Commit
•
c99ba2d
1
Parent(s):
f37dd19
initial commit
Browse files- README.md +12 -12
- all_results.json +9 -9
- eval_results.json +5 -5
- model.safetensors +1 -1
- runs/May01_17-28-34_e5fd9b370bfd/events.out.tfevents.1714584515.e5fd9b370bfd.197.0 +3 -0
- runs/May01_17-28-34_e5fd9b370bfd/events.out.tfevents.1714585252.e5fd9b370bfd.197.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +234 -234
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -15,7 +15,7 @@ model-index:
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
-
name:
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: Testing
|
|
@@ -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
|
|
@@ -31,10 +31,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 31 |
|
| 32 |
# vit-base-brain-xray
|
| 33 |
|
| 34 |
-
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 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.
|
| 75 |
-
| 0.
|
| 76 |
|
| 77 |
|
| 78 |
### Framework versions
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: sartajbhuvaji/Brain-Tumor-Classification
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: Testing
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.6903553299492385
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 31 |
|
| 32 |
# vit-base-brain-xray
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the sartajbhuvaji/Brain-Tumor-Classification dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.9079
|
| 37 |
+
- Accuracy: 0.6904
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 66 |
|
| 67 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 68 |
|:-------------:|:------:|:----:|:---------------:|:--------:|
|
| 69 |
+
| 0.2478 | 0.5556 | 100 | 0.9079 | 0.6904 |
|
| 70 |
+
| 0.1499 | 1.1111 | 200 | 1.1543 | 0.7183 |
|
| 71 |
+
| 0.0872 | 1.6667 | 300 | 1.1469 | 0.7614 |
|
| 72 |
+
| 0.0118 | 2.2222 | 400 | 1.2361 | 0.7259 |
|
| 73 |
+
| 0.0077 | 2.7778 | 500 | 1.2023 | 0.7665 |
|
| 74 |
+
| 0.0057 | 3.3333 | 600 | 1.2470 | 0.7640 |
|
| 75 |
+
| 0.0053 | 3.8889 | 700 | 1.2096 | 0.7766 |
|
| 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": 0.
|
| 5 |
-
"eval_runtime": 6.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 7.
|
| 8 |
"total_flos": 8.896239873264845e+17,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.6903553299492385,
|
| 4 |
+
"eval_loss": 0.907917320728302,
|
| 5 |
+
"eval_runtime": 6.3301,
|
| 6 |
+
"eval_samples_per_second": 62.243,
|
| 7 |
+
"eval_steps_per_second": 7.899,
|
| 8 |
"total_flos": 8.896239873264845e+17,
|
| 9 |
+
"train_loss": 0.12508326765770714,
|
| 10 |
+
"train_runtime": 729.602,
|
| 11 |
+
"train_samples_per_second": 15.735,
|
| 12 |
+
"train_steps_per_second": 0.987
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime": 6.
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 7.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.6903553299492385,
|
| 4 |
+
"eval_loss": 0.907917320728302,
|
| 5 |
+
"eval_runtime": 6.3301,
|
| 6 |
+
"eval_samples_per_second": 62.243,
|
| 7 |
+
"eval_steps_per_second": 7.899
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343230128
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c227906032c442ae079e617ca72129da0a452e7febd5fc0ad1725fb8aa40e959
|
| 3 |
size 343230128
|
runs/May01_17-28-34_e5fd9b370bfd/events.out.tfevents.1714584515.e5fd9b370bfd.197.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3cc6b9a201d2655907073a677e4857c96edc3d867e34b88bff75ade0d4a46458
|
| 3 |
+
size 22559
|
runs/May01_17-28-34_e5fd9b370bfd/events.out.tfevents.1714585252.e5fd9b370bfd.197.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9536aa1016b0f62f3008701a30c91f638a9fb6330414484f8652ef8c0fb4a4cf
|
| 3 |
+
size 411
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
"total_flos": 8.896239873264845e+17,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
"total_flos": 8.896239873264845e+17,
|
| 4 |
+
"train_loss": 0.12508326765770714,
|
| 5 |
+
"train_runtime": 729.602,
|
| 6 |
+
"train_samples_per_second": 15.735,
|
| 7 |
+
"train_steps_per_second": 0.987
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
"best_model_checkpoint": "./vit-base-brain-xray/checkpoint-100",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
|
@@ -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": 3.
|
| 21 |
"learning_rate": 0.00019444444444444446,
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.16666666666666666,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2222222222222222,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.2777777777777778,
|
| 41 |
-
"grad_norm": 1.
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.3333333333333333,
|
| 48 |
-
"grad_norm": 2.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.3888888888888889,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate": 0.
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.4444444444444444,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
-
"learning_rate": 0.
|
| 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": 0.
|
| 85 |
-
"eval_runtime": 5.
|
| 86 |
-
"eval_samples_per_second":
|
| 87 |
-
"eval_steps_per_second": 8.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.6111111111111112,
|
| 92 |
-
"grad_norm":
|
| 93 |
-
"learning_rate": 0.
|
| 94 |
-
"loss": 0.
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.6666666666666666,
|
| 99 |
-
"grad_norm":
|
| 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":
|
| 114 |
-
"learning_rate": 0.
|
| 115 |
-
"loss": 0.
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.8333333333333334,
|
| 120 |
-
"grad_norm": 4.
|
| 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":
|
| 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": 5.
|
| 165 |
-
"eval_samples_per_second":
|
| 166 |
-
"eval_steps_per_second": 8.
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.1666666666666667,
|
| 171 |
-
"grad_norm": 0.
|
| 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":
|
| 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": 0.
|
| 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": 1.
|
| 243 |
-
"eval_runtime": 6.
|
| 244 |
-
"eval_samples_per_second":
|
| 245 |
-
"eval_steps_per_second": 7.
|
| 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": 0.
|
| 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":
|
| 293 |
-
"learning_rate": 9.
|
| 294 |
-
"loss": 0.
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 2.111111111111111,
|
| 299 |
-
"grad_norm":
|
| 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":
|
| 323 |
-
"eval_samples_per_second":
|
| 324 |
-
"eval_steps_per_second":
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 2.2777777777777777,
|
| 329 |
-
"grad_norm":
|
| 330 |
-
"learning_rate": 8.
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 2.3333333333333335,
|
| 336 |
-
"grad_norm":
|
| 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": 0.
|
| 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": 5.
|
| 402 |
-
"eval_samples_per_second":
|
| 403 |
-
"eval_steps_per_second": 8.
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 2.8333333333333335,
|
| 408 |
-
"grad_norm": 0.
|
| 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":
|
| 423 |
"learning_rate": 5.305555555555556e-05,
|
| 424 |
-
"loss": 0.
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 3.0,
|
| 429 |
-
"grad_norm": 0.
|
| 430 |
"learning_rate": 5.027777777777778e-05,
|
| 431 |
-
"loss": 0.
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 3.0555555555555554,
|
| 436 |
-
"grad_norm": 0.
|
| 437 |
"learning_rate": 4.75e-05,
|
| 438 |
-
"loss": 0.
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 3.111111111111111,
|
| 443 |
-
"grad_norm": 0.
|
| 444 |
"learning_rate": 4.472222222222223e-05,
|
| 445 |
-
"loss": 0.
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 3.1666666666666665,
|
| 450 |
-
"grad_norm": 0.
|
| 451 |
"learning_rate": 4.194444444444445e-05,
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 3.2222222222222223,
|
| 457 |
-
"grad_norm": 0.
|
| 458 |
"learning_rate": 3.9166666666666665e-05,
|
| 459 |
-
"loss": 0.
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 3.2777777777777777,
|
| 464 |
-
"grad_norm": 0.
|
| 465 |
"learning_rate": 3.638888888888889e-05,
|
| 466 |
-
"loss": 0.
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 3.3333333333333335,
|
| 471 |
-
"grad_norm": 0.
|
| 472 |
"learning_rate": 3.3611111111111116e-05,
|
| 473 |
-
"loss": 0.
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 3.3333333333333335,
|
| 478 |
-
"eval_accuracy": 0.
|
| 479 |
-
"eval_loss": 1.
|
| 480 |
-
"eval_runtime": 5.
|
| 481 |
-
"eval_samples_per_second":
|
| 482 |
-
"eval_steps_per_second": 8.
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 3.388888888888889,
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
"learning_rate": 3.0833333333333335e-05,
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 3.4444444444444446,
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
"learning_rate": 2.8055555555555557e-05,
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 3.5,
|
| 501 |
-
"grad_norm": 0.
|
| 502 |
"learning_rate": 2.527777777777778e-05,
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 3.5555555555555554,
|
| 508 |
-
"grad_norm": 0.
|
| 509 |
"learning_rate": 2.25e-05,
|
| 510 |
-
"loss": 0.
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 3.611111111111111,
|
| 515 |
-
"grad_norm": 0.
|
| 516 |
"learning_rate": 1.9722222222222224e-05,
|
| 517 |
-
"loss": 0.
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 3.6666666666666665,
|
| 522 |
-
"grad_norm": 0.
|
| 523 |
"learning_rate": 1.6944444444444446e-05,
|
| 524 |
-
"loss": 0.
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 3.7222222222222223,
|
| 529 |
-
"grad_norm": 0.
|
| 530 |
"learning_rate": 1.4166666666666668e-05,
|
| 531 |
-
"loss": 0.
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 3.7777777777777777,
|
| 536 |
-
"grad_norm": 0.
|
| 537 |
"learning_rate": 1.138888888888889e-05,
|
| 538 |
-
"loss": 0.
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 3.8333333333333335,
|
| 543 |
-
"grad_norm": 0.
|
| 544 |
"learning_rate": 8.611111111111112e-06,
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.888888888888889,
|
| 550 |
-
"grad_norm": 0.
|
| 551 |
"learning_rate": 5.833333333333334e-06,
|
| 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": 3.0555555555555556e-06,
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 4.0,
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
"learning_rate": 2.777777777777778e-07,
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 4.0,
|
| 580 |
"step": 720,
|
| 581 |
"total_flos": 8.896239873264845e+17,
|
| 582 |
-
"train_loss": 0.
|
| 583 |
-
"train_runtime":
|
| 584 |
-
"train_samples_per_second":
|
| 585 |
-
"train_steps_per_second":
|
| 586 |
}
|
| 587 |
],
|
| 588 |
"logging_steps": 10,
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.907917320728302,
|
| 3 |
"best_model_checkpoint": "./vit-base-brain-xray/checkpoint-100",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05555555555555555,
|
| 13 |
+
"grad_norm": 5.699078559875488,
|
| 14 |
"learning_rate": 0.00019722222222222225,
|
| 15 |
+
"loss": 1.2918,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1111111111111111,
|
| 20 |
+
"grad_norm": 3.1500043869018555,
|
| 21 |
"learning_rate": 0.00019444444444444446,
|
| 22 |
+
"loss": 0.8867,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.16666666666666666,
|
| 27 |
+
"grad_norm": 3.1283438205718994,
|
| 28 |
+
"learning_rate": 0.00019166666666666667,
|
| 29 |
+
"loss": 0.5707,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2222222222222222,
|
| 34 |
+
"grad_norm": 2.686929225921631,
|
| 35 |
+
"learning_rate": 0.00018888888888888888,
|
| 36 |
+
"loss": 0.3967,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.2777777777777778,
|
| 41 |
+
"grad_norm": 1.111426830291748,
|
| 42 |
+
"learning_rate": 0.00018611111111111112,
|
| 43 |
+
"loss": 0.3759,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.3333333333333333,
|
| 48 |
+
"grad_norm": 2.6397788524627686,
|
| 49 |
+
"learning_rate": 0.00018333333333333334,
|
| 50 |
+
"loss": 0.2927,
|
| 51 |
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.3888888888888889,
|
| 55 |
+
"grad_norm": 0.6170921325683594,
|
| 56 |
+
"learning_rate": 0.00018055555555555557,
|
| 57 |
+
"loss": 0.2122,
|
| 58 |
"step": 70
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.4444444444444444,
|
| 62 |
+
"grad_norm": 0.33072641491889954,
|
| 63 |
+
"learning_rate": 0.00017777777777777779,
|
| 64 |
+
"loss": 0.2795,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.5,
|
| 69 |
+
"grad_norm": 1.4631454944610596,
|
| 70 |
+
"learning_rate": 0.000175,
|
| 71 |
+
"loss": 0.3315,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.5555555555555556,
|
| 76 |
+
"grad_norm": 7.331964492797852,
|
| 77 |
+
"learning_rate": 0.00017222222222222224,
|
| 78 |
+
"loss": 0.2478,
|
| 79 |
"step": 100
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.5555555555555556,
|
| 83 |
+
"eval_accuracy": 0.6903553299492385,
|
| 84 |
+
"eval_loss": 0.907917320728302,
|
| 85 |
+
"eval_runtime": 5.7847,
|
| 86 |
+
"eval_samples_per_second": 68.111,
|
| 87 |
+
"eval_steps_per_second": 8.644,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.6111111111111112,
|
| 92 |
+
"grad_norm": 3.3308780193328857,
|
| 93 |
+
"learning_rate": 0.00016944444444444445,
|
| 94 |
+
"loss": 0.4174,
|
| 95 |
"step": 110
|
| 96 |
},
|
| 97 |
{
|
| 98 |
"epoch": 0.6666666666666666,
|
| 99 |
+
"grad_norm": 3.7708303928375244,
|
| 100 |
+
"learning_rate": 0.0001666666666666667,
|
| 101 |
+
"loss": 0.2337,
|
| 102 |
"step": 120
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 0.7222222222222222,
|
| 106 |
+
"grad_norm": 0.6579460501670837,
|
| 107 |
+
"learning_rate": 0.0001638888888888889,
|
| 108 |
+
"loss": 0.271,
|
| 109 |
"step": 130
|
| 110 |
},
|
| 111 |
{
|
| 112 |
"epoch": 0.7777777777777778,
|
| 113 |
+
"grad_norm": 2.6412248611450195,
|
| 114 |
+
"learning_rate": 0.0001611111111111111,
|
| 115 |
+
"loss": 0.2476,
|
| 116 |
"step": 140
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.8333333333333334,
|
| 120 |
+
"grad_norm": 4.0377197265625,
|
| 121 |
+
"learning_rate": 0.00015833333333333332,
|
| 122 |
+
"loss": 0.1787,
|
| 123 |
"step": 150
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 0.8888888888888888,
|
| 127 |
+
"grad_norm": 5.922617435455322,
|
| 128 |
+
"learning_rate": 0.00015555555555555556,
|
| 129 |
+
"loss": 0.1848,
|
| 130 |
"step": 160
|
| 131 |
},
|
| 132 |
{
|
| 133 |
"epoch": 0.9444444444444444,
|
| 134 |
+
"grad_norm": 4.513651371002197,
|
| 135 |
+
"learning_rate": 0.00015277777777777777,
|
| 136 |
+
"loss": 0.2581,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 1.0,
|
| 141 |
+
"grad_norm": 0.23064616322517395,
|
| 142 |
+
"learning_rate": 0.00015000000000000001,
|
| 143 |
+
"loss": 0.1437,
|
| 144 |
"step": 180
|
| 145 |
},
|
| 146 |
{
|
| 147 |
"epoch": 1.0555555555555556,
|
| 148 |
+
"grad_norm": 2.2603862285614014,
|
| 149 |
+
"learning_rate": 0.00014722222222222223,
|
| 150 |
+
"loss": 0.1052,
|
| 151 |
"step": 190
|
| 152 |
},
|
| 153 |
{
|
| 154 |
"epoch": 1.1111111111111112,
|
| 155 |
+
"grad_norm": 4.710208892822266,
|
| 156 |
+
"learning_rate": 0.00014444444444444444,
|
| 157 |
+
"loss": 0.1499,
|
| 158 |
"step": 200
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 1.1111111111111112,
|
| 162 |
+
"eval_accuracy": 0.7182741116751269,
|
| 163 |
+
"eval_loss": 1.1542534828186035,
|
| 164 |
+
"eval_runtime": 5.8503,
|
| 165 |
+
"eval_samples_per_second": 67.347,
|
| 166 |
+
"eval_steps_per_second": 8.547,
|
| 167 |
"step": 200
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 1.1666666666666667,
|
| 171 |
+
"grad_norm": 0.15027037262916565,
|
| 172 |
+
"learning_rate": 0.00014166666666666668,
|
| 173 |
+
"loss": 0.0747,
|
| 174 |
"step": 210
|
| 175 |
},
|
| 176 |
{
|
| 177 |
"epoch": 1.2222222222222223,
|
| 178 |
+
"grad_norm": 0.3188658356666565,
|
| 179 |
+
"learning_rate": 0.0001388888888888889,
|
| 180 |
+
"loss": 0.1555,
|
| 181 |
"step": 220
|
| 182 |
},
|
| 183 |
{
|
| 184 |
"epoch": 1.2777777777777777,
|
| 185 |
+
"grad_norm": 3.695791482925415,
|
| 186 |
+
"learning_rate": 0.00013611111111111113,
|
| 187 |
+
"loss": 0.1876,
|
| 188 |
"step": 230
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 1.3333333333333333,
|
| 192 |
+
"grad_norm": 2.009079694747925,
|
| 193 |
+
"learning_rate": 0.00013333333333333334,
|
| 194 |
+
"loss": 0.1054,
|
| 195 |
"step": 240
|
| 196 |
},
|
| 197 |
{
|
| 198 |
"epoch": 1.3888888888888888,
|
| 199 |
+
"grad_norm": 0.07869268208742142,
|
| 200 |
+
"learning_rate": 0.00013055555555555555,
|
| 201 |
+
"loss": 0.0381,
|
| 202 |
"step": 250
|
| 203 |
},
|
| 204 |
{
|
| 205 |
"epoch": 1.4444444444444444,
|
| 206 |
+
"grad_norm": 0.9898849725723267,
|
| 207 |
+
"learning_rate": 0.00012777777777777776,
|
| 208 |
+
"loss": 0.1144,
|
| 209 |
"step": 260
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 1.5,
|
| 213 |
+
"grad_norm": 0.509087085723877,
|
| 214 |
+
"learning_rate": 0.000125,
|
| 215 |
+
"loss": 0.1002,
|
| 216 |
"step": 270
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 1.5555555555555556,
|
| 220 |
+
"grad_norm": 0.08736218512058258,
|
| 221 |
+
"learning_rate": 0.00012222222222222224,
|
| 222 |
+
"loss": 0.1103,
|
| 223 |
"step": 280
|
| 224 |
},
|
| 225 |
{
|
| 226 |
"epoch": 1.6111111111111112,
|
| 227 |
+
"grad_norm": 0.24889719486236572,
|
| 228 |
+
"learning_rate": 0.00011944444444444445,
|
| 229 |
+
"loss": 0.0987,
|
| 230 |
"step": 290
|
| 231 |
},
|
| 232 |
{
|
| 233 |
"epoch": 1.6666666666666665,
|
| 234 |
+
"grad_norm": 2.0384163856506348,
|
| 235 |
+
"learning_rate": 0.00011666666666666668,
|
| 236 |
+
"loss": 0.0872,
|
| 237 |
"step": 300
|
| 238 |
},
|
| 239 |
{
|
| 240 |
"epoch": 1.6666666666666665,
|
| 241 |
+
"eval_accuracy": 0.7614213197969543,
|
| 242 |
+
"eval_loss": 1.146877646446228,
|
| 243 |
+
"eval_runtime": 6.3643,
|
| 244 |
+
"eval_samples_per_second": 61.908,
|
| 245 |
+
"eval_steps_per_second": 7.856,
|
| 246 |
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
"epoch": 1.7222222222222223,
|
| 250 |
+
"grad_norm": 0.47840073704719543,
|
| 251 |
+
"learning_rate": 0.00011388888888888889,
|
| 252 |
+
"loss": 0.0468,
|
| 253 |
"step": 310
|
| 254 |
},
|
| 255 |
{
|
| 256 |
"epoch": 1.7777777777777777,
|
| 257 |
+
"grad_norm": 0.06918664276599884,
|
| 258 |
+
"learning_rate": 0.00011111111111111112,
|
| 259 |
+
"loss": 0.0191,
|
| 260 |
"step": 320
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 1.8333333333333335,
|
| 264 |
+
"grad_norm": 0.7651189565658569,
|
| 265 |
+
"learning_rate": 0.00010833333333333333,
|
| 266 |
+
"loss": 0.1305,
|
| 267 |
"step": 330
|
| 268 |
},
|
| 269 |
{
|
| 270 |
"epoch": 1.8888888888888888,
|
| 271 |
+
"grad_norm": 1.8910560607910156,
|
| 272 |
+
"learning_rate": 0.00010555555555555557,
|
| 273 |
+
"loss": 0.0355,
|
| 274 |
"step": 340
|
| 275 |
},
|
| 276 |
{
|
| 277 |
"epoch": 1.9444444444444444,
|
| 278 |
+
"grad_norm": 0.07424870133399963,
|
| 279 |
+
"learning_rate": 0.00010277777777777778,
|
| 280 |
+
"loss": 0.0567,
|
| 281 |
"step": 350
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 2.0,
|
| 285 |
+
"grad_norm": 0.05715026333928108,
|
| 286 |
+
"learning_rate": 0.0001,
|
| 287 |
+
"loss": 0.0991,
|
| 288 |
"step": 360
|
| 289 |
},
|
| 290 |
{
|
| 291 |
"epoch": 2.0555555555555554,
|
| 292 |
+
"grad_norm": 2.061647415161133,
|
| 293 |
+
"learning_rate": 9.722222222222223e-05,
|
| 294 |
+
"loss": 0.0603,
|
| 295 |
"step": 370
|
| 296 |
},
|
| 297 |
{
|
| 298 |
"epoch": 2.111111111111111,
|
| 299 |
+
"grad_norm": 0.04806053638458252,
|
| 300 |
+
"learning_rate": 9.444444444444444e-05,
|
| 301 |
+
"loss": 0.0325,
|
| 302 |
"step": 380
|
| 303 |
},
|
| 304 |
{
|
| 305 |
"epoch": 2.1666666666666665,
|
| 306 |
+
"grad_norm": 0.05517473816871643,
|
| 307 |
+
"learning_rate": 9.166666666666667e-05,
|
| 308 |
+
"loss": 0.0301,
|
| 309 |
"step": 390
|
| 310 |
},
|
| 311 |
{
|
| 312 |
"epoch": 2.2222222222222223,
|
| 313 |
+
"grad_norm": 0.8571930527687073,
|
| 314 |
+
"learning_rate": 8.888888888888889e-05,
|
| 315 |
+
"loss": 0.0118,
|
| 316 |
"step": 400
|
| 317 |
},
|
| 318 |
{
|
| 319 |
"epoch": 2.2222222222222223,
|
| 320 |
+
"eval_accuracy": 0.7258883248730964,
|
| 321 |
+
"eval_loss": 1.2360663414001465,
|
| 322 |
+
"eval_runtime": 6.5101,
|
| 323 |
+
"eval_samples_per_second": 60.522,
|
| 324 |
+
"eval_steps_per_second": 7.68,
|
| 325 |
"step": 400
|
| 326 |
},
|
| 327 |
{
|
| 328 |
"epoch": 2.2777777777777777,
|
| 329 |
+
"grad_norm": 0.040004875510931015,
|
| 330 |
+
"learning_rate": 8.611111111111112e-05,
|
| 331 |
+
"loss": 0.0118,
|
| 332 |
"step": 410
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 2.3333333333333335,
|
| 336 |
+
"grad_norm": 7.572581768035889,
|
| 337 |
+
"learning_rate": 8.333333333333334e-05,
|
| 338 |
+
"loss": 0.0314,
|
| 339 |
"step": 420
|
| 340 |
},
|
| 341 |
{
|
| 342 |
"epoch": 2.388888888888889,
|
| 343 |
+
"grad_norm": 0.037391383200883865,
|
| 344 |
+
"learning_rate": 8.055555555555556e-05,
|
| 345 |
+
"loss": 0.0088,
|
| 346 |
"step": 430
|
| 347 |
},
|
| 348 |
{
|
| 349 |
"epoch": 2.4444444444444446,
|
| 350 |
+
"grad_norm": 0.034802764654159546,
|
| 351 |
+
"learning_rate": 7.777777777777778e-05,
|
| 352 |
+
"loss": 0.0147,
|
| 353 |
"step": 440
|
| 354 |
},
|
| 355 |
{
|
| 356 |
"epoch": 2.5,
|
| 357 |
+
"grad_norm": 0.06982697546482086,
|
| 358 |
+
"learning_rate": 7.500000000000001e-05,
|
| 359 |
+
"loss": 0.0085,
|
| 360 |
"step": 450
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"epoch": 2.5555555555555554,
|
| 364 |
+
"grad_norm": 0.03507501631975174,
|
| 365 |
+
"learning_rate": 7.222222222222222e-05,
|
| 366 |
+
"loss": 0.0087,
|
| 367 |
"step": 460
|
| 368 |
},
|
| 369 |
{
|
| 370 |
"epoch": 2.611111111111111,
|
| 371 |
+
"grad_norm": 4.626717567443848,
|
| 372 |
+
"learning_rate": 6.944444444444444e-05,
|
| 373 |
+
"loss": 0.0216,
|
| 374 |
"step": 470
|
| 375 |
},
|
| 376 |
{
|
| 377 |
"epoch": 2.6666666666666665,
|
| 378 |
+
"grad_norm": 0.0349123477935791,
|
| 379 |
+
"learning_rate": 6.666666666666667e-05,
|
| 380 |
+
"loss": 0.05,
|
| 381 |
"step": 480
|
| 382 |
},
|
| 383 |
{
|
| 384 |
"epoch": 2.7222222222222223,
|
| 385 |
+
"grad_norm": 0.03614797443151474,
|
| 386 |
+
"learning_rate": 6.388888888888888e-05,
|
| 387 |
+
"loss": 0.0085,
|
| 388 |
"step": 490
|
| 389 |
},
|
| 390 |
{
|
| 391 |
"epoch": 2.7777777777777777,
|
| 392 |
+
"grad_norm": 0.06703199446201324,
|
| 393 |
+
"learning_rate": 6.111111111111112e-05,
|
| 394 |
+
"loss": 0.0077,
|
| 395 |
"step": 500
|
| 396 |
},
|
| 397 |
{
|
| 398 |
"epoch": 2.7777777777777777,
|
| 399 |
+
"eval_accuracy": 0.766497461928934,
|
| 400 |
+
"eval_loss": 1.2023228406906128,
|
| 401 |
+
"eval_runtime": 5.9054,
|
| 402 |
+
"eval_samples_per_second": 66.719,
|
| 403 |
+
"eval_steps_per_second": 8.467,
|
| 404 |
"step": 500
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 2.8333333333333335,
|
| 408 |
+
"grad_norm": 0.03837073966860771,
|
| 409 |
+
"learning_rate": 5.833333333333334e-05,
|
| 410 |
+
"loss": 0.007,
|
| 411 |
"step": 510
|
| 412 |
},
|
| 413 |
{
|
| 414 |
"epoch": 2.888888888888889,
|
| 415 |
+
"grad_norm": 0.031188365072011948,
|
| 416 |
+
"learning_rate": 5.555555555555556e-05,
|
| 417 |
+
"loss": 0.0072,
|
| 418 |
"step": 520
|
| 419 |
},
|
| 420 |
{
|
| 421 |
"epoch": 2.9444444444444446,
|
| 422 |
+
"grad_norm": 0.15879875421524048,
|
| 423 |
"learning_rate": 5.305555555555556e-05,
|
| 424 |
+
"loss": 0.0145,
|
| 425 |
"step": 530
|
| 426 |
},
|
| 427 |
{
|
| 428 |
"epoch": 3.0,
|
| 429 |
+
"grad_norm": 0.051313698291778564,
|
| 430 |
"learning_rate": 5.027777777777778e-05,
|
| 431 |
+
"loss": 0.0067,
|
| 432 |
"step": 540
|
| 433 |
},
|
| 434 |
{
|
| 435 |
"epoch": 3.0555555555555554,
|
| 436 |
+
"grad_norm": 0.027085309848189354,
|
| 437 |
"learning_rate": 4.75e-05,
|
| 438 |
+
"loss": 0.0063,
|
| 439 |
"step": 550
|
| 440 |
},
|
| 441 |
{
|
| 442 |
"epoch": 3.111111111111111,
|
| 443 |
+
"grad_norm": 0.02693328447639942,
|
| 444 |
"learning_rate": 4.472222222222223e-05,
|
| 445 |
+
"loss": 0.0066,
|
| 446 |
"step": 560
|
| 447 |
},
|
| 448 |
{
|
| 449 |
"epoch": 3.1666666666666665,
|
| 450 |
+
"grad_norm": 0.027860965579748154,
|
| 451 |
"learning_rate": 4.194444444444445e-05,
|
| 452 |
+
"loss": 0.006,
|
| 453 |
"step": 570
|
| 454 |
},
|
| 455 |
{
|
| 456 |
"epoch": 3.2222222222222223,
|
| 457 |
+
"grad_norm": 0.025451989844441414,
|
| 458 |
"learning_rate": 3.9166666666666665e-05,
|
| 459 |
+
"loss": 0.006,
|
| 460 |
"step": 580
|
| 461 |
},
|
| 462 |
{
|
| 463 |
"epoch": 3.2777777777777777,
|
| 464 |
+
"grad_norm": 0.02524595521390438,
|
| 465 |
"learning_rate": 3.638888888888889e-05,
|
| 466 |
+
"loss": 0.0059,
|
| 467 |
"step": 590
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 3.3333333333333335,
|
| 471 |
+
"grad_norm": 0.02702181413769722,
|
| 472 |
"learning_rate": 3.3611111111111116e-05,
|
| 473 |
+
"loss": 0.0057,
|
| 474 |
"step": 600
|
| 475 |
},
|
| 476 |
{
|
| 477 |
"epoch": 3.3333333333333335,
|
| 478 |
+
"eval_accuracy": 0.7639593908629442,
|
| 479 |
+
"eval_loss": 1.2470098733901978,
|
| 480 |
+
"eval_runtime": 5.9411,
|
| 481 |
+
"eval_samples_per_second": 66.318,
|
| 482 |
+
"eval_steps_per_second": 8.416,
|
| 483 |
"step": 600
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 3.388888888888889,
|
| 487 |
+
"grad_norm": 0.027771031484007835,
|
| 488 |
"learning_rate": 3.0833333333333335e-05,
|
| 489 |
+
"loss": 0.0057,
|
| 490 |
"step": 610
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 3.4444444444444446,
|
| 494 |
+
"grad_norm": 0.02504831738770008,
|
| 495 |
"learning_rate": 2.8055555555555557e-05,
|
| 496 |
+
"loss": 0.0313,
|
| 497 |
"step": 620
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 3.5,
|
| 501 |
+
"grad_norm": 0.027147900313138962,
|
| 502 |
"learning_rate": 2.527777777777778e-05,
|
| 503 |
+
"loss": 0.0056,
|
| 504 |
"step": 630
|
| 505 |
},
|
| 506 |
{
|
| 507 |
"epoch": 3.5555555555555554,
|
| 508 |
+
"grad_norm": 0.026741115376353264,
|
| 509 |
"learning_rate": 2.25e-05,
|
| 510 |
+
"loss": 0.0059,
|
| 511 |
"step": 640
|
| 512 |
},
|
| 513 |
{
|
| 514 |
"epoch": 3.611111111111111,
|
| 515 |
+
"grad_norm": 0.05946130305528641,
|
| 516 |
"learning_rate": 1.9722222222222224e-05,
|
| 517 |
+
"loss": 0.0055,
|
| 518 |
"step": 650
|
| 519 |
},
|
| 520 |
{
|
| 521 |
"epoch": 3.6666666666666665,
|
| 522 |
+
"grad_norm": 0.025428051128983498,
|
| 523 |
"learning_rate": 1.6944444444444446e-05,
|
| 524 |
+
"loss": 0.0054,
|
| 525 |
"step": 660
|
| 526 |
},
|
| 527 |
{
|
| 528 |
"epoch": 3.7222222222222223,
|
| 529 |
+
"grad_norm": 0.023423248901963234,
|
| 530 |
"learning_rate": 1.4166666666666668e-05,
|
| 531 |
+
"loss": 0.0053,
|
| 532 |
"step": 670
|
| 533 |
},
|
| 534 |
{
|
| 535 |
"epoch": 3.7777777777777777,
|
| 536 |
+
"grad_norm": 0.023936165496706963,
|
| 537 |
"learning_rate": 1.138888888888889e-05,
|
| 538 |
+
"loss": 0.0057,
|
| 539 |
"step": 680
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 3.8333333333333335,
|
| 543 |
+
"grad_norm": 0.023328222334384918,
|
| 544 |
"learning_rate": 8.611111111111112e-06,
|
| 545 |
+
"loss": 0.0052,
|
| 546 |
"step": 690
|
| 547 |
},
|
| 548 |
{
|
| 549 |
"epoch": 3.888888888888889,
|
| 550 |
+
"grad_norm": 0.025306882336735725,
|
| 551 |
"learning_rate": 5.833333333333334e-06,
|
| 552 |
+
"loss": 0.0053,
|
| 553 |
"step": 700
|
| 554 |
},
|
| 555 |
{
|
| 556 |
"epoch": 3.888888888888889,
|
| 557 |
+
"eval_accuracy": 0.7766497461928934,
|
| 558 |
+
"eval_loss": 1.209647297859192,
|
| 559 |
+
"eval_runtime": 7.4912,
|
| 560 |
+
"eval_samples_per_second": 52.595,
|
| 561 |
+
"eval_steps_per_second": 6.675,
|
| 562 |
"step": 700
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 3.9444444444444446,
|
| 566 |
+
"grad_norm": 0.023267222568392754,
|
| 567 |
"learning_rate": 3.0555555555555556e-06,
|
| 568 |
+
"loss": 0.0057,
|
| 569 |
"step": 710
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 4.0,
|
| 573 |
+
"grad_norm": 0.03729303553700447,
|
| 574 |
"learning_rate": 2.777777777777778e-07,
|
| 575 |
+
"loss": 0.0057,
|
| 576 |
"step": 720
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 4.0,
|
| 580 |
"step": 720,
|
| 581 |
"total_flos": 8.896239873264845e+17,
|
| 582 |
+
"train_loss": 0.12508326765770714,
|
| 583 |
+
"train_runtime": 729.602,
|
| 584 |
+
"train_samples_per_second": 15.735,
|
| 585 |
+
"train_steps_per_second": 0.987
|
| 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:aa9c86430e116816f9bd3e4f75e40ee3d211f6fccb42b21939614223c8b5b485
|
| 3 |
size 4984
|