Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 8.42,
-    "eval_accuracy": 0.5866666666666667,
-    "eval_loss": 0.9113264679908752,
-    "eval_runtime": 2.6506,
-    "eval_samples_per_second": 28.295,
-    "eval_steps_per_second": 3.773,
-    "total_flos": 9.777525342378394e+16,
-    "train_loss": 0.5095690906047821,
-    "train_runtime": 80.3972,
-    "train_samples_per_second": 18.409,
-    "train_steps_per_second": 0.498
+    "epoch": 9.73,
+    "eval_accuracy": 0.6,
+    "eval_loss": 1.9666080474853516,
+    "eval_runtime": 2.8446,
+    "eval_samples_per_second": 26.366,
+    "eval_steps_per_second": 6.679,
+    "total_flos": 1.1281760010436608e+17,
+    "train_loss": 0.20268239630386234,
+    "train_runtime": 97.0992,
+    "train_samples_per_second": 15.242,
+    "train_steps_per_second": 0.927
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.42,
-    "eval_accuracy": 0.5866666666666667,
-    "eval_loss": 0.9113264679908752,
-    "eval_runtime": 2.6506,
-    "eval_samples_per_second": 28.295,
-    "eval_steps_per_second": 3.773
+    "epoch": 9.73,
+    "eval_accuracy": 0.6,
+    "eval_loss": 1.9666080474853516,
+    "eval_runtime": 2.8446,
+    "eval_samples_per_second": 26.366,
+    "eval_steps_per_second": 6.679
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2fa7c583db0286445600df45d8fa7c50dd9d631e0cc50128622968207d7d504
+oid sha256:1a8a68e6e0036cd2b7cf183a53edb004daa0b046e9ec5b2ec7e5e87dbf233ead
 size 347498816

runs/Dec28_09-46-47_DESKTOP-P9FU4RH/events.out.tfevents.1703775068.DESKTOP-P9FU4RH.34460.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2e771f966771732832d36ff9f96f60c2e543d873c264f7490e7a0ac1d5b0421
+size 722

runs/Dec28_09-52-48_DESKTOP-P9FU4RH/events.out.tfevents.1703775169.DESKTOP-P9FU4RH.34460.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:066b90716e4492aa37c6da5e209082be68858c861fc719b00d64ea3faaf12ba2
+size 5268

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.42,
-    "total_flos": 9.777525342378394e+16,
-    "train_loss": 0.5095690906047821,
-    "train_runtime": 80.3972,
-    "train_samples_per_second": 18.409,
-    "train_steps_per_second": 0.498
+    "epoch": 9.73,
+    "total_flos": 1.1281760010436608e+17,
+    "train_loss": 0.20268239630386234,
+    "train_runtime": 97.0992,
+    "train_samples_per_second": 15.242,
+    "train_steps_per_second": 0.927
 }

trainer_state.json

@@ -1,135 +1,174 @@
 {
   "best_metric": 0.7837837837837838,
-  "best_model_checkpoint": "swin-finetuned-class_mi_a4c\\checkpoint-28",
-  "epoch": 8.421052631578947,
+  "best_model_checkpoint": "swin-finetuned-class_mi_a4c\\checkpoint-64",
+  "epoch": 9.72972972972973,
   "eval_steps": 500,
-  "global_step": 40,
+  "global_step": 90,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.84,
-      "eval_accuracy": 0.6486486486486487,
-      "eval_loss": 0.6846584677696228,
-      "eval_runtime": 1.2742,
-      "eval_samples_per_second": 29.037,
-      "eval_steps_per_second": 3.924,
-      "step": 4
-    },
-    {
-      "epoch": 1.89,
-      "eval_accuracy": 0.6486486486486487,
-      "eval_loss": 0.6533997058868408,
-      "eval_runtime": 1.3318,
-      "eval_samples_per_second": 27.783,
-      "eval_steps_per_second": 3.754,
+      "epoch": 0.97,
+      "eval_accuracy": 0.35135135135135137,
+      "eval_loss": 1.0705137252807617,
+      "eval_runtime": 1.3164,
+      "eval_samples_per_second": 28.107,
+      "eval_steps_per_second": 7.597,
       "step": 9
     },
     {
-      "epoch": 2.11,
-      "learning_rate": 8.333333333333334e-05,
-      "loss": 0.6821,
+      "epoch": 1.08,
+      "learning_rate": 9.876543209876543e-05,
+      "loss": 0.5074,
       "step": 10
     },
     {
-      "epoch": 2.95,
-      "eval_accuracy": 0.6216216216216216,
-      "eval_loss": 0.6845701932907104,
-      "eval_runtime": 1.2931,
-      "eval_samples_per_second": 28.614,
-      "eval_steps_per_second": 3.867,
-      "step": 14
+      "epoch": 1.95,
+      "eval_accuracy": 0.7027027027027027,
+      "eval_loss": 0.5304501056671143,
+      "eval_runtime": 1.3,
+      "eval_samples_per_second": 28.461,
+      "eval_steps_per_second": 7.692,
+      "step": 18
     },
     {
-      "epoch": 4.0,
+      "epoch": 2.16,
+      "learning_rate": 8.641975308641975e-05,
+      "loss": 0.4483,
+      "step": 20
+    },
+    {
+      "epoch": 2.92,
       "eval_accuracy": 0.7567567567567568,
-      "eval_loss": 0.5264998078346252,
-      "eval_runtime": 1.2544,
-      "eval_samples_per_second": 29.497,
-      "eval_steps_per_second": 3.986,
-      "step": 19
+      "eval_loss": 0.6806666254997253,
+      "eval_runtime": 1.3169,
+      "eval_samples_per_second": 28.097,
+      "eval_steps_per_second": 7.594,
+      "step": 27
     },
     {
-      "epoch": 4.21,
-      "learning_rate": 5.555555555555556e-05,
-      "loss": 0.6259,
-      "step": 20
+      "epoch": 3.24,
+      "learning_rate": 7.407407407407407e-05,
+      "loss": 0.3577,
+      "step": 30
     },
     {
-      "epoch": 4.84,
-      "eval_accuracy": 0.7567567567567568,
-      "eval_loss": 0.5310971736907959,
-      "eval_runtime": 1.2613,
-      "eval_samples_per_second": 29.334,
-      "eval_steps_per_second": 3.964,
-      "step": 23
+      "epoch": 4.0,
+      "eval_accuracy": 0.6756756756756757,
+      "eval_loss": 0.8441636562347412,
+      "eval_runtime": 1.3387,
+      "eval_samples_per_second": 27.639,
+      "eval_steps_per_second": 7.47,
+      "step": 37
     },
     {
-      "epoch": 5.89,
-      "eval_accuracy": 0.7837837837837838,
-      "eval_loss": 0.5249072313308716,
-      "eval_runtime": 1.2742,
-      "eval_samples_per_second": 29.037,
-      "eval_steps_per_second": 3.924,
-      "step": 28
+      "epoch": 4.32,
+      "learning_rate": 6.17283950617284e-05,
+      "loss": 0.3117,
+      "step": 40
     },
     {
-      "epoch": 6.32,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.4588,
-      "step": 30
+      "epoch": 4.97,
+      "eval_accuracy": 0.7297297297297297,
+      "eval_loss": 0.8780416250228882,
+      "eval_runtime": 1.3075,
+      "eval_samples_per_second": 28.299,
+      "eval_steps_per_second": 7.648,
+      "step": 46
     },
     {
-      "epoch": 6.95,
-      "eval_accuracy": 0.6756756756756757,
-      "eval_loss": 0.5719398260116577,
-      "eval_runtime": 1.2534,
-      "eval_samples_per_second": 29.52,
-      "eval_steps_per_second": 3.989,
-      "step": 33
+      "epoch": 5.41,
+      "learning_rate": 4.938271604938271e-05,
+      "loss": 0.0626,
+      "step": 50
+    },
+    {
+      "epoch": 5.95,
+      "eval_accuracy": 0.7027027027027027,
+      "eval_loss": 1.5264472961425781,
+      "eval_runtime": 1.3724,
+      "eval_samples_per_second": 26.959,
+      "eval_steps_per_second": 7.286,
+      "step": 55
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.09,
+      "step": 60
+    },
+    {
+      "epoch": 6.92,
+      "eval_accuracy": 0.7837837837837838,
+      "eval_loss": 1.075486183166504,
+      "eval_runtime": 1.3625,
+      "eval_samples_per_second": 27.156,
+      "eval_steps_per_second": 7.339,
+      "step": 64
+    },
+    {
+      "epoch": 7.57,
+      "learning_rate": 2.4691358024691357e-05,
+      "loss": 0.0436,
+      "step": 70
     },
     {
       "epoch": 8.0,
       "eval_accuracy": 0.7027027027027027,
-      "eval_loss": 0.5473979711532593,
-      "eval_runtime": 1.3159,
-      "eval_samples_per_second": 28.118,
-      "eval_steps_per_second": 3.8,
-      "step": 38
+      "eval_loss": 1.428911805152893,
+      "eval_runtime": 1.3263,
+      "eval_samples_per_second": 27.897,
+      "eval_steps_per_second": 7.54,
+      "step": 74
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 1.2345679012345678e-05,
+      "loss": 0.0016,
+      "step": 80
     },
     {
-      "epoch": 8.42,
+      "epoch": 8.97,
+      "eval_accuracy": 0.7837837837837838,
+      "eval_loss": 1.3568603992462158,
+      "eval_runtime": 1.4047,
+      "eval_samples_per_second": 26.341,
+      "eval_steps_per_second": 7.119,
+      "step": 83
+    },
+    {
+      "epoch": 9.73,
       "learning_rate": 0.0,
-      "loss": 0.2715,
-      "step": 40
+      "loss": 0.0014,
+      "step": 90
     },
     {
-      "epoch": 8.42,
-      "eval_accuracy": 0.7567567567567568,
-      "eval_loss": 0.5581153631210327,
-      "eval_runtime": 1.3075,
-      "eval_samples_per_second": 28.299,
-      "eval_steps_per_second": 3.824,
-      "step": 40
+      "epoch": 9.73,
+      "eval_accuracy": 0.7837837837837838,
+      "eval_loss": 1.3536738157272339,
+      "eval_runtime": 1.3288,
+      "eval_samples_per_second": 27.845,
+      "eval_steps_per_second": 7.526,
+      "step": 90
     },
     {
-      "epoch": 8.42,
-      "step": 40,
-      "total_flos": 9.777525342378394e+16,
-      "train_loss": 0.5095690906047821,
-      "train_runtime": 80.3972,
-      "train_samples_per_second": 18.409,
-      "train_steps_per_second": 0.498
+      "epoch": 9.73,
+      "step": 90,
+      "total_flos": 1.1281760010436608e+17,
+      "train_loss": 0.20268239630386234,
+      "train_runtime": 97.0992,
+      "train_samples_per_second": 15.242,
+      "train_steps_per_second": 0.927
     }
   ],
   "logging_steps": 10,
-  "max_steps": 40,
+  "max_steps": 90,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 9.777525342378394e+16,
-  "train_batch_size": 8,
+  "total_flos": 1.1281760010436608e+17,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f7d3d16d0f4adb0244f4fece47cbd128721dba1e2a65f790b60d37599fa7755
+oid sha256:d00e55186ff6e38bf15ed11081ee13c35c869301b885aa2945787966a1bde8a2
 size 4728