|
| 263 |
{
|
| 264 |
-
"epoch": 14.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
"eval_accuracy": 0.6878048780487804,
|
| 266 |
-
"eval_loss": 0.
|
| 267 |
-
"eval_runtime": 3.
|
| 268 |
-
"eval_samples_per_second":
|
| 269 |
-
"eval_steps_per_second": 2.
|
| 270 |
-
"step":
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch":
|
| 274 |
-
"
|
| 275 |
-
"
|
| 276 |
-
"
|
| 277 |
-
|
| 278 |
-
|
| 279 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 280 |
}
|
| 281 |
],
|
| 282 |
"logging_steps": 10,
|
| 283 |
-
"max_steps":
|
| 284 |
"num_input_tokens_seen": 0,
|
| 285 |
-
"num_train_epochs":
|
| 286 |
"save_steps": 500,
|
| 287 |
-
"total_flos":
|
| 288 |
"train_batch_size": 32,
|
| 289 |
"trial_name": null,
|
| 290 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7121951219512195,
|
| 3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-exp-max/checkpoint-159",
|
| 4 |
+
"epoch": 24.137931034482758,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 350,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.69,
|
| 13 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 14 |
+
"loss": 1.6068,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.97,
|
| 19 |
"eval_accuracy": 0.5414634146341464,
|
| 20 |
+
"eval_loss": 1.5808794498443604,
|
| 21 |
+
"eval_runtime": 3.5265,
|
| 22 |
+
"eval_samples_per_second": 58.131,
|
| 23 |
+
"eval_steps_per_second": 1.985,
|
| 24 |
"step": 14
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 1.38,
|
| 28 |
+
"learning_rate": 2.857142857142857e-05,
|
| 29 |
+
"loss": 1.56,
|
| 30 |
"step": 20
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 2.0,
|
| 34 |
"eval_accuracy": 0.5414634146341464,
|
| 35 |
+
"eval_loss": 1.2830290794372559,
|
| 36 |
+
"eval_runtime": 3.4914,
|
| 37 |
+
"eval_samples_per_second": 58.716,
|
| 38 |
+
"eval_steps_per_second": 2.005,
|
| 39 |
"step": 29
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.07,
|
| 43 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 44 |
+
"loss": 1.3595,
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 2.76,
|
| 49 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 50 |
+
"loss": 1.1852,
|
| 51 |
"step": 40
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 2.97,
|
| 55 |
+
"eval_accuracy": 0.5414634146341464,
|
| 56 |
+
"eval_loss": 1.0793886184692383,
|
| 57 |
+
"eval_runtime": 3.5623,
|
| 58 |
+
"eval_samples_per_second": 57.547,
|
| 59 |
+
"eval_steps_per_second": 1.965,
|
| 60 |
"step": 43
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 3.45,
|
| 64 |
+
"learning_rate": 4.761904761904762e-05,
|
| 65 |
+
"loss": 1.1132,
|
| 66 |
"step": 50
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 4.0,
|
| 70 |
+
"eval_accuracy": 0.6487804878048781,
|
| 71 |
+
"eval_loss": 0.9314356446266174,
|
| 72 |
+
"eval_runtime": 3.4589,
|
| 73 |
+
"eval_samples_per_second": 59.268,
|
| 74 |
+
"eval_steps_per_second": 2.024,
|
| 75 |
"step": 58
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 4.14,
|
| 79 |
+
"learning_rate": 4.603174603174603e-05,
|
| 80 |
+
"loss": 1.024,
|
| 81 |
"step": 60
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 4.83,
|
| 85 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 86 |
+
"loss": 0.9416,
|
| 87 |
"step": 70
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 4.97,
|
| 91 |
+
"eval_accuracy": 0.6341463414634146,
|
| 92 |
+
"eval_loss": 0.8935254216194153,
|
| 93 |
+
"eval_runtime": 3.4888,
|
| 94 |
+
"eval_samples_per_second": 58.759,
|
| 95 |
+
"eval_steps_per_second": 2.006,
|
| 96 |
"step": 72
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 5.52,
|
| 100 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 101 |
+
"loss": 0.9143,
|
| 102 |
"step": 80
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 6.0,
|
| 106 |
+
"eval_accuracy": 0.6829268292682927,
|
| 107 |
+
"eval_loss": 0.8009229898452759,
|
| 108 |
+
"eval_runtime": 3.465,
|
| 109 |
+
"eval_samples_per_second": 59.163,
|
| 110 |
+
"eval_steps_per_second": 2.02,
|
| 111 |
"step": 87
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 6.21,
|
| 115 |
+
"learning_rate": 4.126984126984127e-05,
|
| 116 |
+
"loss": 0.8868,
|
| 117 |
"step": 90
|
| 118 |
},
|
| 119 |
{
|
| 120 |
"epoch": 6.9,
|
| 121 |
+
"learning_rate": 3.968253968253968e-05,
|
| 122 |
+
"loss": 0.8243,
|
| 123 |
"step": 100
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 6.97,
|
| 127 |
+
"eval_accuracy": 0.6634146341463415,
|
| 128 |
+
"eval_loss": 0.8067137002944946,
|
| 129 |
+
"eval_runtime": 3.5324,
|
| 130 |
+
"eval_samples_per_second": 58.034,
|
| 131 |
+
"eval_steps_per_second": 1.982,
|
| 132 |
"step": 101
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 7.59,
|
| 136 |
+
"learning_rate": 3.809523809523809e-05,
|
| 137 |
+
"loss": 0.8171,
|
| 138 |
"step": 110
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 8.0,
|
| 142 |
+
"eval_accuracy": 0.6780487804878049,
|
| 143 |
+
"eval_loss": 0.7783121466636658,
|
| 144 |
+
"eval_runtime": 3.4599,
|
| 145 |
+
"eval_samples_per_second": 59.25,
|
| 146 |
+
"eval_steps_per_second": 2.023,
|
| 147 |
"step": 116
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 8.28,
|
| 151 |
+
"learning_rate": 3.650793650793651e-05,
|
| 152 |
+
"loss": 0.8161,
|
| 153 |
"step": 120
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 8.97,
|
| 157 |
+
"learning_rate": 3.492063492063492e-05,
|
| 158 |
+
"loss": 0.7901,
|
| 159 |
"step": 130
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 8.97,
|
| 163 |
+
"eval_accuracy": 0.6585365853658537,
|
| 164 |
+
"eval_loss": 0.7871080636978149,
|
| 165 |
+
"eval_runtime": 3.4561,
|
| 166 |
+
"eval_samples_per_second": 59.316,
|
| 167 |
+
"eval_steps_per_second": 2.025,
|
| 168 |
"step": 130
|
| 169 |
},
|
| 170 |
{
|
| 171 |
"epoch": 9.66,
|
| 172 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 173 |
+
"loss": 0.7944,
|
| 174 |
"step": 140
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 10.0,
|
| 178 |
+
"eval_accuracy": 0.697560975609756,
|
| 179 |
+
"eval_loss": 0.7413551211357117,
|
| 180 |
+
"eval_runtime": 3.4339,
|
| 181 |
+
"eval_samples_per_second": 59.699,
|
| 182 |
+
"eval_steps_per_second": 2.038,
|
| 183 |
"step": 145
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 10.34,
|
| 187 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 188 |
+
"loss": 0.7669,
|
| 189 |
"step": 150
|
| 190 |
},
|
| 191 |
{
|
| 192 |
"epoch": 10.97,
|
| 193 |
"eval_accuracy": 0.7121951219512195,
|
| 194 |
+
"eval_loss": 0.6976904273033142,
|
| 195 |
+
"eval_runtime": 3.4718,
|
| 196 |
+
"eval_samples_per_second": 59.047,
|
| 197 |
+
"eval_steps_per_second": 2.016,
|
| 198 |
"step": 159
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 11.03,
|
| 202 |
+
"learning_rate": 3.0158730158730158e-05,
|
| 203 |
+
"loss": 0.7961,
|
| 204 |
"step": 160
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 11.72,
|
| 208 |
+
"learning_rate": 2.857142857142857e-05,
|
| 209 |
+
"loss": 0.7478,
|
| 210 |
"step": 170
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 12.0,
|
| 214 |
+
"eval_accuracy": 0.7121951219512195,
|
| 215 |
+
"eval_loss": 0.7042645812034607,
|
| 216 |
+
"eval_runtime": 3.4485,
|
| 217 |
+
"eval_samples_per_second": 59.447,
|
| 218 |
+
"eval_steps_per_second": 2.03,
|
| 219 |
"step": 174
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 12.41,
|
| 223 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 224 |
+
"loss": 0.766,
|
| 225 |
"step": 180
|
| 226 |
},
|
| 227 |
{
|
| 228 |
"epoch": 12.97,
|
| 229 |
+
"eval_accuracy": 0.6585365853658537,
|
| 230 |
+
"eval_loss": 0.7778439521789551,
|
| 231 |
+
"eval_runtime": 3.6047,
|
| 232 |
+
"eval_samples_per_second": 56.87,
|
| 233 |
+
"eval_steps_per_second": 1.942,
|
| 234 |
"step": 188
|
| 235 |
},
|
| 236 |
{
|
| 237 |
"epoch": 13.1,
|
| 238 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 239 |
+
"loss": 0.7691,
|
| 240 |
"step": 190
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 13.79,
|
| 244 |
+
"learning_rate": 2.380952380952381e-05,
|
| 245 |
+
"loss": 0.7322,
|
| 246 |
"step": 200
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 14.0,
|
| 250 |
+
"eval_accuracy": 0.6780487804878049,
|
| 251 |
+
"eval_loss": 0.7503620386123657,
|
| 252 |
+
"eval_runtime": 3.4523,
|
| 253 |
+
"eval_samples_per_second": 59.381,
|
| 254 |
+
"eval_steps_per_second": 2.028,
|
| 255 |
"step": 203
|
| 256 |
},
|
| 257 |
{
|
| 258 |
"epoch": 14.48,
|
| 259 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 260 |
+
"loss": 0.7242,
|
| 261 |
"step": 210
|
| 262 |
},
|
| 263 |
{
|
| 264 |
+
"epoch": 14.97,
|
| 265 |
+
"eval_accuracy": 0.6829268292682927,
|
| 266 |
+
"eval_loss": 0.7290918827056885,
|
| 267 |
+
"eval_runtime": 3.5373,
|
| 268 |
+
"eval_samples_per_second": 57.954,
|
| 269 |
+
"eval_steps_per_second": 1.979,
|
| 270 |
+
"step": 217
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 15.17,
|
| 274 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 275 |
+
"loss": 0.7172,
|
| 276 |
+
"step": 220
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 15.86,
|
| 280 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 281 |
+
"loss": 0.7554,
|
| 282 |
+
"step": 230
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 16.0,
|
| 286 |
+
"eval_accuracy": 0.6634146341463415,
|
| 287 |
+
"eval_loss": 0.7694374918937683,
|
| 288 |
+
"eval_runtime": 3.4718,
|
| 289 |
+
"eval_samples_per_second": 59.047,
|
| 290 |
+
"eval_steps_per_second": 2.016,
|
| 291 |
+
"step": 232
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 16.55,
|
| 295 |
+
"learning_rate": 1.746031746031746e-05,
|
| 296 |
+
"loss": 0.7422,
|
| 297 |
+
"step": 240
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 16.97,
|
| 301 |
+
"eval_accuracy": 0.6829268292682927,
|
| 302 |
+
"eval_loss": 0.7568630576133728,
|
| 303 |
+
"eval_runtime": 3.5281,
|
| 304 |
+
"eval_samples_per_second": 58.104,
|
| 305 |
+
"eval_steps_per_second": 1.984,
|
| 306 |
+
"step": 246
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 17.24,
|
| 310 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 311 |
+
"loss": 0.7324,
|
| 312 |
+
"step": 250
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 17.93,
|
| 316 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 317 |
+
"loss": 0.7292,
|
| 318 |
+
"step": 260
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 18.0,
|
| 322 |
+
"eval_accuracy": 0.6780487804878049,
|
| 323 |
+
"eval_loss": 0.7389385104179382,
|
| 324 |
+
"eval_runtime": 3.4574,
|
| 325 |
+
"eval_samples_per_second": 59.292,
|
| 326 |
+
"eval_steps_per_second": 2.025,
|
| 327 |
+
"step": 261
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 18.62,
|
| 331 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 332 |
+
"loss": 0.7354,
|
| 333 |
+
"step": 270
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 18.97,
|
| 337 |
+
"eval_accuracy": 0.7121951219512195,
|
| 338 |
+
"eval_loss": 0.668440580368042,
|
| 339 |
+
"eval_runtime": 3.6943,
|
| 340 |
+
"eval_samples_per_second": 55.491,
|
| 341 |
+
"eval_steps_per_second": 1.895,
|
| 342 |
+
"step": 275
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 19.31,
|
| 346 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 347 |
+
"loss": 0.7274,
|
| 348 |
+
"step": 280
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 20.0,
|
| 352 |
+
"learning_rate": 9.523809523809523e-06,
|
| 353 |
+
"loss": 0.6847,
|
| 354 |
+
"step": 290
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 20.0,
|
| 358 |
+
"eval_accuracy": 0.7121951219512195,
|
| 359 |
+
"eval_loss": 0.6821430325508118,
|
| 360 |
+
"eval_runtime": 3.4768,
|
| 361 |
+
"eval_samples_per_second": 58.962,
|
| 362 |
+
"eval_steps_per_second": 2.013,
|
| 363 |
+
"step": 290
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 20.69,
|
| 367 |
+
"learning_rate": 7.936507936507936e-06,
|
| 368 |
+
"loss": 0.7231,
|
| 369 |
+
"step": 300
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 20.97,
|
| 373 |
+
"eval_accuracy": 0.7024390243902439,
|
| 374 |
+
"eval_loss": 0.6839069128036499,
|
| 375 |
+
"eval_runtime": 3.53,
|
| 376 |
+
"eval_samples_per_second": 58.074,
|
| 377 |
+
"eval_steps_per_second": 1.983,
|
| 378 |
+
"step": 304
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 21.38,
|
| 382 |
+
"learning_rate": 6.349206349206349e-06,
|
| 383 |
+
"loss": 0.6962,
|
| 384 |
+
"step": 310
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 22.0,
|
| 388 |
"eval_accuracy": 0.6878048780487804,
|
| 389 |
+
"eval_loss": 0.6958089470863342,
|
| 390 |
+
"eval_runtime": 3.4679,
|
| 391 |
+
"eval_samples_per_second": 59.113,
|
| 392 |
+
"eval_steps_per_second": 2.018,
|
| 393 |
+
"step": 319
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"epoch": 22.07,
|
| 397 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 398 |
+
"loss": 0.6995,
|
| 399 |
+
"step": 320
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 22.76,
|
| 403 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 404 |
+
"loss": 0.7079,
|
| 405 |
+
"step": 330
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 22.97,
|
| 409 |
+
"eval_accuracy": 0.6878048780487804,
|
| 410 |
+
"eval_loss": 0.7039469480514526,
|
| 411 |
+
"eval_runtime": 3.4654,
|
| 412 |
+
"eval_samples_per_second": 59.156,
|
| 413 |
+
"eval_steps_per_second": 2.02,
|
| 414 |
+
"step": 333
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 23.45,
|
| 418 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 419 |
+
"loss": 0.7088,
|
| 420 |
+
"step": 340
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 24.0,
|
| 424 |
+
"eval_accuracy": 0.6878048780487804,
|
| 425 |
+
"eval_loss": 0.6974486112594604,
|
| 426 |
+
"eval_runtime": 4.2005,
|
| 427 |
+
"eval_samples_per_second": 48.803,
|
| 428 |
+
"eval_steps_per_second": 1.666,
|
| 429 |
+
"step": 348
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 24.14,
|
| 433 |
+
"learning_rate": 0.0,
|
| 434 |
+
"loss": 0.7106,
|
| 435 |
+
"step": 350
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 24.14,
|
| 439 |
+
"eval_accuracy": 0.6878048780487804,
|
| 440 |
+
"eval_loss": 0.6975364089012146,
|
| 441 |
+
"eval_runtime": 3.9957,
|
| 442 |
+
"eval_samples_per_second": 51.305,
|
| 443 |
+
"eval_steps_per_second": 1.752,
|
| 444 |
+
"step": 350
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 24.14,
|
| 448 |
+
"step": 350,
|
| 449 |
+
"total_flos": 1.444384721662378e+18,
|
| 450 |
+
"train_loss": 0.8573012270246233,
|
| 451 |
+
"train_runtime": 2334.0273,
|
| 452 |
+
"train_samples_per_second": 19.698,
|
| 453 |
+
"train_steps_per_second": 0.15
|
| 454 |
}
|
| 455 |
],
|
| 456 |
"logging_steps": 10,
|
| 457 |
+
"max_steps": 350,
|
| 458 |
"num_input_tokens_seen": 0,
|
| 459 |
+
"num_train_epochs": 25,
|
| 460 |
"save_steps": 500,
|
| 461 |
+
"total_flos": 1.444384721662378e+18,
|
| 462 |
"train_batch_size": 32,
|
| 463 |
"trial_name": null,
|
| 464 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4792
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f2b62bbe9a6d50250e169300e55cdbb15de7e0a611ef74fc73949faedf4cc42
|
| 3 |
size 4792
|