File size: 29,436 Bytes
a9eed45 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 |
2025-04-29,17:42:33 | INFO | No latest resume checkpoint found in /mnt/personal/zhudongy/datacomp_results/medium/low_inter_only/checkpoints.
2025-04-29,17:42:35 | INFO | Running in distributed mode with multiple processes. Device: cuda:0.Process (global: 0, local 0), total 2.
2025-04-29,17:42:35 | INFO | Loaded ViT-B-32 model config.
2025-04-29,17:42:36 | INFO | Model:
2025-04-29,17:42:36 | INFO | CLIP(
(visual): VisionTransformer(
(patchnorm_pre_ln): Identity()
(conv1): Conv2d(3, 768, kernel_size=(32, 32), stride=(32, 32), bias=False)
(patch_dropout): Identity()
(ln_pre): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(transformer): Transformer(
(resblocks): ModuleList(
(0): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(1): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(2): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(3): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(4): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(5): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(6): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(7): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(8): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(9): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(10): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
(11): ResidualAttentionBlock(
(ln_1): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=768, out_features=768, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=768, out_features=3072, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=3072, out_features=768, bias=True)
)
(ls_2): Identity()
)
)
)
(ln_post): LayerNorm((768,), eps=1e-05, elementwise_affine=True)
)
(transformer): Transformer(
(resblocks): ModuleList(
(0): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(1): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(2): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(3): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(4): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(5): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(6): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(7): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(8): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(9): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(10): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
(11): ResidualAttentionBlock(
(ln_1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(attn): MultiheadAttention(
(out_proj): NonDynamicallyQuantizableLinear(in_features=512, out_features=512, bias=True)
)
(ls_1): Identity()
(ln_2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
(mlp): Sequential(
(c_fc): Linear(in_features=512, out_features=2048, bias=True)
(gelu): GELU(approximate='none')
(c_proj): Linear(in_features=2048, out_features=512, bias=True)
)
(ls_2): Identity()
)
)
)
(token_embedding): Embedding(49408, 512)
(ln_final): LayerNorm((512,), eps=1e-05, elementwise_affine=True)
)
2025-04-29,17:42:36 | INFO | Params:
2025-04-29,17:42:36 | INFO | accum_freq: 1
2025-04-29,17:42:36 | INFO | aug_cfg: {}
2025-04-29,17:42:36 | INFO | batch_size: 2048
2025-04-29,17:42:36 | INFO | beta1: 0.9
2025-04-29,17:42:36 | INFO | beta2: 0.98
2025-04-29,17:42:36 | INFO | checkpoint_path: /mnt/personal/zhudongy/datacomp_results/medium/low_inter_only/checkpoints
2025-04-29,17:42:36 | INFO | coca_caption_loss_weight: 2.0
2025-04-29,17:42:36 | INFO | coca_contrastive_loss_weight: 1.0
2025-04-29,17:42:36 | INFO | copy_codebase: False
2025-04-29,17:42:36 | INFO | csv_caption_key: title
2025-04-29,17:42:36 | INFO | csv_img_key: filepath
2025-04-29,17:42:36 | INFO | csv_separator:
2025-04-29,17:42:36 | INFO | dataset_resampled: True
2025-04-29,17:42:36 | INFO | dataset_type: webdataset
2025-04-29,17:42:36 | INFO | ddp_static_graph: True
2025-04-29,17:42:36 | INFO | debug: False
2025-04-29,17:42:36 | INFO | delete_previous_checkpoint: False
2025-04-29,17:42:36 | INFO | device: cuda:0
2025-04-29,17:42:36 | INFO | dist_backend: nccl
2025-04-29,17:42:36 | INFO | dist_url: env://
2025-04-29,17:42:36 | INFO | distill: False
2025-04-29,17:42:36 | INFO | distill_model: None
2025-04-29,17:42:36 | INFO | distill_pretrained: None
2025-04-29,17:42:36 | INFO | distributed: True
2025-04-29,17:42:36 | INFO | epochs: 8
2025-04-29,17:42:36 | INFO | epochs_cooldown: None
2025-04-29,17:42:36 | INFO | eps: 1e-06
2025-04-29,17:42:36 | INFO | force_custom_text: False
2025-04-29,17:42:36 | INFO | force_image_size: None
2025-04-29,17:42:36 | INFO | force_patch_dropout: None
2025-04-29,17:42:36 | INFO | force_quick_gelu: False
2025-04-29,17:42:36 | INFO | gather_with_grad: True
2025-04-29,17:42:36 | INFO | grad_checkpointing: True
2025-04-29,17:42:36 | INFO | grad_clip_norm: None
2025-04-29,17:42:36 | INFO | horovod: False
2025-04-29,17:42:36 | INFO | image_mean: None
2025-04-29,17:42:36 | INFO | image_std: None
2025-04-29,17:42:36 | INFO | imagenet_v2: None
2025-04-29,17:42:36 | INFO | imagenet_val: None
2025-04-29,17:42:36 | INFO | local_loss: True
2025-04-29,17:42:36 | INFO | local_rank: 0
2025-04-29,17:42:36 | INFO | lock_image: False
2025-04-29,17:42:36 | INFO | lock_image_freeze_bn_stats: False
2025-04-29,17:42:36 | INFO | lock_image_unlocked_groups: 0
2025-04-29,17:42:36 | INFO | lock_text: False
2025-04-29,17:42:36 | INFO | lock_text_freeze_layer_norm: False
2025-04-29,17:42:36 | INFO | lock_text_unlocked_layers: 0
2025-04-29,17:42:36 | INFO | log_every_n_steps: 100
2025-04-29,17:42:36 | INFO | log_level: 20
2025-04-29,17:42:36 | INFO | log_local: False
2025-04-29,17:42:36 | INFO | log_path: /mnt/personal/zhudongy/datacomp_results/medium/low_inter_only/out.log
2025-04-29,17:42:36 | INFO | logs: /mnt/personal/zhudongy/datacomp_results/medium
2025-04-29,17:42:36 | INFO | lr: 0.0005
2025-04-29,17:42:36 | INFO | lr_cooldown_end: 0.0
2025-04-29,17:42:36 | INFO | lr_cooldown_power: 1.0
2025-04-29,17:42:36 | INFO | lr_scheduler: cosine
2025-04-29,17:42:36 | INFO | model: ViT-B-32
2025-04-29,17:42:36 | INFO | name: low_inter_only
2025-04-29,17:42:36 | INFO | no_set_device_rank: False
2025-04-29,17:42:36 | INFO | precision: amp_bfloat16
2025-04-29,17:42:36 | INFO | pretrained:
2025-04-29,17:42:36 | INFO | pretrained_image: False
2025-04-29,17:42:36 | INFO | rank: 0
2025-04-29,17:42:36 | INFO | remote_sync: None
2025-04-29,17:42:36 | INFO | remote_sync_frequency: 300
2025-04-29,17:42:36 | INFO | remote_sync_protocol: s3
2025-04-29,17:42:36 | INFO | report_to:
2025-04-29,17:42:36 | INFO | resume: None
2025-04-29,17:42:36 | INFO | save_frequency: 0
2025-04-29,17:42:36 | INFO | save_most_recent: True
2025-04-29,17:42:36 | INFO | seed: 0
2025-04-29,17:42:36 | INFO | skip_scheduler: False
2025-04-29,17:42:36 | INFO | tensorboard: False
2025-04-29,17:42:36 | INFO | tensorboard_path:
2025-04-29,17:42:36 | INFO | torchscript: False
2025-04-29,17:42:36 | INFO | trace: False
2025-04-29,17:42:36 | INFO | train_data: /mnt/personal/zhudongy/datacomp-medium/shards/0000{0000..6126}.tar
2025-04-29,17:42:36 | INFO | train_data_upsampling_factors: None
2025-04-29,17:42:36 | INFO | train_num_samples: 16000000
2025-04-29,17:42:36 | INFO | use_bn_sync: False
2025-04-29,17:42:36 | INFO | val_data: None
2025-04-29,17:42:36 | INFO | val_frequency: 1
2025-04-29,17:42:36 | INFO | val_num_samples: None
2025-04-29,17:42:36 | INFO | wandb: False
2025-04-29,17:42:36 | INFO | wandb_notes:
2025-04-29,17:42:36 | INFO | wandb_project_name: open-clip
2025-04-29,17:42:36 | INFO | warmup: 500
2025-04-29,17:42:36 | INFO | wd: 0.2
2025-04-29,17:42:36 | INFO | workers: 16
2025-04-29,17:42:36 | INFO | world_size: 2
2025-04-29,17:42:36 | INFO | zeroshot_frequency: 2
2025-04-29,17:42:36 | INFO | Start epoch 0
2025-04-29,17:43:01 | INFO | Train Epoch: 0 [ 4096/16056320 (0%)] Data (t): 21.925 Batch (t): 24.947, 164.187/s, 82.0936/s/gpu LR: 0.000001 Logit Scale: 14.286 Contrastive_loss: 8.3837 (8.3837) Loss: 8.3837 (8.3837)
2025-04-29,17:43:04 | INFO | Reducer buckets have been rebuilt in this iteration.
2025-04-29,17:47:03 | INFO | Train Epoch: 0 [ 413696/16056320 (3%)] Data (t): 0.367 Batch (t): 2.418, 1701.28/s, 850.642/s/gpu LR: 0.000101 Logit Scale: 14.261 Contrastive_loss: 8.1890 (8.2863) Loss: 8.1890 (8.2863)
2025-04-29,17:51:07 | INFO | Train Epoch: 0 [ 823296/16056320 (5%)] Data (t): 0.400 Batch (t): 2.442, 1697.43/s, 848.716/s/gpu LR: 0.000201 Logit Scale: 14.237 Contrastive_loss: 8.0558 (8.2095) Loss: 8.0558 (8.2095)
2025-04-29,17:55:15 | INFO | Train Epoch: 0 [ 1232896/16056320 (8%)] Data (t): 0.436 Batch (t): 2.479, 1693.55/s, 846.774/s/gpu LR: 0.000301 Logit Scale: 14.210 Contrastive_loss: 7.9360 (8.1411) Loss: 7.9360 (8.1411)
2025-04-29,17:59:16 | INFO | Train Epoch: 0 [ 1642496/16056320 (10%)] Data (t): 0.372 Batch (t): 2.411, 1693.62/s, 846.809/s/gpu LR: 0.000401 Logit Scale: 14.185 Contrastive_loss: 7.8394 (8.0808) Loss: 7.8394 (8.0808)
2025-04-29,18:03:17 | INFO | Train Epoch: 0 [ 2052096/16056320 (13%)] Data (t): 0.364 Batch (t): 2.407, 1698.74/s, 849.369/s/gpu LR: 0.000500 Logit Scale: 14.182 Contrastive_loss: 7.7843 (8.0314) Loss: 7.7843 (8.0314)
2025-04-29,18:07:18 | INFO | Train Epoch: 0 [ 2461696/16056320 (15%)] Data (t): 0.369 Batch (t): 2.413, 1692.95/s, 846.474/s/gpu LR: 0.000500 Logit Scale: 14.201 Contrastive_loss: 7.6211 (7.9728) Loss: 7.6211 (7.9728)
2025-04-29,18:11:19 | INFO | Train Epoch: 0 [ 2871296/16056320 (18%)] Data (t): 0.365 Batch (t): 2.407, 1707.98/s, 853.988/s/gpu LR: 0.000500 Logit Scale: 14.260 Contrastive_loss: 7.5299 (7.9174) Loss: 7.5299 (7.9174)
2025-04-29,18:15:20 | INFO | Train Epoch: 0 [ 3280896/16056320 (20%)] Data (t): 0.369 Batch (t): 2.410, 1690.42/s, 845.211/s/gpu LR: 0.000500 Logit Scale: 14.340 Contrastive_loss: 7.4407 (7.8644) Loss: 7.4407 (7.8644)
2025-04-29,18:19:24 | INFO | Train Epoch: 0 [ 3690496/16056320 (23%)] Data (t): 0.393 Batch (t): 2.436, 1701.12/s, 850.559/s/gpu LR: 0.000500 Logit Scale: 14.441 Contrastive_loss: 7.4649 (7.8245) Loss: 7.4649 (7.8245)
2025-04-29,18:23:25 | INFO | Train Epoch: 0 [ 4100096/16056320 (26%)] Data (t): 0.378 Batch (t): 2.417, 1712.22/s, 856.110/s/gpu LR: 0.000500 Logit Scale: 14.582 Contrastive_loss: 7.3469 (7.7811) Loss: 7.3469 (7.7811)
2025-04-29,18:27:27 | INFO | Train Epoch: 0 [ 4509696/16056320 (28%)] Data (t): 0.372 Batch (t): 2.413, 1699.89/s, 849.943/s/gpu LR: 0.000500 Logit Scale: 14.727 Contrastive_loss: 7.2976 (7.7408) Loss: 7.2976 (7.7408)
2025-04-29,18:31:29 | INFO | Train Epoch: 0 [ 4919296/16056320 (31%)] Data (t): 0.378 Batch (t): 2.418, 1680.76/s, 840.382/s/gpu LR: 0.000499 Logit Scale: 14.942 Contrastive_loss: 7.2187 (7.7006) Loss: 7.2187 (7.7006)
2025-04-29,18:35:30 | INFO | Train Epoch: 0 [ 5328896/16056320 (33%)] Data (t): 0.375 Batch (t): 2.417, 1677.33/s, 838.665/s/gpu LR: 0.000499 Logit Scale: 15.128 Contrastive_loss: 7.1773 (7.6632) Loss: 7.1773 (7.6632)
2025-04-29,18:39:32 | INFO | Train Epoch: 0 [ 5738496/16056320 (36%)] Data (t): 0.373 Batch (t): 2.416, 1698.13/s, 849.067/s/gpu LR: 0.000499 Logit Scale: 15.344 Contrastive_loss: 7.0983 (7.6256) Loss: 7.0983 (7.6256)
2025-04-29,18:43:35 | INFO | Train Epoch: 0 [ 6148096/16056320 (38%)] Data (t): 0.390 Batch (t): 2.433, 1681.07/s, 840.537/s/gpu LR: 0.000499 Logit Scale: 15.595 Contrastive_loss: 7.0212 (7.5878) Loss: 7.0212 (7.5878)
2025-04-29,18:47:38 | INFO | Train Epoch: 0 [ 6557696/16056320 (41%)] Data (t): 0.390 Batch (t): 2.430, 1683.60/s, 841.802/s/gpu LR: 0.000498 Logit Scale: 15.826 Contrastive_loss: 6.9587 (7.5508) Loss: 6.9587 (7.5508)
2025-04-29,18:51:40 | INFO | Train Epoch: 0 [ 6967296/16056320 (43%)] Data (t): 0.378 Batch (t): 2.418, 1698.91/s, 849.455/s/gpu LR: 0.000498 Logit Scale: 16.115 Contrastive_loss: 6.8671 (7.5128) Loss: 6.8671 (7.5128)
2025-04-29,18:55:42 | INFO | Train Epoch: 0 [ 7376896/16056320 (46%)] Data (t): 0.380 Batch (t): 2.423, 1679.70/s, 839.848/s/gpu LR: 0.000498 Logit Scale: 16.413 Contrastive_loss: 6.9072 (7.4809) Loss: 6.9072 (7.4809)
2025-04-29,18:59:46 | INFO | Train Epoch: 0 [ 7786496/16056320 (48%)] Data (t): 0.390 Batch (t): 2.432, 1697.74/s, 848.868/s/gpu LR: 0.000497 Logit Scale: 16.734 Contrastive_loss: 6.8222 (7.4480) Loss: 6.8222 (7.4480)
2025-04-29,19:03:50 | INFO | Train Epoch: 0 [ 8196096/16056320 (51%)] Data (t): 0.392 Batch (t): 2.440, 1684.68/s, 842.340/s/gpu LR: 0.000497 Logit Scale: 17.017 Contrastive_loss: 6.7487 (7.4147) Loss: 6.7487 (7.4147)
2025-04-29,19:07:55 | INFO | Train Epoch: 0 [ 8605696/16056320 (54%)] Data (t): 0.414 Batch (t): 2.455, 1699.34/s, 849.669/s/gpu LR: 0.000497 Logit Scale: 17.283 Contrastive_loss: 6.6749 (7.3811) Loss: 6.6749 (7.3811)
2025-04-29,19:11:58 | INFO | Train Epoch: 0 [ 9015296/16056320 (56%)] Data (t): 0.384 Batch (t): 2.425, 1704.23/s, 852.116/s/gpu LR: 0.000496 Logit Scale: 17.559 Contrastive_loss: 6.5998 (7.3471) Loss: 6.5998 (7.3471)
2025-04-29,19:16:01 | INFO | Train Epoch: 0 [ 9424896/16056320 (59%)] Data (t): 0.387 Batch (t): 2.438, 1666.65/s, 833.323/s/gpu LR: 0.000496 Logit Scale: 17.832 Contrastive_loss: 6.5839 (7.3153) Loss: 6.5839 (7.3153)
2025-04-29,19:20:05 | INFO | Train Epoch: 0 [ 9834496/16056320 (61%)] Data (t): 0.393 Batch (t): 2.432, 1699.78/s, 849.890/s/gpu LR: 0.000495 Logit Scale: 18.053 Contrastive_loss: 6.4805 (7.2819) Loss: 6.4805 (7.2819)
2025-04-29,19:24:07 | INFO | Train Epoch: 0 [10244096/16056320 (64%)] Data (t): 0.382 Batch (t): 2.425, 1685.49/s, 842.747/s/gpu LR: 0.000495 Logit Scale: 18.360 Contrastive_loss: 6.5788 (7.2549) Loss: 6.5788 (7.2549)
2025-04-29,19:28:10 | INFO | Train Epoch: 0 [10653696/16056320 (66%)] Data (t): 0.391 Batch (t): 2.431, 1693.68/s, 846.839/s/gpu LR: 0.000494 Logit Scale: 18.648 Contrastive_loss: 6.4210 (7.2240) Loss: 6.4210 (7.2240)
2025-04-29,19:32:14 | INFO | Train Epoch: 0 [11063296/16056320 (69%)] Data (t): 0.396 Batch (t): 2.438, 1692.32/s, 846.162/s/gpu LR: 0.000494 Logit Scale: 18.913 Contrastive_loss: 6.4979 (7.1981) Loss: 6.4979 (7.1981)
2025-04-29,19:36:17 | INFO | Train Epoch: 0 [11472896/16056320 (71%)] Data (t): 0.390 Batch (t): 2.429, 1678.49/s, 839.246/s/gpu LR: 0.000493 Logit Scale: 19.224 Contrastive_loss: 6.4005 (7.1706) Loss: 6.4005 (7.1706)
2025-04-29,19:40:20 | INFO | Train Epoch: 0 [11882496/16056320 (74%)] Data (t): 0.387 Batch (t): 2.430, 1678.26/s, 839.130/s/gpu LR: 0.000493 Logit Scale: 19.531 Contrastive_loss: 6.3682 (7.1438) Loss: 6.3682 (7.1438)
2025-04-29,19:44:23 | INFO | Train Epoch: 0 [12292096/16056320 (77%)] Data (t): 0.396 Batch (t): 2.436, 1677.17/s, 838.587/s/gpu LR: 0.000492 Logit Scale: 19.819 Contrastive_loss: 6.1071 (7.1104) Loss: 6.1071 (7.1104)
2025-04-29,19:48:27 | INFO | Train Epoch: 0 [12701696/16056320 (79%)] Data (t): 0.398 Batch (t): 2.438, 1696.85/s, 848.426/s/gpu LR: 0.000491 Logit Scale: 20.042 Contrastive_loss: 6.3618 (7.0870) Loss: 6.3618 (7.0870)
2025-04-29,19:52:31 | INFO | Train Epoch: 0 [13111296/16056320 (82%)] Data (t): 0.398 Batch (t): 2.439, 1606.99/s, 803.494/s/gpu LR: 0.000491 Logit Scale: 20.290 Contrastive_loss: 6.2285 (7.0610) Loss: 6.2285 (7.0610)
2025-04-29,19:56:35 | INFO | Train Epoch: 0 [13520896/16056320 (84%)] Data (t): 0.402 Batch (t): 2.443, 1668.95/s, 834.473/s/gpu LR: 0.000490 Logit Scale: 20.522 Contrastive_loss: 6.0085 (7.0300) Loss: 6.0085 (7.0300)
2025-04-29,20:00:38 | INFO | Train Epoch: 0 [13930496/16056320 (87%)] Data (t): 0.389 Batch (t): 2.429, 1705.41/s, 852.704/s/gpu LR: 0.000489 Logit Scale: 20.775 Contrastive_loss: 6.2236 (7.0070) Loss: 6.2236 (7.0070)
2025-04-29,20:04:42 | INFO | Train Epoch: 0 [14340096/16056320 (89%)] Data (t): 0.397 Batch (t): 2.439, 1616.17/s, 808.085/s/gpu LR: 0.000488 Logit Scale: 21.015 Contrastive_loss: 6.1521 (6.9832) Loss: 6.1521 (6.9832)
2025-04-29,20:08:46 | INFO | Train Epoch: 0 [14749696/16056320 (92%)] Data (t): 0.396 Batch (t): 2.438, 1700.78/s, 850.392/s/gpu LR: 0.000488 Logit Scale: 21.230 Contrastive_loss: 6.1918 (6.9618) Loss: 6.1918 (6.9618)
2025-04-29,20:12:49 | INFO | Train Epoch: 0 [15159296/16056320 (94%)] Data (t): 0.390 Batch (t): 2.432, 1695.14/s, 847.571/s/gpu LR: 0.000487 Logit Scale: 21.468 Contrastive_loss: 6.1201 (6.9397) Loss: 6.1201 (6.9397)
2025-04-29,20:16:52 | INFO | Train Epoch: 0 [15568896/16056320 (97%)] Data (t): 0.390 Batch (t): 2.432, 1704.81/s, 852.404/s/gpu LR: 0.000486 Logit Scale: 21.638 Contrastive_loss: 6.0887 (6.9179) Loss: 6.0887 (6.9179)
|