Upload folder using huggingface_hub
Browse files- checkpoint-100/model.safetensors +1 -1
- checkpoint-100/optimizer.pt +1 -1
- checkpoint-100/trainer_state.json +15 -15
- checkpoint-1000/model.safetensors +1 -1
- checkpoint-1000/optimizer.pt +1 -1
- checkpoint-1000/trainer_state.json +141 -141
- checkpoint-200/model.safetensors +1 -1
- checkpoint-200/optimizer.pt +1 -1
- checkpoint-200/trainer_state.json +29 -29
- checkpoint-300/model.safetensors +1 -1
- checkpoint-300/optimizer.pt +1 -1
- checkpoint-300/trainer_state.json +43 -43
- checkpoint-400/model.safetensors +1 -1
- checkpoint-400/optimizer.pt +1 -1
- checkpoint-400/trainer_state.json +57 -57
- checkpoint-500/model.safetensors +1 -1
- checkpoint-500/optimizer.pt +1 -1
- checkpoint-500/trainer_state.json +71 -71
- checkpoint-600/model.safetensors +1 -1
- checkpoint-600/optimizer.pt +1 -1
- checkpoint-600/trainer_state.json +85 -85
- checkpoint-700/model.safetensors +1 -1
- checkpoint-700/optimizer.pt +1 -1
- checkpoint-700/trainer_state.json +99 -99
- checkpoint-800/model.safetensors +1 -1
- checkpoint-800/optimizer.pt +1 -1
- checkpoint-800/trainer_state.json +115 -115
- checkpoint-900/model.safetensors +1 -1
- checkpoint-900/optimizer.pt +1 -1
- checkpoint-900/trainer_state.json +127 -127
- model.safetensors +1 -1
checkpoint-100/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44e8376d0df2030baceee2c4c7807b7c56568d28cc5cd0da0e52c2820bcd8fc8
|
| 3 |
size 124209532
|
checkpoint-100/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:322014de21646a346ec9ad3e448f269fc57980c2445d1e2821136b875c795f7b
|
| 3 |
size 248356555
|
checkpoint-100/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 100,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-100",
|
| 5 |
"epoch": 1.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,40 +11,40 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
}
|
| 50 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 100,
|
| 3 |
+
"best_metric": 0.5377293432044794,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-100",
|
| 5 |
"epoch": 1.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
}
|
| 50 |
],
|
checkpoint-1000/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b8ca815c4d3ef32947703903f06f8164304b43454932abcec8e328437ab81cd
|
| 3 |
size 124209532
|
checkpoint-1000/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f28e26b4242bde5906f4768eff11f94b6d8d0117d7b960e04a988c0d49aa6b7
|
| 3 |
size 248356555
|
checkpoint-1000/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 1000,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-1000",
|
| 5 |
"epoch": 10.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,382 +11,382 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
-
"grad_norm":
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
-
"grad_norm":
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
-
"grad_norm":
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
-
"grad_norm":
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
-
"eval_dice": 0.
|
| 195 |
-
"eval_iou": 0.
|
| 196 |
-
"eval_loss": 0.
|
| 197 |
-
"eval_runtime": 4.
|
| 198 |
-
"eval_samples_per_second": 22.
|
| 199 |
-
"eval_steps_per_second": 2.
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
-
"grad_norm": 9.
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
-
"grad_norm": 3.
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
-
"loss": 0.
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
-
"grad_norm":
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
-
"grad_norm": 3.
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
-
"eval_dice": 0.
|
| 233 |
-
"eval_iou": 0.
|
| 234 |
-
"eval_loss": 0.
|
| 235 |
-
"eval_runtime": 4.
|
| 236 |
-
"eval_samples_per_second": 22.
|
| 237 |
-
"eval_steps_per_second": 2.
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
-
"grad_norm":
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
-
"grad_norm":
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
-
"grad_norm":
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
-
"grad_norm":
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
-
"eval_dice": 0.
|
| 271 |
-
"eval_iou": 0.
|
| 272 |
-
"eval_loss": 0.
|
| 273 |
-
"eval_runtime": 4.
|
| 274 |
-
"eval_samples_per_second":
|
| 275 |
-
"eval_steps_per_second": 2.
|
| 276 |
"step": 700
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 7.25,
|
| 280 |
-
"grad_norm":
|
| 281 |
"learning_rate": 2.7600000000000003e-05,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 725
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 7.5,
|
| 287 |
-
"grad_norm": 1.
|
| 288 |
"learning_rate": 2.51e-05,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 750
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 7.75,
|
| 294 |
-
"grad_norm": 4.
|
| 295 |
"learning_rate": 2.26e-05,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 775
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 8.0,
|
| 301 |
-
"grad_norm": 3.
|
| 302 |
"learning_rate": 2.01e-05,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 800
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 8.0,
|
| 308 |
-
"eval_dice": 0.
|
| 309 |
-
"eval_iou": 0.
|
| 310 |
-
"eval_loss": 0.
|
| 311 |
-
"eval_runtime": 4.
|
| 312 |
-
"eval_samples_per_second": 21.
|
| 313 |
-
"eval_steps_per_second": 2.
|
| 314 |
"step": 800
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 8.25,
|
| 318 |
-
"grad_norm":
|
| 319 |
"learning_rate": 1.76e-05,
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 825
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 8.5,
|
| 325 |
-
"grad_norm":
|
| 326 |
"learning_rate": 1.51e-05,
|
| 327 |
-
"loss": 0.
|
| 328 |
"step": 850
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 8.75,
|
| 332 |
-
"grad_norm":
|
| 333 |
"learning_rate": 1.2600000000000001e-05,
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 875
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 9.0,
|
| 339 |
-
"grad_norm":
|
| 340 |
"learning_rate": 1.0100000000000002e-05,
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 900
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 9.0,
|
| 346 |
-
"eval_dice": 0.
|
| 347 |
-
"eval_iou": 0.
|
| 348 |
-
"eval_loss": 0.
|
| 349 |
-
"eval_runtime": 4.
|
| 350 |
-
"eval_samples_per_second":
|
| 351 |
-
"eval_steps_per_second": 2.
|
| 352 |
"step": 900
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 9.25,
|
| 356 |
-
"grad_norm": 3.
|
| 357 |
"learning_rate": 7.6e-06,
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 925
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 9.5,
|
| 363 |
-
"grad_norm": 2.
|
| 364 |
"learning_rate": 5.1e-06,
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 950
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 9.75,
|
| 370 |
-
"grad_norm":
|
| 371 |
"learning_rate": 2.6e-06,
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 975
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 10.0,
|
| 377 |
-
"grad_norm": 1.
|
| 378 |
"learning_rate": 1.0000000000000001e-07,
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 1000
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 10.0,
|
| 384 |
-
"eval_dice": 0.
|
| 385 |
-
"eval_iou": 0.
|
| 386 |
-
"eval_loss": 0.
|
| 387 |
-
"eval_runtime": 4.
|
| 388 |
-
"eval_samples_per_second":
|
| 389 |
-
"eval_steps_per_second": 2.
|
| 390 |
"step": 1000
|
| 391 |
}
|
| 392 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 1000,
|
| 3 |
+
"best_metric": 0.769044628133352,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-1000",
|
| 5 |
"epoch": 10.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
+
"grad_norm": 2.91507625579834,
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
+
"loss": 0.817935562133789,
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
+
"grad_norm": 6.0939226150512695,
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
+
"loss": 0.8120758056640625,
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
+
"grad_norm": 2.495445728302002,
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
+
"loss": 0.7950348663330078,
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
+
"grad_norm": 4.1540961265563965,
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
+
"loss": 0.8125821685791016,
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
+
"eval_dice": 0.6808156703817527,
|
| 195 |
+
"eval_iou": 0.5160885243741951,
|
| 196 |
+
"eval_loss": 0.8397173285484314,
|
| 197 |
+
"eval_runtime": 4.4587,
|
| 198 |
+
"eval_samples_per_second": 22.428,
|
| 199 |
+
"eval_steps_per_second": 2.916,
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
+
"grad_norm": 9.038034439086914,
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
+
"loss": 0.7783580780029297,
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
+
"grad_norm": 3.4795265197753906,
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
+
"loss": 0.7784259796142579,
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
+
"grad_norm": 2.7996904850006104,
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
+
"loss": 0.7458997344970704,
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
+
"grad_norm": 3.0424861907958984,
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
+
"loss": 0.7311380767822265,
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
+
"eval_dice": 0.7141547601985387,
|
| 233 |
+
"eval_iou": 0.5553973025470366,
|
| 234 |
+
"eval_loss": 0.7741679549217224,
|
| 235 |
+
"eval_runtime": 4.4797,
|
| 236 |
+
"eval_samples_per_second": 22.323,
|
| 237 |
+
"eval_steps_per_second": 2.902,
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
+
"grad_norm": 2.969958782196045,
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
+
"loss": 0.7293343353271484,
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
+
"grad_norm": 3.2517130374908447,
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
+
"loss": 0.7013224029541015,
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
+
"grad_norm": 2.587013006210327,
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
+
"loss": 0.6944268798828125,
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
+
"grad_norm": 2.868363380432129,
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
+
"loss": 0.7179935455322266,
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
+
"eval_dice": 0.7389087921045768,
|
| 271 |
+
"eval_iou": 0.5859282924237521,
|
| 272 |
+
"eval_loss": 0.7406055927276611,
|
| 273 |
+
"eval_runtime": 4.5995,
|
| 274 |
+
"eval_samples_per_second": 21.741,
|
| 275 |
+
"eval_steps_per_second": 2.826,
|
| 276 |
"step": 700
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 7.25,
|
| 280 |
+
"grad_norm": 4.949324131011963,
|
| 281 |
"learning_rate": 2.7600000000000003e-05,
|
| 282 |
+
"loss": 0.7042760467529297,
|
| 283 |
"step": 725
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 7.5,
|
| 287 |
+
"grad_norm": 1.5597704648971558,
|
| 288 |
"learning_rate": 2.51e-05,
|
| 289 |
+
"loss": 0.6764234161376953,
|
| 290 |
"step": 750
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 7.75,
|
| 294 |
+
"grad_norm": 4.044253349304199,
|
| 295 |
"learning_rate": 2.26e-05,
|
| 296 |
+
"loss": 0.6907909393310547,
|
| 297 |
"step": 775
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 8.0,
|
| 301 |
+
"grad_norm": 3.0667364597320557,
|
| 302 |
"learning_rate": 2.01e-05,
|
| 303 |
+
"loss": 0.6335317611694335,
|
| 304 |
"step": 800
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 8.0,
|
| 308 |
+
"eval_dice": 0.7353989116647303,
|
| 309 |
+
"eval_iou": 0.5815265551170842,
|
| 310 |
+
"eval_loss": 0.7389959692955017,
|
| 311 |
+
"eval_runtime": 4.6414,
|
| 312 |
+
"eval_samples_per_second": 21.545,
|
| 313 |
+
"eval_steps_per_second": 2.801,
|
| 314 |
"step": 800
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 8.25,
|
| 318 |
+
"grad_norm": 8.7631196975708,
|
| 319 |
"learning_rate": 1.76e-05,
|
| 320 |
+
"loss": 0.6445667266845703,
|
| 321 |
"step": 825
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 8.5,
|
| 325 |
+
"grad_norm": 3.8787753582000732,
|
| 326 |
"learning_rate": 1.51e-05,
|
| 327 |
+
"loss": 0.6259555816650391,
|
| 328 |
"step": 850
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 8.75,
|
| 332 |
+
"grad_norm": 2.609173536300659,
|
| 333 |
"learning_rate": 1.2600000000000001e-05,
|
| 334 |
+
"loss": 0.6285885620117188,
|
| 335 |
"step": 875
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 9.0,
|
| 339 |
+
"grad_norm": 3.2756707668304443,
|
| 340 |
"learning_rate": 1.0100000000000002e-05,
|
| 341 |
+
"loss": 0.643200912475586,
|
| 342 |
"step": 900
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 9.0,
|
| 346 |
+
"eval_dice": 0.7648601527792854,
|
| 347 |
+
"eval_iou": 0.6192499971116917,
|
| 348 |
+
"eval_loss": 0.6945236921310425,
|
| 349 |
+
"eval_runtime": 4.6242,
|
| 350 |
+
"eval_samples_per_second": 21.626,
|
| 351 |
+
"eval_steps_per_second": 2.811,
|
| 352 |
"step": 900
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 9.25,
|
| 356 |
+
"grad_norm": 3.0061779022216797,
|
| 357 |
"learning_rate": 7.6e-06,
|
| 358 |
+
"loss": 0.625957908630371,
|
| 359 |
"step": 925
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 9.5,
|
| 363 |
+
"grad_norm": 2.281498432159424,
|
| 364 |
"learning_rate": 5.1e-06,
|
| 365 |
+
"loss": 0.6123639678955078,
|
| 366 |
"step": 950
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 9.75,
|
| 370 |
+
"grad_norm": 2.7292234897613525,
|
| 371 |
"learning_rate": 2.6e-06,
|
| 372 |
+
"loss": 0.6031129455566406,
|
| 373 |
"step": 975
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 10.0,
|
| 377 |
+
"grad_norm": 1.900956630706787,
|
| 378 |
"learning_rate": 1.0000000000000001e-07,
|
| 379 |
+
"loss": 0.5966448211669921,
|
| 380 |
"step": 1000
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 10.0,
|
| 384 |
+
"eval_dice": 0.769044628133352,
|
| 385 |
+
"eval_iou": 0.6247544303329668,
|
| 386 |
+
"eval_loss": 0.6924489736557007,
|
| 387 |
+
"eval_runtime": 4.5712,
|
| 388 |
+
"eval_samples_per_second": 21.876,
|
| 389 |
+
"eval_steps_per_second": 2.844,
|
| 390 |
"step": 1000
|
| 391 |
}
|
| 392 |
],
|
checkpoint-200/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:392289cbc34e2d29eec5e104d526387f07b7ed0c30e1125b854b9082c48d87ae
|
| 3 |
size 124209532
|
checkpoint-200/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0469a16e48630ed021703d9a6d804f97b9ff7f06a86be78bc4446a856c7b7094
|
| 3 |
size 248356555
|
checkpoint-200/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 200,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-200",
|
| 5 |
"epoch": 2.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,78 +11,78 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
}
|
| 88 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 200,
|
| 3 |
+
"best_metric": 0.6019400298029258,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-200",
|
| 5 |
"epoch": 2.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
}
|
| 88 |
],
|
checkpoint-300/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62ba7376e446b67312c31798248cb91d858ee7294eba002624ea87caf691c39a
|
| 3 |
size 124209532
|
checkpoint-300/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:328cc8c79225c39496cb4794b3c339da79925ae706f6dabcaa5d9bacc79e1a20
|
| 3 |
size 248356555
|
checkpoint-300/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 300,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-300",
|
| 5 |
"epoch": 3.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,116 +11,116 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
}
|
| 126 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 300,
|
| 3 |
+
"best_metric": 0.6144931842511224,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-300",
|
| 5 |
"epoch": 3.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
}
|
| 126 |
],
|
checkpoint-400/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be938384f1b3ee5f32bb7c7b722086a26a46481a74bae9acf609449ed8c7a469
|
| 3 |
size 124209532
|
checkpoint-400/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c38e382f91215aad8b16124472fe59fce773cb90ba2c558aebd87247c354ed37
|
| 3 |
size 248356555
|
checkpoint-400/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 400,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-400",
|
| 5 |
"epoch": 4.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,154 +11,154 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
}
|
| 164 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 400,
|
| 3 |
+
"best_metric": 0.6718166921867913,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-400",
|
| 5 |
"epoch": 4.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
}
|
| 164 |
],
|
checkpoint-500/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f3ff4f5322db23faaf01ca762f093fbd616b3cfd0e15f4f82b8b6af82dd92bc2
|
| 3 |
size 124209532
|
checkpoint-500/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91a6ec708e4d9fad3f41dbdf353955c0a21858f35905b7cb2ae81d6ed41c0f9a
|
| 3 |
size 248356555
|
checkpoint-500/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 500,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-500",
|
| 5 |
"epoch": 5.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,192 +11,192 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
-
"grad_norm":
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
-
"grad_norm":
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
-
"grad_norm":
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
-
"grad_norm":
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
-
"eval_dice": 0.
|
| 195 |
-
"eval_iou": 0.
|
| 196 |
-
"eval_loss": 0.
|
| 197 |
-
"eval_runtime": 4.
|
| 198 |
-
"eval_samples_per_second": 22.
|
| 199 |
-
"eval_steps_per_second": 2.
|
| 200 |
"step": 500
|
| 201 |
}
|
| 202 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 500,
|
| 3 |
+
"best_metric": 0.6808156703817527,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-500",
|
| 5 |
"epoch": 5.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
+
"grad_norm": 2.91507625579834,
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
+
"loss": 0.817935562133789,
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
+
"grad_norm": 6.0939226150512695,
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
+
"loss": 0.8120758056640625,
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
+
"grad_norm": 2.495445728302002,
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
+
"loss": 0.7950348663330078,
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
+
"grad_norm": 4.1540961265563965,
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
+
"loss": 0.8125821685791016,
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
+
"eval_dice": 0.6808156703817527,
|
| 195 |
+
"eval_iou": 0.5160885243741951,
|
| 196 |
+
"eval_loss": 0.8397173285484314,
|
| 197 |
+
"eval_runtime": 4.4587,
|
| 198 |
+
"eval_samples_per_second": 22.428,
|
| 199 |
+
"eval_steps_per_second": 2.916,
|
| 200 |
"step": 500
|
| 201 |
}
|
| 202 |
],
|
checkpoint-600/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51b3f2009c156cb159536c5b11040edda05684c2c676aa9788d0bac50e7682f4
|
| 3 |
size 124209532
|
checkpoint-600/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53868c5cd3aee3c5ccfa2f43e7345a86ca266ef55cc46dafeb01faf9478bcf1a
|
| 3 |
size 248356555
|
checkpoint-600/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 600,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-600",
|
| 5 |
"epoch": 6.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,230 +11,230 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
-
"grad_norm":
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
-
"grad_norm":
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
-
"grad_norm":
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
-
"grad_norm":
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
-
"eval_dice": 0.
|
| 195 |
-
"eval_iou": 0.
|
| 196 |
-
"eval_loss": 0.
|
| 197 |
-
"eval_runtime": 4.
|
| 198 |
-
"eval_samples_per_second": 22.
|
| 199 |
-
"eval_steps_per_second": 2.
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
-
"grad_norm": 9.
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
-
"grad_norm": 3.
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
-
"loss": 0.
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
-
"grad_norm":
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
-
"grad_norm": 3.
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
-
"eval_dice": 0.
|
| 233 |
-
"eval_iou": 0.
|
| 234 |
-
"eval_loss": 0.
|
| 235 |
-
"eval_runtime": 4.
|
| 236 |
-
"eval_samples_per_second": 22.
|
| 237 |
-
"eval_steps_per_second": 2.
|
| 238 |
"step": 600
|
| 239 |
}
|
| 240 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 600,
|
| 3 |
+
"best_metric": 0.7141547601985387,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-600",
|
| 5 |
"epoch": 6.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
+
"grad_norm": 2.91507625579834,
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
+
"loss": 0.817935562133789,
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
+
"grad_norm": 6.0939226150512695,
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
+
"loss": 0.8120758056640625,
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
+
"grad_norm": 2.495445728302002,
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
+
"loss": 0.7950348663330078,
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
+
"grad_norm": 4.1540961265563965,
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
+
"loss": 0.8125821685791016,
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
+
"eval_dice": 0.6808156703817527,
|
| 195 |
+
"eval_iou": 0.5160885243741951,
|
| 196 |
+
"eval_loss": 0.8397173285484314,
|
| 197 |
+
"eval_runtime": 4.4587,
|
| 198 |
+
"eval_samples_per_second": 22.428,
|
| 199 |
+
"eval_steps_per_second": 2.916,
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
+
"grad_norm": 9.038034439086914,
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
+
"loss": 0.7783580780029297,
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
+
"grad_norm": 3.4795265197753906,
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
+
"loss": 0.7784259796142579,
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
+
"grad_norm": 2.7996904850006104,
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
+
"loss": 0.7458997344970704,
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
+
"grad_norm": 3.0424861907958984,
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
+
"loss": 0.7311380767822265,
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
+
"eval_dice": 0.7141547601985387,
|
| 233 |
+
"eval_iou": 0.5553973025470366,
|
| 234 |
+
"eval_loss": 0.7741679549217224,
|
| 235 |
+
"eval_runtime": 4.4797,
|
| 236 |
+
"eval_samples_per_second": 22.323,
|
| 237 |
+
"eval_steps_per_second": 2.902,
|
| 238 |
"step": 600
|
| 239 |
}
|
| 240 |
],
|
checkpoint-700/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c78dc02da1051d31752a22275033ac4ba561afbe4c8d737f8f0f8d91a907bc2
|
| 3 |
size 124209532
|
checkpoint-700/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de3550f641e0214883fa72b91576b70ae140154e379e082f499c788c728ed41c
|
| 3 |
size 248356555
|
checkpoint-700/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 700,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-700",
|
| 5 |
"epoch": 7.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,268 +11,268 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
-
"grad_norm":
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
-
"grad_norm":
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
-
"grad_norm":
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
-
"grad_norm":
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
-
"eval_dice": 0.
|
| 195 |
-
"eval_iou": 0.
|
| 196 |
-
"eval_loss": 0.
|
| 197 |
-
"eval_runtime": 4.
|
| 198 |
-
"eval_samples_per_second": 22.
|
| 199 |
-
"eval_steps_per_second": 2.
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
-
"grad_norm": 9.
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
-
"grad_norm": 3.
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
-
"loss": 0.
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
-
"grad_norm":
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
-
"grad_norm": 3.
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
-
"eval_dice": 0.
|
| 233 |
-
"eval_iou": 0.
|
| 234 |
-
"eval_loss": 0.
|
| 235 |
-
"eval_runtime": 4.
|
| 236 |
-
"eval_samples_per_second": 22.
|
| 237 |
-
"eval_steps_per_second": 2.
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
-
"grad_norm":
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
-
"grad_norm":
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
-
"grad_norm":
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
-
"grad_norm":
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
-
"eval_dice": 0.
|
| 271 |
-
"eval_iou": 0.
|
| 272 |
-
"eval_loss": 0.
|
| 273 |
-
"eval_runtime": 4.
|
| 274 |
-
"eval_samples_per_second":
|
| 275 |
-
"eval_steps_per_second": 2.
|
| 276 |
"step": 700
|
| 277 |
}
|
| 278 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 700,
|
| 3 |
+
"best_metric": 0.7389087921045768,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-700",
|
| 5 |
"epoch": 7.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
+
"grad_norm": 2.91507625579834,
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
+
"loss": 0.817935562133789,
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
+
"grad_norm": 6.0939226150512695,
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
+
"loss": 0.8120758056640625,
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
+
"grad_norm": 2.495445728302002,
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
+
"loss": 0.7950348663330078,
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
+
"grad_norm": 4.1540961265563965,
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
+
"loss": 0.8125821685791016,
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
+
"eval_dice": 0.6808156703817527,
|
| 195 |
+
"eval_iou": 0.5160885243741951,
|
| 196 |
+
"eval_loss": 0.8397173285484314,
|
| 197 |
+
"eval_runtime": 4.4587,
|
| 198 |
+
"eval_samples_per_second": 22.428,
|
| 199 |
+
"eval_steps_per_second": 2.916,
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
+
"grad_norm": 9.038034439086914,
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
+
"loss": 0.7783580780029297,
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
+
"grad_norm": 3.4795265197753906,
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
+
"loss": 0.7784259796142579,
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
+
"grad_norm": 2.7996904850006104,
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
+
"loss": 0.7458997344970704,
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
+
"grad_norm": 3.0424861907958984,
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
+
"loss": 0.7311380767822265,
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
+
"eval_dice": 0.7141547601985387,
|
| 233 |
+
"eval_iou": 0.5553973025470366,
|
| 234 |
+
"eval_loss": 0.7741679549217224,
|
| 235 |
+
"eval_runtime": 4.4797,
|
| 236 |
+
"eval_samples_per_second": 22.323,
|
| 237 |
+
"eval_steps_per_second": 2.902,
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
+
"grad_norm": 2.969958782196045,
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
+
"loss": 0.7293343353271484,
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
+
"grad_norm": 3.2517130374908447,
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
+
"loss": 0.7013224029541015,
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
+
"grad_norm": 2.587013006210327,
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
+
"loss": 0.6944268798828125,
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
+
"grad_norm": 2.868363380432129,
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
+
"loss": 0.7179935455322266,
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
+
"eval_dice": 0.7389087921045768,
|
| 271 |
+
"eval_iou": 0.5859282924237521,
|
| 272 |
+
"eval_loss": 0.7406055927276611,
|
| 273 |
+
"eval_runtime": 4.5995,
|
| 274 |
+
"eval_samples_per_second": 21.741,
|
| 275 |
+
"eval_steps_per_second": 2.826,
|
| 276 |
"step": 700
|
| 277 |
}
|
| 278 |
],
|
checkpoint-800/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e20f3d0a8d25991beb05f02ea00daf969984dde673ce9e16a0d31ac088fed91
|
| 3 |
size 124209532
|
checkpoint-800/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:38e7c7189fbf51523d950ece45ab5127c9717a309d8c8c4e8d1bbcfc48e23030
|
| 3 |
size 248356555
|
checkpoint-800/trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"best_global_step":
|
| 3 |
-
"best_metric": 0.
|
| 4 |
-
"best_model_checkpoint": "./kvasir-unet/checkpoint-
|
| 5 |
"epoch": 8.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
"global_step": 800,
|
|
@@ -11,306 +11,306 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
-
"grad_norm":
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
-
"grad_norm":
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
-
"grad_norm":
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
-
"grad_norm":
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
-
"eval_dice": 0.
|
| 195 |
-
"eval_iou": 0.
|
| 196 |
-
"eval_loss": 0.
|
| 197 |
-
"eval_runtime": 4.
|
| 198 |
-
"eval_samples_per_second": 22.
|
| 199 |
-
"eval_steps_per_second": 2.
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
-
"grad_norm": 9.
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
-
"grad_norm": 3.
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
-
"loss": 0.
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
-
"grad_norm":
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
-
"grad_norm": 3.
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
-
"eval_dice": 0.
|
| 233 |
-
"eval_iou": 0.
|
| 234 |
-
"eval_loss": 0.
|
| 235 |
-
"eval_runtime": 4.
|
| 236 |
-
"eval_samples_per_second": 22.
|
| 237 |
-
"eval_steps_per_second": 2.
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
-
"grad_norm":
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
-
"grad_norm":
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
-
"grad_norm":
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
-
"grad_norm":
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
-
"eval_dice": 0.
|
| 271 |
-
"eval_iou": 0.
|
| 272 |
-
"eval_loss": 0.
|
| 273 |
-
"eval_runtime": 4.
|
| 274 |
-
"eval_samples_per_second":
|
| 275 |
-
"eval_steps_per_second": 2.
|
| 276 |
"step": 700
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 7.25,
|
| 280 |
-
"grad_norm":
|
| 281 |
"learning_rate": 2.7600000000000003e-05,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 725
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 7.5,
|
| 287 |
-
"grad_norm": 1.
|
| 288 |
"learning_rate": 2.51e-05,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 750
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 7.75,
|
| 294 |
-
"grad_norm": 4.
|
| 295 |
"learning_rate": 2.26e-05,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 775
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 8.0,
|
| 301 |
-
"grad_norm": 3.
|
| 302 |
"learning_rate": 2.01e-05,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 800
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 8.0,
|
| 308 |
-
"eval_dice": 0.
|
| 309 |
-
"eval_iou": 0.
|
| 310 |
-
"eval_loss": 0.
|
| 311 |
-
"eval_runtime": 4.
|
| 312 |
-
"eval_samples_per_second": 21.
|
| 313 |
-
"eval_steps_per_second": 2.
|
| 314 |
"step": 800
|
| 315 |
}
|
| 316 |
],
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_global_step": 700,
|
| 3 |
+
"best_metric": 0.7389087921045768,
|
| 4 |
+
"best_model_checkpoint": "./kvasir-unet/checkpoint-700",
|
| 5 |
"epoch": 8.0,
|
| 6 |
"eval_steps": 500,
|
| 7 |
"global_step": 800,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
+
"grad_norm": 2.91507625579834,
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
+
"loss": 0.817935562133789,
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
+
"grad_norm": 6.0939226150512695,
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
+
"loss": 0.8120758056640625,
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
+
"grad_norm": 2.495445728302002,
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
+
"loss": 0.7950348663330078,
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
+
"grad_norm": 4.1540961265563965,
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
+
"loss": 0.8125821685791016,
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
+
"eval_dice": 0.6808156703817527,
|
| 195 |
+
"eval_iou": 0.5160885243741951,
|
| 196 |
+
"eval_loss": 0.8397173285484314,
|
| 197 |
+
"eval_runtime": 4.4587,
|
| 198 |
+
"eval_samples_per_second": 22.428,
|
| 199 |
+
"eval_steps_per_second": 2.916,
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
+
"grad_norm": 9.038034439086914,
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
+
"loss": 0.7783580780029297,
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
+
"grad_norm": 3.4795265197753906,
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
+
"loss": 0.7784259796142579,
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
+
"grad_norm": 2.7996904850006104,
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
+
"loss": 0.7458997344970704,
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
+
"grad_norm": 3.0424861907958984,
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
+
"loss": 0.7311380767822265,
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
+
"eval_dice": 0.7141547601985387,
|
| 233 |
+
"eval_iou": 0.5553973025470366,
|
| 234 |
+
"eval_loss": 0.7741679549217224,
|
| 235 |
+
"eval_runtime": 4.4797,
|
| 236 |
+
"eval_samples_per_second": 22.323,
|
| 237 |
+
"eval_steps_per_second": 2.902,
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
+
"grad_norm": 2.969958782196045,
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
+
"loss": 0.7293343353271484,
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
+
"grad_norm": 3.2517130374908447,
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
+
"loss": 0.7013224029541015,
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
+
"grad_norm": 2.587013006210327,
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
+
"loss": 0.6944268798828125,
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
+
"grad_norm": 2.868363380432129,
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
+
"loss": 0.7179935455322266,
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
+
"eval_dice": 0.7389087921045768,
|
| 271 |
+
"eval_iou": 0.5859282924237521,
|
| 272 |
+
"eval_loss": 0.7406055927276611,
|
| 273 |
+
"eval_runtime": 4.5995,
|
| 274 |
+
"eval_samples_per_second": 21.741,
|
| 275 |
+
"eval_steps_per_second": 2.826,
|
| 276 |
"step": 700
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 7.25,
|
| 280 |
+
"grad_norm": 4.949324131011963,
|
| 281 |
"learning_rate": 2.7600000000000003e-05,
|
| 282 |
+
"loss": 0.7042760467529297,
|
| 283 |
"step": 725
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 7.5,
|
| 287 |
+
"grad_norm": 1.5597704648971558,
|
| 288 |
"learning_rate": 2.51e-05,
|
| 289 |
+
"loss": 0.6764234161376953,
|
| 290 |
"step": 750
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 7.75,
|
| 294 |
+
"grad_norm": 4.044253349304199,
|
| 295 |
"learning_rate": 2.26e-05,
|
| 296 |
+
"loss": 0.6907909393310547,
|
| 297 |
"step": 775
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 8.0,
|
| 301 |
+
"grad_norm": 3.0667364597320557,
|
| 302 |
"learning_rate": 2.01e-05,
|
| 303 |
+
"loss": 0.6335317611694335,
|
| 304 |
"step": 800
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 8.0,
|
| 308 |
+
"eval_dice": 0.7353989116647303,
|
| 309 |
+
"eval_iou": 0.5815265551170842,
|
| 310 |
+
"eval_loss": 0.7389959692955017,
|
| 311 |
+
"eval_runtime": 4.6414,
|
| 312 |
+
"eval_samples_per_second": 21.545,
|
| 313 |
+
"eval_steps_per_second": 2.801,
|
| 314 |
"step": 800
|
| 315 |
}
|
| 316 |
],
|
checkpoint-900/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209532
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:967f01bab1ffb76a03b02e6b4f4fed6b68e97a04ff1785a9bc54ce04852a718e
|
| 3 |
size 124209532
|
checkpoint-900/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 248356555
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3e771f08d4ecd7de24e68a978dee208ba08d7aefcb6f59b6be835568ecaf03a0
|
| 3 |
size 248356555
|
checkpoint-900/trainer_state.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"best_global_step": 900,
|
| 3 |
-
"best_metric": 0.
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-900",
|
| 5 |
"epoch": 9.0,
|
| 6 |
"eval_steps": 500,
|
|
@@ -11,344 +11,344 @@
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
-
"grad_norm": 1.
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
-
"loss": 1.
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
-
"grad_norm": 1.
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
-
"loss": 1.
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
-
"grad_norm": 1.
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
-
"loss": 1.
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
-
"grad_norm":
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
-
"loss": 1.
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
-
"eval_dice": 0.
|
| 43 |
-
"eval_iou": 0.
|
| 44 |
-
"eval_loss": 1.
|
| 45 |
-
"eval_runtime": 4.
|
| 46 |
-
"eval_samples_per_second": 22.
|
| 47 |
-
"eval_steps_per_second": 2.
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
-
"grad_norm": 2.
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
-
"loss":
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
-
"grad_norm": 2.
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
-
"loss":
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
-
"grad_norm": 2.
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
-
"loss":
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
-
"grad_norm": 3.
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
-
"loss": 0.
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
-
"eval_dice": 0.
|
| 81 |
-
"eval_iou": 0.
|
| 82 |
-
"eval_loss": 0.
|
| 83 |
-
"eval_runtime": 4.
|
| 84 |
-
"eval_samples_per_second":
|
| 85 |
-
"eval_steps_per_second":
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
-
"grad_norm": 4.
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
-
"grad_norm": 1.
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
-
"loss": 0.
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
-
"loss": 0.
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
-
"grad_norm": 1.
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
-
"eval_dice": 0.
|
| 119 |
-
"eval_iou": 0.
|
| 120 |
-
"eval_loss": 0.
|
| 121 |
-
"eval_runtime": 4.
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 2.
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
-
"grad_norm":
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
-
"loss": 0.
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
-
"grad_norm":
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
-
"grad_norm":
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
-
"loss": 0.
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
-
"grad_norm": 4.
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
-
"eval_dice": 0.
|
| 157 |
-
"eval_iou": 0.
|
| 158 |
-
"eval_loss": 0.
|
| 159 |
-
"eval_runtime": 4.
|
| 160 |
-
"eval_samples_per_second": 22.
|
| 161 |
-
"eval_steps_per_second": 2.
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
-
"grad_norm":
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
-
"grad_norm":
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
-
"grad_norm":
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
-
"grad_norm":
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
-
"loss": 0.
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
-
"eval_dice": 0.
|
| 195 |
-
"eval_iou": 0.
|
| 196 |
-
"eval_loss": 0.
|
| 197 |
-
"eval_runtime": 4.
|
| 198 |
-
"eval_samples_per_second": 22.
|
| 199 |
-
"eval_steps_per_second": 2.
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
-
"grad_norm": 9.
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
-
"grad_norm": 3.
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
-
"loss": 0.
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
-
"grad_norm":
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
-
"loss": 0.
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
-
"grad_norm": 3.
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
-
"eval_dice": 0.
|
| 233 |
-
"eval_iou": 0.
|
| 234 |
-
"eval_loss": 0.
|
| 235 |
-
"eval_runtime": 4.
|
| 236 |
-
"eval_samples_per_second": 22.
|
| 237 |
-
"eval_steps_per_second": 2.
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
-
"grad_norm":
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
-
"grad_norm":
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
-
"grad_norm":
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
-
"loss": 0.
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
-
"grad_norm":
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
-
"loss": 0.
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
-
"eval_dice": 0.
|
| 271 |
-
"eval_iou": 0.
|
| 272 |
-
"eval_loss": 0.
|
| 273 |
-
"eval_runtime": 4.
|
| 274 |
-
"eval_samples_per_second":
|
| 275 |
-
"eval_steps_per_second": 2.
|
| 276 |
"step": 700
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 7.25,
|
| 280 |
-
"grad_norm":
|
| 281 |
"learning_rate": 2.7600000000000003e-05,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 725
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 7.5,
|
| 287 |
-
"grad_norm": 1.
|
| 288 |
"learning_rate": 2.51e-05,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 750
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 7.75,
|
| 294 |
-
"grad_norm": 4.
|
| 295 |
"learning_rate": 2.26e-05,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 775
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 8.0,
|
| 301 |
-
"grad_norm": 3.
|
| 302 |
"learning_rate": 2.01e-05,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 800
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 8.0,
|
| 308 |
-
"eval_dice": 0.
|
| 309 |
-
"eval_iou": 0.
|
| 310 |
-
"eval_loss": 0.
|
| 311 |
-
"eval_runtime": 4.
|
| 312 |
-
"eval_samples_per_second": 21.
|
| 313 |
-
"eval_steps_per_second": 2.
|
| 314 |
"step": 800
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 8.25,
|
| 318 |
-
"grad_norm":
|
| 319 |
"learning_rate": 1.76e-05,
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 825
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 8.5,
|
| 325 |
-
"grad_norm":
|
| 326 |
"learning_rate": 1.51e-05,
|
| 327 |
-
"loss": 0.
|
| 328 |
"step": 850
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 8.75,
|
| 332 |
-
"grad_norm":
|
| 333 |
"learning_rate": 1.2600000000000001e-05,
|
| 334 |
-
"loss": 0.
|
| 335 |
"step": 875
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 9.0,
|
| 339 |
-
"grad_norm":
|
| 340 |
"learning_rate": 1.0100000000000002e-05,
|
| 341 |
-
"loss": 0.
|
| 342 |
"step": 900
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 9.0,
|
| 346 |
-
"eval_dice": 0.
|
| 347 |
-
"eval_iou": 0.
|
| 348 |
-
"eval_loss": 0.
|
| 349 |
-
"eval_runtime": 4.
|
| 350 |
-
"eval_samples_per_second":
|
| 351 |
-
"eval_steps_per_second": 2.
|
| 352 |
"step": 900
|
| 353 |
}
|
| 354 |
],
|
|
|
|
| 1 |
{
|
| 2 |
"best_global_step": 900,
|
| 3 |
+
"best_metric": 0.7648601527792854,
|
| 4 |
"best_model_checkpoint": "./kvasir-unet/checkpoint-900",
|
| 5 |
"epoch": 9.0,
|
| 6 |
"eval_steps": 500,
|
|
|
|
| 11 |
"log_history": [
|
| 12 |
{
|
| 13 |
"epoch": 0.25,
|
| 14 |
+
"grad_norm": 1.2136750221252441,
|
| 15 |
"learning_rate": 9.76e-05,
|
| 16 |
+
"loss": 1.2960072326660157,
|
| 17 |
"step": 25
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 0.5,
|
| 21 |
+
"grad_norm": 1.6375166177749634,
|
| 22 |
"learning_rate": 9.51e-05,
|
| 23 |
+
"loss": 1.1861843109130858,
|
| 24 |
"step": 50
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.75,
|
| 28 |
+
"grad_norm": 1.6723055839538574,
|
| 29 |
"learning_rate": 9.260000000000001e-05,
|
| 30 |
+
"loss": 1.146585922241211,
|
| 31 |
"step": 75
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 1.0,
|
| 35 |
+
"grad_norm": 3.804276943206787,
|
| 36 |
"learning_rate": 9.010000000000001e-05,
|
| 37 |
+
"loss": 1.067328872680664,
|
| 38 |
"step": 100
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.0,
|
| 42 |
+
"eval_dice": 0.5377293432044794,
|
| 43 |
+
"eval_iou": 0.36773605249770436,
|
| 44 |
+
"eval_loss": 1.0997637510299683,
|
| 45 |
+
"eval_runtime": 4.4666,
|
| 46 |
+
"eval_samples_per_second": 22.388,
|
| 47 |
+
"eval_steps_per_second": 2.91,
|
| 48 |
"step": 100
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.25,
|
| 52 |
+
"grad_norm": 2.740241050720215,
|
| 53 |
"learning_rate": 8.76e-05,
|
| 54 |
+
"loss": 1.0599410247802734,
|
| 55 |
"step": 125
|
| 56 |
},
|
| 57 |
{
|
| 58 |
"epoch": 1.5,
|
| 59 |
+
"grad_norm": 2.035609722137451,
|
| 60 |
"learning_rate": 8.510000000000001e-05,
|
| 61 |
+
"loss": 1.025189437866211,
|
| 62 |
"step": 150
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 1.75,
|
| 66 |
+
"grad_norm": 2.5915679931640625,
|
| 67 |
"learning_rate": 8.26e-05,
|
| 68 |
+
"loss": 1.0165634155273438,
|
| 69 |
"step": 175
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 2.0,
|
| 73 |
+
"grad_norm": 3.354675054550171,
|
| 74 |
"learning_rate": 8.010000000000001e-05,
|
| 75 |
+
"loss": 0.9938540649414063,
|
| 76 |
"step": 200
|
| 77 |
},
|
| 78 |
{
|
| 79 |
"epoch": 2.0,
|
| 80 |
+
"eval_dice": 0.6019400298029258,
|
| 81 |
+
"eval_iou": 0.4305539757991033,
|
| 82 |
+
"eval_loss": 0.9929438233375549,
|
| 83 |
+
"eval_runtime": 4.2981,
|
| 84 |
+
"eval_samples_per_second": 23.266,
|
| 85 |
+
"eval_steps_per_second": 3.025,
|
| 86 |
"step": 200
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 2.25,
|
| 90 |
+
"grad_norm": 4.807499408721924,
|
| 91 |
"learning_rate": 7.76e-05,
|
| 92 |
+
"loss": 0.9634833526611328,
|
| 93 |
"step": 225
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.5,
|
| 97 |
+
"grad_norm": 1.9283570051193237,
|
| 98 |
"learning_rate": 7.510000000000001e-05,
|
| 99 |
+
"loss": 0.9624813842773438,
|
| 100 |
"step": 250
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 2.75,
|
| 104 |
+
"grad_norm": 1.9197851419448853,
|
| 105 |
"learning_rate": 7.26e-05,
|
| 106 |
+
"loss": 0.9242520904541016,
|
| 107 |
"step": 275
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 3.0,
|
| 111 |
+
"grad_norm": 1.9761945009231567,
|
| 112 |
"learning_rate": 7.01e-05,
|
| 113 |
+
"loss": 0.9074368286132812,
|
| 114 |
"step": 300
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.0,
|
| 118 |
+
"eval_dice": 0.6144931842511224,
|
| 119 |
+
"eval_iou": 0.4435152564919117,
|
| 120 |
+
"eval_loss": 0.9622752666473389,
|
| 121 |
+
"eval_runtime": 4.4082,
|
| 122 |
+
"eval_samples_per_second": 22.685,
|
| 123 |
+
"eval_steps_per_second": 2.949,
|
| 124 |
"step": 300
|
| 125 |
},
|
| 126 |
{
|
| 127 |
"epoch": 3.25,
|
| 128 |
+
"grad_norm": 3.118398427963257,
|
| 129 |
"learning_rate": 6.76e-05,
|
| 130 |
+
"loss": 0.9109857177734375,
|
| 131 |
"step": 325
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 3.5,
|
| 135 |
+
"grad_norm": 1.3993834257125854,
|
| 136 |
"learning_rate": 6.510000000000001e-05,
|
| 137 |
+
"loss": 0.8557350158691406,
|
| 138 |
"step": 350
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 3.75,
|
| 142 |
+
"grad_norm": 7.050769805908203,
|
| 143 |
"learning_rate": 6.26e-05,
|
| 144 |
+
"loss": 0.8826413726806641,
|
| 145 |
"step": 375
|
| 146 |
},
|
| 147 |
{
|
| 148 |
"epoch": 4.0,
|
| 149 |
+
"grad_norm": 4.970097541809082,
|
| 150 |
"learning_rate": 6.0100000000000004e-05,
|
| 151 |
+
"loss": 0.8279270172119141,
|
| 152 |
"step": 400
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 4.0,
|
| 156 |
+
"eval_dice": 0.6718166921867913,
|
| 157 |
+
"eval_iou": 0.5058164193098597,
|
| 158 |
+
"eval_loss": 0.8655270934104919,
|
| 159 |
+
"eval_runtime": 4.3994,
|
| 160 |
+
"eval_samples_per_second": 22.731,
|
| 161 |
+
"eval_steps_per_second": 2.955,
|
| 162 |
"step": 400
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 4.25,
|
| 166 |
+
"grad_norm": 2.91507625579834,
|
| 167 |
"learning_rate": 5.76e-05,
|
| 168 |
+
"loss": 0.817935562133789,
|
| 169 |
"step": 425
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 4.5,
|
| 173 |
+
"grad_norm": 6.0939226150512695,
|
| 174 |
"learning_rate": 5.5100000000000004e-05,
|
| 175 |
+
"loss": 0.8120758056640625,
|
| 176 |
"step": 450
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 4.75,
|
| 180 |
+
"grad_norm": 2.495445728302002,
|
| 181 |
"learning_rate": 5.2600000000000005e-05,
|
| 182 |
+
"loss": 0.7950348663330078,
|
| 183 |
"step": 475
|
| 184 |
},
|
| 185 |
{
|
| 186 |
"epoch": 5.0,
|
| 187 |
+
"grad_norm": 4.1540961265563965,
|
| 188 |
"learning_rate": 5.0100000000000005e-05,
|
| 189 |
+
"loss": 0.8125821685791016,
|
| 190 |
"step": 500
|
| 191 |
},
|
| 192 |
{
|
| 193 |
"epoch": 5.0,
|
| 194 |
+
"eval_dice": 0.6808156703817527,
|
| 195 |
+
"eval_iou": 0.5160885243741951,
|
| 196 |
+
"eval_loss": 0.8397173285484314,
|
| 197 |
+
"eval_runtime": 4.4587,
|
| 198 |
+
"eval_samples_per_second": 22.428,
|
| 199 |
+
"eval_steps_per_second": 2.916,
|
| 200 |
"step": 500
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 5.25,
|
| 204 |
+
"grad_norm": 9.038034439086914,
|
| 205 |
"learning_rate": 4.76e-05,
|
| 206 |
+
"loss": 0.7783580780029297,
|
| 207 |
"step": 525
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 5.5,
|
| 211 |
+
"grad_norm": 3.4795265197753906,
|
| 212 |
"learning_rate": 4.5100000000000005e-05,
|
| 213 |
+
"loss": 0.7784259796142579,
|
| 214 |
"step": 550
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 5.75,
|
| 218 |
+
"grad_norm": 2.7996904850006104,
|
| 219 |
"learning_rate": 4.26e-05,
|
| 220 |
+
"loss": 0.7458997344970704,
|
| 221 |
"step": 575
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 6.0,
|
| 225 |
+
"grad_norm": 3.0424861907958984,
|
| 226 |
"learning_rate": 4.0100000000000006e-05,
|
| 227 |
+
"loss": 0.7311380767822265,
|
| 228 |
"step": 600
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 6.0,
|
| 232 |
+
"eval_dice": 0.7141547601985387,
|
| 233 |
+
"eval_iou": 0.5553973025470366,
|
| 234 |
+
"eval_loss": 0.7741679549217224,
|
| 235 |
+
"eval_runtime": 4.4797,
|
| 236 |
+
"eval_samples_per_second": 22.323,
|
| 237 |
+
"eval_steps_per_second": 2.902,
|
| 238 |
"step": 600
|
| 239 |
},
|
| 240 |
{
|
| 241 |
"epoch": 6.25,
|
| 242 |
+
"grad_norm": 2.969958782196045,
|
| 243 |
"learning_rate": 3.76e-05,
|
| 244 |
+
"loss": 0.7293343353271484,
|
| 245 |
"step": 625
|
| 246 |
},
|
| 247 |
{
|
| 248 |
"epoch": 6.5,
|
| 249 |
+
"grad_norm": 3.2517130374908447,
|
| 250 |
"learning_rate": 3.51e-05,
|
| 251 |
+
"loss": 0.7013224029541015,
|
| 252 |
"step": 650
|
| 253 |
},
|
| 254 |
{
|
| 255 |
"epoch": 6.75,
|
| 256 |
+
"grad_norm": 2.587013006210327,
|
| 257 |
"learning_rate": 3.26e-05,
|
| 258 |
+
"loss": 0.6944268798828125,
|
| 259 |
"step": 675
|
| 260 |
},
|
| 261 |
{
|
| 262 |
"epoch": 7.0,
|
| 263 |
+
"grad_norm": 2.868363380432129,
|
| 264 |
"learning_rate": 3.01e-05,
|
| 265 |
+
"loss": 0.7179935455322266,
|
| 266 |
"step": 700
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 7.0,
|
| 270 |
+
"eval_dice": 0.7389087921045768,
|
| 271 |
+
"eval_iou": 0.5859282924237521,
|
| 272 |
+
"eval_loss": 0.7406055927276611,
|
| 273 |
+
"eval_runtime": 4.5995,
|
| 274 |
+
"eval_samples_per_second": 21.741,
|
| 275 |
+
"eval_steps_per_second": 2.826,
|
| 276 |
"step": 700
|
| 277 |
},
|
| 278 |
{
|
| 279 |
"epoch": 7.25,
|
| 280 |
+
"grad_norm": 4.949324131011963,
|
| 281 |
"learning_rate": 2.7600000000000003e-05,
|
| 282 |
+
"loss": 0.7042760467529297,
|
| 283 |
"step": 725
|
| 284 |
},
|
| 285 |
{
|
| 286 |
"epoch": 7.5,
|
| 287 |
+
"grad_norm": 1.5597704648971558,
|
| 288 |
"learning_rate": 2.51e-05,
|
| 289 |
+
"loss": 0.6764234161376953,
|
| 290 |
"step": 750
|
| 291 |
},
|
| 292 |
{
|
| 293 |
"epoch": 7.75,
|
| 294 |
+
"grad_norm": 4.044253349304199,
|
| 295 |
"learning_rate": 2.26e-05,
|
| 296 |
+
"loss": 0.6907909393310547,
|
| 297 |
"step": 775
|
| 298 |
},
|
| 299 |
{
|
| 300 |
"epoch": 8.0,
|
| 301 |
+
"grad_norm": 3.0667364597320557,
|
| 302 |
"learning_rate": 2.01e-05,
|
| 303 |
+
"loss": 0.6335317611694335,
|
| 304 |
"step": 800
|
| 305 |
},
|
| 306 |
{
|
| 307 |
"epoch": 8.0,
|
| 308 |
+
"eval_dice": 0.7353989116647303,
|
| 309 |
+
"eval_iou": 0.5815265551170842,
|
| 310 |
+
"eval_loss": 0.7389959692955017,
|
| 311 |
+
"eval_runtime": 4.6414,
|
| 312 |
+
"eval_samples_per_second": 21.545,
|
| 313 |
+
"eval_steps_per_second": 2.801,
|
| 314 |
"step": 800
|
| 315 |
},
|
| 316 |
{
|
| 317 |
"epoch": 8.25,
|
| 318 |
+
"grad_norm": 8.7631196975708,
|
| 319 |
"learning_rate": 1.76e-05,
|
| 320 |
+
"loss": 0.6445667266845703,
|
| 321 |
"step": 825
|
| 322 |
},
|
| 323 |
{
|
| 324 |
"epoch": 8.5,
|
| 325 |
+
"grad_norm": 3.8787753582000732,
|
| 326 |
"learning_rate": 1.51e-05,
|
| 327 |
+
"loss": 0.6259555816650391,
|
| 328 |
"step": 850
|
| 329 |
},
|
| 330 |
{
|
| 331 |
"epoch": 8.75,
|
| 332 |
+
"grad_norm": 2.609173536300659,
|
| 333 |
"learning_rate": 1.2600000000000001e-05,
|
| 334 |
+
"loss": 0.6285885620117188,
|
| 335 |
"step": 875
|
| 336 |
},
|
| 337 |
{
|
| 338 |
"epoch": 9.0,
|
| 339 |
+
"grad_norm": 3.2756707668304443,
|
| 340 |
"learning_rate": 1.0100000000000002e-05,
|
| 341 |
+
"loss": 0.643200912475586,
|
| 342 |
"step": 900
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 9.0,
|
| 346 |
+
"eval_dice": 0.7648601527792854,
|
| 347 |
+
"eval_iou": 0.6192499971116917,
|
| 348 |
+
"eval_loss": 0.6945236921310425,
|
| 349 |
+
"eval_runtime": 4.6242,
|
| 350 |
+
"eval_samples_per_second": 21.626,
|
| 351 |
+
"eval_steps_per_second": 2.811,
|
| 352 |
"step": 900
|
| 353 |
}
|
| 354 |
],
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 124209500
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e54f2a0721f4b7d6e27ea0849e85ff1b5d3b66e6fdac2d875aa510d4324fe5f7
|
| 3 |
size 124209500
|