initial commit
Browse files- README.md +10 -10
- all_results.json +9 -9
- eval_results.json +5 -5
- model.safetensors +1 -1
- runs/May01_18-09-22_e5fd9b370bfd/events.out.tfevents.1714586963.e5fd9b370bfd.8190.2 +3 -0
- runs/May01_18-27-41_e5fd9b370bfd/events.out.tfevents.1714588064.e5fd9b370bfd.8190.3 +3 -0
- runs/May01_18-28-50_e5fd9b370bfd/events.out.tfevents.1714588131.e5fd9b370bfd.17713.0 +3 -0
- runs/May01_18-28-50_e5fd9b370bfd/events.out.tfevents.1714589250.e5fd9b370bfd.17713.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +247 -247
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-large-patch32-224-in21k](https://huggingface.co/google/vit-large-patch32-224-in21k) on the sartajbhuvaji/Brain-Tumor-Classification dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss:
|
| 37 |
-
- Accuracy: 0.
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
@@ -66,13 +66,13 @@ The following hyperparameters were used during training:
|
|
| 66 |
|
| 67 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.0066 | 3.3333 | 600 | 1.
|
| 75 |
-
| 0.
|
| 76 |
|
| 77 |
|
| 78 |
### Framework versions
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.7741116751269036
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-large-patch32-224-in21k](https://huggingface.co/google/vit-large-patch32-224-in21k) on the sartajbhuvaji/Brain-Tumor-Classification dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.9050
|
| 37 |
+
- Accuracy: 0.7741
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 66 |
|
| 67 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 69 |
+
| 0.352 | 0.5556 | 100 | 1.2267 | 0.6294 |
|
| 70 |
+
| 0.1612 | 1.1111 | 200 | 1.0895 | 0.7538 |
|
| 71 |
+
| 0.0473 | 1.6667 | 300 | 0.9050 | 0.7741 |
|
| 72 |
+
| 0.0525 | 2.2222 | 400 | 1.0663 | 0.7690 |
|
| 73 |
+
| 0.0123 | 2.7778 | 500 | 1.2450 | 0.7462 |
|
| 74 |
+
| 0.0066 | 3.3333 | 600 | 1.1283 | 0.7817 |
|
| 75 |
+
| 0.0126 | 3.8889 | 700 | 1.1717 | 0.7843 |
|
| 76 |
|
| 77 |
|
| 78 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 7.
|
| 8 |
"total_flos": 3.16768696086528e+18,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7741116751269036,
|
| 4 |
+
"eval_loss": 0.9049533605575562,
|
| 5 |
+
"eval_runtime": 6.5562,
|
| 6 |
+
"eval_samples_per_second": 60.096,
|
| 7 |
+
"eval_steps_per_second": 7.626,
|
| 8 |
"total_flos": 3.16768696086528e+18,
|
| 9 |
+
"train_loss": 0.16034429804939362,
|
| 10 |
+
"train_runtime": 1070.1524,
|
| 11 |
+
"train_samples_per_second": 10.727,
|
| 12 |
+
"train_steps_per_second": 0.673
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 7.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7741116751269036,
|
| 4 |
+
"eval_loss": 0.9049533605575562,
|
| 5 |
+
"eval_runtime": 6.5562,
|
| 6 |
+
"eval_samples_per_second": 60.096,
|
| 7 |
+
"eval_steps_per_second": 7.626
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1222104568
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fe26c731c183216b927e1ce0b6f662aab1cefb27ec9b8a9be42fd07a7c6eb23
|
| 3 |
size 1222104568
|
runs/May01_18-09-22_e5fd9b370bfd/events.out.tfevents.1714586963.e5fd9b370bfd.8190.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbf87b52d8a80ef0a15dba296f126c9dd2b3831e3da9ca245e22305477f7356a
|
| 3 |
+
size 4810
|
runs/May01_18-27-41_e5fd9b370bfd/events.out.tfevents.1714588064.e5fd9b370bfd.8190.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6aa0d6375f444aca8fb66acfe45a77dfcb2aad5239e72af858e31655f0d83a5a
|
| 3 |
+
size 8280
|
runs/May01_18-28-50_e5fd9b370bfd/events.out.tfevents.1714588131.e5fd9b370bfd.17713.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86a630185358ab232804b9446b10a6182b03da53b403c346476bfdef71c754d8
|
| 3 |
+
size 22564
|
runs/May01_18-28-50_e5fd9b370bfd/events.out.tfevents.1714589250.e5fd9b370bfd.17713.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73d9f02977269441d4a3178c448d5e1fc91fa22705f7b1199e2403267f68ab98
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
"total_flos": 3.16768696086528e+18,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
"total_flos": 3.16768696086528e+18,
|
| 4 |
+
"train_loss": 0.16034429804939362,
|
| 5 |
+
"train_runtime": 1070.1524,
|
| 6 |
+
"train_samples_per_second": 10.727,
|
| 7 |
+
"train_steps_per_second": 0.673
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./vit-large-brain-xray/checkpoint-
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
"global_step": 720,
|
|
@@ -10,579 +10,579 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05555555555555555,
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 0.00019722222222222225,
|
| 15 |
-
"loss": 1.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1111111111111111,
|
| 20 |
-
"grad_norm": 1.
|
| 21 |
"learning_rate": 0.00019444444444444446,
|
| 22 |
-
"loss":
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.16666666666666666,
|
| 27 |
-
"grad_norm": 3.
|
| 28 |
"learning_rate": 0.00019166666666666667,
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2222222222222222,
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 0.00018888888888888888,
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.2777777777777778,
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 0.00018611111111111112,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.3333333333333333,
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 0.00018333333333333334,
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.3888888888888889,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 0.00018055555555555557,
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.4444444444444444,
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 0.00017777777777777779,
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.5,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.5555555555555556,
|
| 76 |
-
"grad_norm":
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 0.
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5555555555555556,
|
| 83 |
-
"eval_accuracy": 0.
|
| 84 |
-
"eval_loss": 1.
|
| 85 |
-
"eval_runtime":
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second": 6.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.6111111111111112,
|
| 92 |
-
"grad_norm": 1.
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.6666666666666666,
|
| 99 |
-
"grad_norm": 0.
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.7222222222222222,
|
| 106 |
-
"grad_norm":
|
| 107 |
-
"learning_rate": 0.
|
| 108 |
-
"loss": 0.
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.7777777777777778,
|
| 113 |
-
"grad_norm": 0.
|
| 114 |
-
"learning_rate": 0.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.8333333333333334,
|
| 120 |
-
"grad_norm":
|
| 121 |
-
"learning_rate": 0.
|
| 122 |
-
"loss": 0.
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.8888888888888888,
|
| 127 |
-
"grad_norm":
|
| 128 |
-
"learning_rate": 0.
|
| 129 |
-
"loss": 0.
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.9444444444444444,
|
| 134 |
-
"grad_norm":
|
| 135 |
-
"learning_rate": 0.
|
| 136 |
-
"loss": 0.
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.0,
|
| 141 |
-
"grad_norm": 0.
|
| 142 |
-
"learning_rate": 0.
|
| 143 |
-
"loss": 0.
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.0555555555555556,
|
| 148 |
-
"grad_norm": 0.
|
| 149 |
-
"learning_rate": 0.
|
| 150 |
-
"loss": 0.
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 1.1111111111111112,
|
| 155 |
-
"grad_norm":
|
| 156 |
-
"learning_rate": 0.
|
| 157 |
-
"loss": 0.
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 1.1111111111111112,
|
| 162 |
-
"eval_accuracy": 0.
|
| 163 |
-
"eval_loss": 1.
|
| 164 |
-
"eval_runtime":
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second":
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.1666666666666667,
|
| 171 |
-
"grad_norm":
|
| 172 |
-
"learning_rate": 0.
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.2222222222222223,
|
| 178 |
-
"grad_norm": 0.
|
| 179 |
-
"learning_rate": 0.
|
| 180 |
-
"loss": 0.
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 1.2777777777777777,
|
| 185 |
-
"grad_norm":
|
| 186 |
-
"learning_rate": 0.
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 1.3333333333333333,
|
| 192 |
-
"grad_norm":
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 1.3888888888888888,
|
| 199 |
-
"grad_norm":
|
| 200 |
-
"learning_rate": 0.
|
| 201 |
-
"loss": 0.
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 1.4444444444444444,
|
| 206 |
-
"grad_norm": 0.
|
| 207 |
-
"learning_rate": 0.
|
| 208 |
-
"loss": 0.
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 1.5,
|
| 213 |
-
"grad_norm": 3.
|
| 214 |
-
"learning_rate": 0.
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.5555555555555556,
|
| 220 |
-
"grad_norm": 0.
|
| 221 |
-
"learning_rate": 0.
|
| 222 |
-
"loss": 0.
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.6111111111111112,
|
| 227 |
-
"grad_norm":
|
| 228 |
-
"learning_rate": 0.
|
| 229 |
-
"loss": 0.
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.6666666666666665,
|
| 234 |
-
"grad_norm":
|
| 235 |
-
"learning_rate": 0.
|
| 236 |
-
"loss": 0.
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.6666666666666665,
|
| 241 |
-
"eval_accuracy": 0.
|
| 242 |
-
"eval_loss":
|
| 243 |
-
"eval_runtime": 6.
|
| 244 |
-
"eval_samples_per_second":
|
| 245 |
-
"eval_steps_per_second":
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.7222222222222223,
|
| 250 |
-
"grad_norm":
|
| 251 |
-
"learning_rate": 0.
|
| 252 |
-
"loss": 0.
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.7777777777777777,
|
| 257 |
-
"grad_norm": 0.
|
| 258 |
-
"learning_rate": 0.
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.8333333333333335,
|
| 264 |
-
"grad_norm": 0.
|
| 265 |
-
"learning_rate": 0.
|
| 266 |
-
"loss": 0.
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.8888888888888888,
|
| 271 |
-
"grad_norm":
|
| 272 |
-
"learning_rate": 0.
|
| 273 |
-
"loss": 0.
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.9444444444444444,
|
| 278 |
-
"grad_norm":
|
| 279 |
-
"learning_rate": 0.
|
| 280 |
-
"loss": 0.
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 2.0,
|
| 285 |
-
"grad_norm":
|
| 286 |
-
"learning_rate": 0.
|
| 287 |
-
"loss": 0.
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 2.0555555555555554,
|
| 292 |
-
"grad_norm": 1.
|
| 293 |
-
"learning_rate": 9.
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 2.111111111111111,
|
| 299 |
-
"grad_norm": 0.
|
| 300 |
-
"learning_rate": 9.
|
| 301 |
-
"loss": 0.
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 2.1666666666666665,
|
| 306 |
-
"grad_norm":
|
| 307 |
-
"learning_rate": 9.
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 2.2222222222222223,
|
| 313 |
-
"grad_norm": 0.
|
| 314 |
-
"learning_rate": 8.
|
| 315 |
-
"loss": 0.
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 2.2222222222222223,
|
| 320 |
-
"eval_accuracy": 0.
|
| 321 |
-
"eval_loss": 1.
|
| 322 |
-
"eval_runtime": 6.
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 2.2777777777777777,
|
| 329 |
-
"grad_norm": 0.
|
| 330 |
-
"learning_rate": 8.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 2.3333333333333335,
|
| 336 |
-
"grad_norm": 0.
|
| 337 |
-
"learning_rate": 8.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.388888888888889,
|
| 343 |
-
"grad_norm": 0.
|
| 344 |
-
"learning_rate": 8.
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 2.4444444444444446,
|
| 350 |
-
"grad_norm": 0.
|
| 351 |
-
"learning_rate": 7.
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 2.5,
|
| 357 |
-
"grad_norm":
|
| 358 |
-
"learning_rate": 7.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 2.5555555555555554,
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
-
"learning_rate": 7.
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 2.611111111111111,
|
| 371 |
-
"grad_norm":
|
| 372 |
-
"learning_rate": 6.
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 2.6666666666666665,
|
| 378 |
-
"grad_norm": 0.
|
| 379 |
-
"learning_rate": 6.
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.7222222222222223,
|
| 385 |
-
"grad_norm": 0.
|
| 386 |
-
"learning_rate": 6.
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 2.7777777777777777,
|
| 392 |
-
"grad_norm": 0.
|
| 393 |
-
"learning_rate": 6.
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 2.7777777777777777,
|
| 399 |
-
"eval_accuracy": 0.
|
| 400 |
-
"eval_loss": 1.
|
| 401 |
-
"eval_runtime":
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second":
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 2.8333333333333335,
|
| 408 |
-
"grad_norm":
|
| 409 |
-
"learning_rate": 5.
|
| 410 |
-
"loss": 0.
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.888888888888889,
|
| 415 |
-
"grad_norm": 0.
|
| 416 |
-
"learning_rate": 5.
|
| 417 |
-
"loss": 0.
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 2.9444444444444446,
|
| 422 |
-
"grad_norm": 0.
|
| 423 |
-
"learning_rate": 5.
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 3.0,
|
| 429 |
-
"grad_norm": 0.
|
| 430 |
-
"learning_rate":
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 3.0555555555555554,
|
| 436 |
-
"grad_norm": 0.
|
| 437 |
-
"learning_rate": 4.
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 3.111111111111111,
|
| 443 |
-
"grad_norm":
|
| 444 |
-
"learning_rate": 4.
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 3.1666666666666665,
|
| 450 |
-
"grad_norm": 0.
|
| 451 |
-
"learning_rate": 4.
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 3.2222222222222223,
|
| 457 |
-
"grad_norm": 0.
|
| 458 |
-
"learning_rate": 3.
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 3.2777777777777777,
|
| 464 |
-
"grad_norm": 0.
|
| 465 |
-
"learning_rate": 3.
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 3.3333333333333335,
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
-
"learning_rate": 3.
|
| 473 |
"loss": 0.0066,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 3.3333333333333335,
|
| 478 |
-
"eval_accuracy": 0.
|
| 479 |
-
"eval_loss": 1.
|
| 480 |
-
"eval_runtime": 6.
|
| 481 |
-
"eval_samples_per_second":
|
| 482 |
-
"eval_steps_per_second": 7.
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 3.388888888888889,
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
-
"learning_rate": 3.
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 3.4444444444444446,
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate": 2.
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 3.5,
|
| 501 |
-
"grad_norm": 0.
|
| 502 |
-
"learning_rate": 2.
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 3.5555555555555554,
|
| 508 |
-
"grad_norm": 0.
|
| 509 |
-
"learning_rate": 2.
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 3.611111111111111,
|
| 515 |
-
"grad_norm": 0.
|
| 516 |
-
"learning_rate": 1.
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 3.6666666666666665,
|
| 522 |
-
"grad_norm": 0.
|
| 523 |
-
"learning_rate": 1.
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 3.7222222222222223,
|
| 529 |
-
"grad_norm": 0.
|
| 530 |
-
"learning_rate": 1.
|
| 531 |
"loss": 0.0062,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 3.7777777777777777,
|
| 536 |
-
"grad_norm": 0.
|
| 537 |
-
"learning_rate": 1.
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 3.8333333333333335,
|
| 543 |
-
"grad_norm": 0.
|
| 544 |
-
"learning_rate": 8.
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.888888888888889,
|
| 550 |
-
"grad_norm":
|
| 551 |
-
"learning_rate": 5.
|
| 552 |
-
"loss": 0.
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 3.888888888888889,
|
| 557 |
-
"eval_accuracy": 0.
|
| 558 |
-
"eval_loss": 1.
|
| 559 |
-
"eval_runtime":
|
| 560 |
-
"eval_samples_per_second":
|
| 561 |
-
"eval_steps_per_second":
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 3.9444444444444446,
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
-
"learning_rate":
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 4.0,
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
-
"learning_rate":
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 4.0,
|
| 580 |
"step": 720,
|
| 581 |
"total_flos": 3.16768696086528e+18,
|
| 582 |
-
"train_loss": 0.
|
| 583 |
-
"train_runtime":
|
| 584 |
-
"train_samples_per_second":
|
| 585 |
-
"train_steps_per_second": 0.
|
| 586 |
}
|
| 587 |
],
|
| 588 |
"logging_steps": 10,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9049533605575562,
|
| 3 |
+
"best_model_checkpoint": "./vit-large-brain-xray/checkpoint-300",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
"global_step": 720,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05555555555555555,
|
| 13 |
+
"grad_norm": 0.6702606678009033,
|
| 14 |
"learning_rate": 0.00019722222222222225,
|
| 15 |
+
"loss": 1.3662,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1111111111111111,
|
| 20 |
+
"grad_norm": 1.5160739421844482,
|
| 21 |
"learning_rate": 0.00019444444444444446,
|
| 22 |
+
"loss": 1.2341,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.16666666666666666,
|
| 27 |
+
"grad_norm": 3.668222665786743,
|
| 28 |
"learning_rate": 0.00019166666666666667,
|
| 29 |
+
"loss": 0.9384,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2222222222222222,
|
| 34 |
+
"grad_norm": 2.245746612548828,
|
| 35 |
"learning_rate": 0.00018888888888888888,
|
| 36 |
+
"loss": 0.6538,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.2777777777777778,
|
| 41 |
+
"grad_norm": 2.4727349281311035,
|
| 42 |
"learning_rate": 0.00018611111111111112,
|
| 43 |
+
"loss": 0.4997,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.3333333333333333,
|
| 48 |
+
"grad_norm": 3.7319023609161377,
|
| 49 |
"learning_rate": 0.00018333333333333334,
|
| 50 |
+
"loss": 0.3898,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.3888888888888889,
|
| 55 |
+
"grad_norm": 0.6027088165283203,
|
| 56 |
"learning_rate": 0.00018055555555555557,
|
| 57 |
+
"loss": 0.3304,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.4444444444444444,
|
| 62 |
+
"grad_norm": 0.37244492769241333,
|
| 63 |
"learning_rate": 0.00017777777777777779,
|
| 64 |
+
"loss": 0.2543,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.5,
|
| 69 |
+
"grad_norm": 1.6335736513137817,
|
| 70 |
+
"learning_rate": 0.000175,
|
| 71 |
+
"loss": 0.3538,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.5555555555555556,
|
| 76 |
+
"grad_norm": 1.6272715330123901,
|
| 77 |
+
"learning_rate": 0.00017222222222222224,
|
| 78 |
+
"loss": 0.352,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5555555555555556,
|
| 83 |
+
"eval_accuracy": 0.6294416243654822,
|
| 84 |
+
"eval_loss": 1.2266901731491089,
|
| 85 |
+
"eval_runtime": 8.2003,
|
| 86 |
+
"eval_samples_per_second": 48.047,
|
| 87 |
+
"eval_steps_per_second": 6.097,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.6111111111111112,
|
| 92 |
+
"grad_norm": 1.5688170194625854,
|
| 93 |
+
"learning_rate": 0.00016944444444444445,
|
| 94 |
+
"loss": 0.3773,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.6666666666666666,
|
| 99 |
+
"grad_norm": 0.8265367150306702,
|
| 100 |
+
"learning_rate": 0.0001666666666666667,
|
| 101 |
+
"loss": 0.226,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.7222222222222222,
|
| 106 |
+
"grad_norm": 2.6476309299468994,
|
| 107 |
+
"learning_rate": 0.0001638888888888889,
|
| 108 |
+
"loss": 0.3515,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.7777777777777778,
|
| 113 |
+
"grad_norm": 0.8978700637817383,
|
| 114 |
+
"learning_rate": 0.0001611111111111111,
|
| 115 |
+
"loss": 0.2367,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.8333333333333334,
|
| 120 |
+
"grad_norm": 1.1820647716522217,
|
| 121 |
+
"learning_rate": 0.00015833333333333332,
|
| 122 |
+
"loss": 0.3112,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.8888888888888888,
|
| 127 |
+
"grad_norm": 2.701751708984375,
|
| 128 |
+
"learning_rate": 0.00015555555555555556,
|
| 129 |
+
"loss": 0.2741,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.9444444444444444,
|
| 134 |
+
"grad_norm": 1.249694585800171,
|
| 135 |
+
"learning_rate": 0.00015277777777777777,
|
| 136 |
+
"loss": 0.2529,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.0,
|
| 141 |
+
"grad_norm": 0.23323917388916016,
|
| 142 |
+
"learning_rate": 0.00015000000000000001,
|
| 143 |
+
"loss": 0.2239,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.0555555555555556,
|
| 148 |
+
"grad_norm": 0.4952305853366852,
|
| 149 |
+
"learning_rate": 0.00014722222222222223,
|
| 150 |
+
"loss": 0.1749,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 1.1111111111111112,
|
| 155 |
+
"grad_norm": 0.2073395550251007,
|
| 156 |
+
"learning_rate": 0.00014444444444444444,
|
| 157 |
+
"loss": 0.1612,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 1.1111111111111112,
|
| 162 |
+
"eval_accuracy": 0.7538071065989848,
|
| 163 |
+
"eval_loss": 1.0894657373428345,
|
| 164 |
+
"eval_runtime": 7.609,
|
| 165 |
+
"eval_samples_per_second": 51.781,
|
| 166 |
+
"eval_steps_per_second": 6.571,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.1666666666666667,
|
| 171 |
+
"grad_norm": 0.16568297147750854,
|
| 172 |
+
"learning_rate": 0.00014166666666666668,
|
| 173 |
+
"loss": 0.0718,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.2222222222222223,
|
| 178 |
+
"grad_norm": 0.0984596461057663,
|
| 179 |
+
"learning_rate": 0.0001388888888888889,
|
| 180 |
+
"loss": 0.1395,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 1.2777777777777777,
|
| 185 |
+
"grad_norm": 0.6732985973358154,
|
| 186 |
+
"learning_rate": 0.00013611111111111113,
|
| 187 |
+
"loss": 0.1421,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 1.3333333333333333,
|
| 192 |
+
"grad_norm": 2.5877740383148193,
|
| 193 |
+
"learning_rate": 0.00013333333333333334,
|
| 194 |
+
"loss": 0.1118,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 1.3888888888888888,
|
| 199 |
+
"grad_norm": 3.4756956100463867,
|
| 200 |
+
"learning_rate": 0.00013055555555555555,
|
| 201 |
+
"loss": 0.1562,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 1.4444444444444444,
|
| 206 |
+
"grad_norm": 0.09811172634363174,
|
| 207 |
+
"learning_rate": 0.00012777777777777776,
|
| 208 |
+
"loss": 0.1463,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 1.5,
|
| 213 |
+
"grad_norm": 3.1999194622039795,
|
| 214 |
+
"learning_rate": 0.000125,
|
| 215 |
+
"loss": 0.1179,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.5555555555555556,
|
| 220 |
+
"grad_norm": 0.45012134313583374,
|
| 221 |
+
"learning_rate": 0.00012222222222222224,
|
| 222 |
+
"loss": 0.0929,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.6111111111111112,
|
| 227 |
+
"grad_norm": 1.862654209136963,
|
| 228 |
+
"learning_rate": 0.00011944444444444445,
|
| 229 |
+
"loss": 0.186,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.6666666666666665,
|
| 234 |
+
"grad_norm": 1.39608895778656,
|
| 235 |
+
"learning_rate": 0.00011666666666666668,
|
| 236 |
+
"loss": 0.0473,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.6666666666666665,
|
| 241 |
+
"eval_accuracy": 0.7741116751269036,
|
| 242 |
+
"eval_loss": 0.9049533605575562,
|
| 243 |
+
"eval_runtime": 6.1753,
|
| 244 |
+
"eval_samples_per_second": 63.803,
|
| 245 |
+
"eval_steps_per_second": 8.097,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.7222222222222223,
|
| 250 |
+
"grad_norm": 0.0962180569767952,
|
| 251 |
+
"learning_rate": 0.00011388888888888889,
|
| 252 |
+
"loss": 0.0858,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.7777777777777777,
|
| 257 |
+
"grad_norm": 0.08585009723901749,
|
| 258 |
+
"learning_rate": 0.00011111111111111112,
|
| 259 |
+
"loss": 0.0519,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.8333333333333335,
|
| 264 |
+
"grad_norm": 0.06303343176841736,
|
| 265 |
+
"learning_rate": 0.00010833333333333333,
|
| 266 |
+
"loss": 0.0221,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.8888888888888888,
|
| 271 |
+
"grad_norm": 5.68204402923584,
|
| 272 |
+
"learning_rate": 0.00010555555555555557,
|
| 273 |
+
"loss": 0.0589,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.9444444444444444,
|
| 278 |
+
"grad_norm": 5.385427474975586,
|
| 279 |
+
"learning_rate": 0.00010277777777777778,
|
| 280 |
+
"loss": 0.1881,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 2.0,
|
| 285 |
+
"grad_norm": 3.201244831085205,
|
| 286 |
+
"learning_rate": 0.0001,
|
| 287 |
+
"loss": 0.0758,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 2.0555555555555554,
|
| 292 |
+
"grad_norm": 1.2587229013442993,
|
| 293 |
+
"learning_rate": 9.722222222222223e-05,
|
| 294 |
+
"loss": 0.0461,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 2.111111111111111,
|
| 299 |
+
"grad_norm": 0.07729563117027283,
|
| 300 |
+
"learning_rate": 9.444444444444444e-05,
|
| 301 |
+
"loss": 0.0155,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 2.1666666666666665,
|
| 306 |
+
"grad_norm": 5.077848434448242,
|
| 307 |
+
"learning_rate": 9.166666666666667e-05,
|
| 308 |
+
"loss": 0.0724,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 2.2222222222222223,
|
| 313 |
+
"grad_norm": 0.05171338841319084,
|
| 314 |
+
"learning_rate": 8.888888888888889e-05,
|
| 315 |
+
"loss": 0.0525,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 2.2222222222222223,
|
| 320 |
+
"eval_accuracy": 0.7690355329949239,
|
| 321 |
+
"eval_loss": 1.0663037300109863,
|
| 322 |
+
"eval_runtime": 6.6314,
|
| 323 |
+
"eval_samples_per_second": 59.415,
|
| 324 |
+
"eval_steps_per_second": 7.54,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 2.2777777777777777,
|
| 329 |
+
"grad_norm": 0.056168586015701294,
|
| 330 |
+
"learning_rate": 8.611111111111112e-05,
|
| 331 |
+
"loss": 0.1214,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 2.3333333333333335,
|
| 336 |
+
"grad_norm": 0.2696777582168579,
|
| 337 |
+
"learning_rate": 8.333333333333334e-05,
|
| 338 |
+
"loss": 0.0148,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.388888888888889,
|
| 343 |
+
"grad_norm": 0.045138537883758545,
|
| 344 |
+
"learning_rate": 8.055555555555556e-05,
|
| 345 |
+
"loss": 0.0175,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 2.4444444444444446,
|
| 350 |
+
"grad_norm": 0.03756405785679817,
|
| 351 |
+
"learning_rate": 7.777777777777778e-05,
|
| 352 |
+
"loss": 0.0264,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 2.5,
|
| 357 |
+
"grad_norm": 0.17634020745754242,
|
| 358 |
+
"learning_rate": 7.500000000000001e-05,
|
| 359 |
+
"loss": 0.0101,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 2.5555555555555554,
|
| 364 |
+
"grad_norm": 0.037890926003456116,
|
| 365 |
+
"learning_rate": 7.222222222222222e-05,
|
| 366 |
+
"loss": 0.0484,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 2.611111111111111,
|
| 371 |
+
"grad_norm": 4.751524448394775,
|
| 372 |
+
"learning_rate": 6.944444444444444e-05,
|
| 373 |
+
"loss": 0.0525,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 2.6666666666666665,
|
| 378 |
+
"grad_norm": 0.15853020548820496,
|
| 379 |
+
"learning_rate": 6.666666666666667e-05,
|
| 380 |
+
"loss": 0.0719,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.7222222222222223,
|
| 385 |
+
"grad_norm": 0.039081115275621414,
|
| 386 |
+
"learning_rate": 6.388888888888888e-05,
|
| 387 |
+
"loss": 0.0085,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 2.7777777777777777,
|
| 392 |
+
"grad_norm": 0.4480770230293274,
|
| 393 |
+
"learning_rate": 6.111111111111112e-05,
|
| 394 |
+
"loss": 0.0123,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 2.7777777777777777,
|
| 399 |
+
"eval_accuracy": 0.7461928934010152,
|
| 400 |
+
"eval_loss": 1.2449774742126465,
|
| 401 |
+
"eval_runtime": 5.9167,
|
| 402 |
+
"eval_samples_per_second": 66.591,
|
| 403 |
+
"eval_steps_per_second": 8.451,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 2.8333333333333335,
|
| 408 |
+
"grad_norm": 7.5741801261901855,
|
| 409 |
+
"learning_rate": 5.833333333333334e-05,
|
| 410 |
+
"loss": 0.0278,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.888888888888889,
|
| 415 |
+
"grad_norm": 0.1201184019446373,
|
| 416 |
+
"learning_rate": 5.555555555555556e-05,
|
| 417 |
+
"loss": 0.0091,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 2.9444444444444446,
|
| 422 |
+
"grad_norm": 0.032710809260606766,
|
| 423 |
+
"learning_rate": 5.2777777777777784e-05,
|
| 424 |
+
"loss": 0.0077,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 3.0,
|
| 429 |
+
"grad_norm": 0.03236711025238037,
|
| 430 |
+
"learning_rate": 5e-05,
|
| 431 |
+
"loss": 0.0557,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 3.0555555555555554,
|
| 436 |
+
"grad_norm": 0.034722838550806046,
|
| 437 |
+
"learning_rate": 4.722222222222222e-05,
|
| 438 |
+
"loss": 0.0191,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 3.111111111111111,
|
| 443 |
+
"grad_norm": 4.018179416656494,
|
| 444 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 445 |
+
"loss": 0.0176,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 3.1666666666666665,
|
| 450 |
+
"grad_norm": 0.5732712745666504,
|
| 451 |
+
"learning_rate": 4.166666666666667e-05,
|
| 452 |
+
"loss": 0.0087,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 3.2222222222222223,
|
| 457 |
+
"grad_norm": 0.027404414489865303,
|
| 458 |
+
"learning_rate": 3.888888888888889e-05,
|
| 459 |
+
"loss": 0.0079,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 3.2777777777777777,
|
| 464 |
+
"grad_norm": 0.02965979278087616,
|
| 465 |
+
"learning_rate": 3.611111111111111e-05,
|
| 466 |
+
"loss": 0.0068,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 3.3333333333333335,
|
| 471 |
+
"grad_norm": 0.026871565729379654,
|
| 472 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 473 |
"loss": 0.0066,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 3.3333333333333335,
|
| 478 |
+
"eval_accuracy": 0.7817258883248731,
|
| 479 |
+
"eval_loss": 1.1282514333724976,
|
| 480 |
+
"eval_runtime": 6.7045,
|
| 481 |
+
"eval_samples_per_second": 58.767,
|
| 482 |
+
"eval_steps_per_second": 7.458,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 3.388888888888889,
|
| 487 |
+
"grad_norm": 0.03278065472841263,
|
| 488 |
+
"learning_rate": 3.055555555555556e-05,
|
| 489 |
+
"loss": 0.0086,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 3.4444444444444446,
|
| 494 |
+
"grad_norm": 0.07111264020204544,
|
| 495 |
+
"learning_rate": 2.777777777777778e-05,
|
| 496 |
+
"loss": 0.0295,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 3.5,
|
| 501 |
+
"grad_norm": 0.028257286176085472,
|
| 502 |
+
"learning_rate": 2.5e-05,
|
| 503 |
+
"loss": 0.0065,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 3.5555555555555554,
|
| 508 |
+
"grad_norm": 0.02719848044216633,
|
| 509 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 510 |
+
"loss": 0.0418,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 3.611111111111111,
|
| 515 |
+
"grad_norm": 0.026137089356780052,
|
| 516 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 517 |
+
"loss": 0.0073,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 3.6666666666666665,
|
| 522 |
+
"grad_norm": 0.030431417748332024,
|
| 523 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 524 |
+
"loss": 0.0101,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 3.7222222222222223,
|
| 529 |
+
"grad_norm": 0.025364473462104797,
|
| 530 |
+
"learning_rate": 1.388888888888889e-05,
|
| 531 |
"loss": 0.0062,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 3.7777777777777777,
|
| 536 |
+
"grad_norm": 0.02630157209932804,
|
| 537 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 538 |
+
"loss": 0.0076,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 3.8333333333333335,
|
| 543 |
+
"grad_norm": 0.025917503982782364,
|
| 544 |
+
"learning_rate": 8.333333333333334e-06,
|
| 545 |
+
"loss": 0.0062,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.888888888888889,
|
| 550 |
+
"grad_norm": 6.676637649536133,
|
| 551 |
+
"learning_rate": 5.555555555555556e-06,
|
| 552 |
+
"loss": 0.0126,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 3.888888888888889,
|
| 557 |
+
"eval_accuracy": 0.7842639593908629,
|
| 558 |
+
"eval_loss": 1.1716859340667725,
|
| 559 |
+
"eval_runtime": 6.745,
|
| 560 |
+
"eval_samples_per_second": 58.414,
|
| 561 |
+
"eval_steps_per_second": 7.413,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 3.9444444444444446,
|
| 566 |
+
"grad_norm": 0.025717712938785553,
|
| 567 |
+
"learning_rate": 2.777777777777778e-06,
|
| 568 |
+
"loss": 0.008,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 4.0,
|
| 573 |
+
"grad_norm": 0.07326529920101166,
|
| 574 |
+
"learning_rate": 0.0,
|
| 575 |
+
"loss": 0.0065,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 4.0,
|
| 580 |
"step": 720,
|
| 581 |
"total_flos": 3.16768696086528e+18,
|
| 582 |
+
"train_loss": 0.16034429804939362,
|
| 583 |
+
"train_runtime": 1070.1524,
|
| 584 |
+
"train_samples_per_second": 10.727,
|
| 585 |
+
"train_steps_per_second": 0.673
|
| 586 |
}
|
| 587 |
],
|
| 588 |
"logging_steps": 10,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fce6746ccd146e07105259c2e2e4af9d52f0db85b3fc814d35a71a4f602c62a9
|
| 3 |
size 4984
|