Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 9.0,
-    "eval_accuracy": 0.9851851851851852,
-    "eval_loss": 0.063043974339962,
-    "eval_runtime": 0.7297,
-    "eval_samples_per_second": 185.014,
-    "eval_steps_per_second": 6.852,
-    "total_flos": 1.4093739557001216e+17,
-    "train_loss": 0.3985253996319241,
-    "train_runtime": 73.599,
-    "train_samples_per_second": 171.198,
-    "train_steps_per_second": 1.359
+    "epoch": 8.0,
+    "eval_accuracy": 0.8484848484848485,
+    "eval_loss": 0.5320981740951538,
+    "eval_runtime": 0.1987,
+    "eval_samples_per_second": 166.048,
+    "eval_steps_per_second": 10.064,
+    "total_flos": 3.121999450369229e+16,
+    "train_loss": 0.7512132167816162,
+    "train_runtime": 20.6283,
+    "train_samples_per_second": 76.109,
+    "train_steps_per_second": 0.485
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.0,
-    "eval_accuracy": 0.9851851851851852,
-    "eval_loss": 0.063043974339962,
-    "eval_runtime": 0.7297,
-    "eval_samples_per_second": 185.014,
-    "eval_steps_per_second": 6.852
+    "epoch": 8.0,
+    "eval_accuracy": 0.9090909090909091,
+    "eval_loss": 0.5003632307052612,
+    "eval_runtime": 0.1947,
+    "eval_samples_per_second": 169.491,
+    "eval_steps_per_second": 10.272
 }

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.8484848484848485,
+    "eval_loss": 0.5320981740951538,
+    "eval_runtime": 0.1987,
+    "eval_samples_per_second": 166.048,
+    "eval_steps_per_second": 10.064
+}

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.0,
-    "total_flos": 1.4093739557001216e+17,
-    "train_loss": 0.3985253996319241,
-    "train_runtime": 73.599,
-    "train_samples_per_second": 171.198,
-    "train_steps_per_second": 1.359
+    "epoch": 8.0,
+    "total_flos": 3.121999450369229e+16,
+    "train_loss": 0.7512132167816162,
+    "train_runtime": 20.6283,
+    "train_samples_per_second": 76.109,
+    "train_steps_per_second": 0.485
 }

trainer_state.json

@@ -1,182 +1,157 @@
 {
-  "best_metric": 0.9851851851851852,
-  "best_model_checkpoint": "/kaggle/working/swin-brain-plane-classification/checkpoint-45",
-  "epoch": 9.0,
+  "best_metric": 0.9090909090909091,
+  "best_model_checkpoint": "/kaggle/working/swin-brain-plane-classification/checkpoint-6",
+  "epoch": 8.0,
   "eval_steps": 500,
-  "global_step": 45,
+  "global_step": 10,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "grad_norm": 5.971311092376709,
-      "learning_rate": 2.5e-05,
-      "loss": 1.0732,
-      "step": 5
+      "epoch": 0.8,
+      "grad_norm": 7.547826766967773,
+      "learning_rate": 5e-05,
+      "loss": 1.3313,
+      "step": 1
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 0.9284645318984985,
-      "eval_runtime": 0.7118,
-      "eval_samples_per_second": 189.658,
-      "eval_steps_per_second": 7.024,
-      "step": 5
+      "epoch": 0.8,
+      "eval_accuracy": 0.30303030303030304,
+      "eval_loss": 1.0681976079940796,
+      "eval_runtime": 0.1993,
+      "eval_samples_per_second": 165.581,
+      "eval_steps_per_second": 10.035,
+      "step": 1
     },
     {
-      "epoch": 2.0,
-      "grad_norm": 10.538399696350098,
-      "learning_rate": 5e-05,
-      "loss": 0.829,
-      "step": 10
+      "epoch": 1.6,
+      "grad_norm": 4.482707500457764,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.2576,
+      "step": 2
     },
     {
-      "epoch": 2.0,
-      "eval_accuracy": 0.8592592592592593,
-      "eval_loss": 0.527397871017456,
-      "eval_runtime": 0.7119,
-      "eval_samples_per_second": 189.625,
-      "eval_steps_per_second": 7.023,
-      "step": 10
+      "epoch": 1.6,
+      "eval_accuracy": 0.7878787878787878,
+      "eval_loss": 0.8504449129104614,
+      "eval_runtime": 0.202,
+      "eval_samples_per_second": 163.354,
+      "eval_steps_per_second": 9.9,
+      "step": 2
     },
     {
-      "epoch": 3.0,
-      "grad_norm": 6.285182476043701,
-      "learning_rate": 4.722222222222222e-05,
-      "loss": 0.5189,
-      "step": 15
+      "epoch": 2.4,
+      "grad_norm": 4.672352313995361,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 1.0651,
+      "step": 3
     },
     {
-      "epoch": 3.0,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.24730348587036133,
-      "eval_runtime": 0.6998,
-      "eval_samples_per_second": 192.917,
-      "eval_steps_per_second": 7.145,
-      "step": 15
+      "epoch": 2.4,
+      "eval_accuracy": 0.7575757575757576,
+      "eval_loss": 0.7269154191017151,
+      "eval_runtime": 0.202,
+      "eval_samples_per_second": 163.337,
+      "eval_steps_per_second": 9.899,
+      "step": 3
     },
     {
       "epoch": 4.0,
-      "grad_norm": 9.056550979614258,
-      "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.3517,
-      "step": 20
+      "grad_norm": 6.733062267303467,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.4774,
+      "step": 5
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9629629629629629,
-      "eval_loss": 0.1404803842306137,
-      "eval_runtime": 0.7065,
-      "eval_samples_per_second": 191.084,
-      "eval_steps_per_second": 7.077,
-      "step": 20
+      "eval_accuracy": 0.8484848484848485,
+      "eval_loss": 0.5601949691772461,
+      "eval_runtime": 0.1975,
+      "eval_samples_per_second": 167.092,
+      "eval_steps_per_second": 10.127,
+      "step": 5
     },
     {
-      "epoch": 5.0,
-      "grad_norm": 7.212975025177002,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.2075,
-      "step": 25
+      "epoch": 4.8,
+      "grad_norm": 5.887814521789551,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.8084,
+      "step": 6
     },
     {
-      "epoch": 5.0,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.17972101271152496,
-      "eval_runtime": 0.703,
-      "eval_samples_per_second": 192.032,
-      "eval_steps_per_second": 7.112,
-      "step": 25
+      "epoch": 4.8,
+      "eval_accuracy": 0.9090909090909091,
+      "eval_loss": 0.5003632307052612,
+      "eval_runtime": 0.2283,
+      "eval_samples_per_second": 144.536,
+      "eval_steps_per_second": 8.76,
+      "step": 6
     },
     {
-      "epoch": 6.0,
-      "grad_norm": 5.206702709197998,
-      "learning_rate": 3.888888888888889e-05,
-      "loss": 0.1871,
-      "step": 30
+      "epoch": 5.6,
+      "grad_norm": 5.614295959472656,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.7194,
+      "step": 7
     },
     {
-      "epoch": 6.0,
-      "eval_accuracy": 0.9703703703703703,
-      "eval_loss": 0.09362433105707169,
-      "eval_runtime": 0.759,
-      "eval_samples_per_second": 177.874,
-      "eval_steps_per_second": 6.588,
-      "step": 30
+      "epoch": 5.6,
+      "eval_accuracy": 0.9090909090909091,
+      "eval_loss": 0.4589892327785492,
+      "eval_runtime": 0.1953,
+      "eval_samples_per_second": 168.965,
+      "eval_steps_per_second": 10.24,
+      "step": 7
     },
     {
-      "epoch": 7.0,
-      "grad_norm": 5.035118579864502,
-      "learning_rate": 3.611111111111111e-05,
-      "loss": 0.1567,
-      "step": 35
+      "epoch": 6.4,
+      "grad_norm": 7.968162536621094,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.6712,
+      "step": 8
     },
     {
-      "epoch": 7.0,
-      "eval_accuracy": 0.9703703703703703,
-      "eval_loss": 0.1113051250576973,
-      "eval_runtime": 0.7175,
-      "eval_samples_per_second": 188.151,
-      "eval_steps_per_second": 6.969,
-      "step": 35
+      "epoch": 6.4,
+      "eval_accuracy": 0.9090909090909091,
+      "eval_loss": 0.4304216802120209,
+      "eval_runtime": 0.2015,
+      "eval_samples_per_second": 163.754,
+      "eval_steps_per_second": 9.924,
+      "step": 8
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 6.553189754486084,
+      "learning_rate": 0.0,
+      "loss": 0.3522,
+      "step": 10
     },
     {
       "epoch": 8.0,
-      "grad_norm": 4.850553035736084,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.1482,
-      "step": 40
+      "eval_accuracy": 0.9090909090909091,
+      "eval_loss": 0.4053419828414917,
+      "eval_runtime": 0.2196,
+      "eval_samples_per_second": 150.303,
+      "eval_steps_per_second": 9.109,
+      "step": 10
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9777777777777777,
-      "eval_loss": 0.0790533721446991,
-      "eval_runtime": 0.7241,
-      "eval_samples_per_second": 186.438,
-      "eval_steps_per_second": 6.905,
-      "step": 40
-    },
-    {
-      "epoch": 9.0,
-      "grad_norm": 4.634683132171631,
-      "learning_rate": 3.055555555555556e-05,
-      "loss": 0.1143,
-      "step": 45
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.9851851851851852,
-      "eval_loss": 0.063043974339962,
-      "eval_runtime": 0.7126,
-      "eval_samples_per_second": 189.438,
-      "eval_steps_per_second": 7.016,
-      "step": 45
-    },
-    {
-      "epoch": 9.0,
-      "step": 45,
-      "total_flos": 1.4093739557001216e+17,
-      "train_loss": 0.3985253996319241,
-      "train_runtime": 73.599,
-      "train_samples_per_second": 171.198,
-      "train_steps_per_second": 1.359
+      "step": 10,
+      "total_flos": 3.121999450369229e+16,
+      "train_loss": 0.7512132167816162,
+      "train_runtime": 20.6283,
+      "train_samples_per_second": 76.109,
+      "train_steps_per_second": 0.485
     }
   ],
-  "logging_steps": 500,
-  "max_steps": 100,
+  "logging_steps": 1,
+  "max_steps": 10,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 20,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
-    "CustomEarlyStoppingCallback": {
-      "args": {
-        "early_stopping_patience": 1,
-        "early_stopping_threshold": 0.0
-      },
-      "attributes": {
-        "early_stopping_patience_counter": 0
-      }
-    },
     "TrainerControl": {
       "args": {
         "should_epoch_stop": false,
@@ -188,7 +163,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.4093739557001216e+17,
+  "total_flos": 3.121999450369229e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null