Training in progress, epoch 1

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 10.0,
+    "epoch": 2.0,
     "eval_accuracy": 0.2283464566929134,
     "eval_f1": 0.31496062992125984,
-    "eval_loss": 2.054736191474893e+16,
-    "eval_runtime": 2.0937,
-    "eval_samples_per_second": 60.658,
-    "eval_steps_per_second": 1.91,
-    "train_loss": 1.794394348418353e+16,
-    "train_runtime": 289.567,
-    "train_samples_per_second": 39.438,
-    "train_steps_per_second": 0.311
+    "eval_loss": NaN,
+    "eval_runtime": 2.2244,
+    "eval_samples_per_second": 57.094,
+    "eval_steps_per_second": 1.798,
+    "train_loss": 0.0,
+    "train_runtime": 58.4627,
+    "train_samples_per_second": 39.068,
+    "train_steps_per_second": 0.308
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
+    "epoch": 2.0,
     "eval_accuracy": 0.2283464566929134,
-    "eval_loss": 2.054736191474893e+16,
-    "eval_runtime": 2.0937,
-    "eval_samples_per_second": 60.658,
-    "eval_steps_per_second": 1.91
+    "eval_loss": NaN,
+    "eval_runtime": 2.2244,
+    "eval_samples_per_second": 57.094,
+    "eval_steps_per_second": 1.798
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae1aa792a538dcb8b33fc25aefa03df36e0b84e26b800ef5115a1b9a6c32b2dc
+oid sha256:7ab8c3c8874807f211a42edf9001413a2d5a24898f3a99c384a3fb9350bc1760
 size 110355136

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 10.0,
-    "train_loss": 1.794394348418353e+16,
-    "train_runtime": 289.567,
-    "train_samples_per_second": 39.438,
-    "train_steps_per_second": 0.311
+    "epoch": 2.0,
+    "train_loss": 0.0,
+    "train_runtime": 58.4627,
+    "train_samples_per_second": 39.068,
+    "train_steps_per_second": 0.308
 }

trainer_state.json

@@ -1,173 +1,53 @@
 {
   "best_metric": 0.2283464566929134,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-mgasior-2024/checkpoint-27",
-  "epoch": 10.0,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-mgasior-2024/checkpoint-9",
+  "epoch": 2.0,
   "eval_steps": 500,
-  "global_step": 90,
+  "global_step": 18,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 1.2739158615261184e+16,
-      "eval_runtime": 1.7678,
-      "eval_samples_per_second": 71.839,
-      "eval_steps_per_second": 2.263,
+      "eval_accuracy": 0.2283464566929134,
+      "eval_loss": NaN,
+      "eval_runtime": 1.7274,
+      "eval_samples_per_second": 73.521,
+      "eval_steps_per_second": 2.316,
       "step": 9
     },
     {
       "epoch": 1.11,
-      "learning_rate": 0.04938271604938271,
-      "loss": 4.9989753141356136e+16,
+      "learning_rate": 0.025,
+      "loss": 0.0,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 1.669814028337152e+16,
-      "eval_runtime": 2.1815,
-      "eval_samples_per_second": 58.216,
-      "eval_steps_per_second": 1.834,
-      "step": 18
-    },
-    {
-      "epoch": 2.22,
-      "learning_rate": 0.043209876543209874,
-      "loss": 3.340994242821489e+16,
-      "step": 20
-    },
-    {
-      "epoch": 3.0,
       "eval_accuracy": 0.2283464566929134,
-      "eval_loss": 2.054736191474893e+16,
-      "eval_runtime": 1.827,
-      "eval_samples_per_second": 69.513,
-      "eval_steps_per_second": 2.189,
-      "step": 27
-    },
-    {
-      "epoch": 3.33,
-      "learning_rate": 0.037037037037037035,
-      "loss": 1.7126365917400268e+16,
-      "step": 30
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 1.1163631467102208e+16,
-      "eval_runtime": 1.8924,
-      "eval_samples_per_second": 67.109,
-      "eval_steps_per_second": 2.114,
-      "step": 36
-    },
-    {
-      "epoch": 4.44,
-      "learning_rate": 0.030864197530864196,
-      "loss": 1.1563202153073868e+16,
-      "step": 40
-    },
-    {
-      "epoch": 5.0,
-      "eval_accuracy": 0.2283464566929134,
-      "eval_loss": 1.2371938911453184e+16,
-      "eval_runtime": 2.0238,
-      "eval_samples_per_second": 62.753,
-      "eval_steps_per_second": 1.976,
-      "step": 45
-    },
-    {
-      "epoch": 5.56,
-      "learning_rate": 0.024691358024691357,
-      "loss": 1.0920572825370624e+16,
-      "step": 50
-    },
-    {
-      "epoch": 6.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 9788674996699136.0,
-      "eval_runtime": 1.761,
-      "eval_samples_per_second": 72.116,
-      "eval_steps_per_second": 2.271,
-      "step": 54
-    },
-    {
-      "epoch": 6.67,
-      "learning_rate": 0.018518518518518517,
-      "loss": 9718191576894668.0,
-      "step": 60
-    },
-    {
-      "epoch": 7.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 9368427479171072.0,
-      "eval_runtime": 1.8769,
-      "eval_samples_per_second": 67.666,
-      "eval_steps_per_second": 2.131,
-      "step": 63
-    },
-    {
-      "epoch": 7.78,
-      "learning_rate": 0.012345679012345678,
-      "loss": 1.0115938190255718e+16,
-      "step": 70
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 9327990362079232.0,
-      "eval_runtime": 1.765,
-      "eval_samples_per_second": 71.954,
-      "eval_steps_per_second": 2.266,
-      "step": 72
-    },
-    {
-      "epoch": 8.89,
-      "learning_rate": 0.006172839506172839,
-      "loss": 9208841097340518.0,
-      "step": 80
-    },
-    {
-      "epoch": 9.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 9341989807980544.0,
-      "eval_runtime": 1.8195,
-      "eval_samples_per_second": 69.798,
-      "eval_steps_per_second": 2.198,
-      "step": 81
-    },
-    {
-      "epoch": 10.0,
-      "learning_rate": 0.0,
-      "loss": 9442684027745076.0,
-      "step": 90
-    },
-    {
-      "epoch": 10.0,
-      "eval_accuracy": 0.2047244094488189,
-      "eval_loss": 9249980871081984.0,
-      "eval_runtime": 1.8668,
-      "eval_samples_per_second": 68.031,
-      "eval_steps_per_second": 2.143,
-      "step": 90
+      "eval_loss": NaN,
+      "eval_runtime": 1.7869,
+      "eval_samples_per_second": 71.072,
+      "eval_steps_per_second": 2.238,
+      "step": 18
     },
     {
-      "epoch": 10.0,
-      "step": 90,
-      "total_flos": 2.8388712063172608e+17,
-      "train_loss": 1.794394348418353e+16,
-      "train_runtime": 289.567,
-      "train_samples_per_second": 39.438,
-      "train_steps_per_second": 0.311
+      "epoch": 2.0,
+      "step": 18,
+      "total_flos": 5.677742412634522e+16,
+      "train_loss": 0.0,
+      "train_runtime": 58.4627,
+      "train_samples_per_second": 39.068,
+      "train_steps_per_second": 0.308
     }
   ],
   "logging_steps": 10,
-  "max_steps": 90,
+  "max_steps": 18,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 2,
   "save_steps": 500,
-  "total_flos": 2.8388712063172608e+17,
+  "total_flos": 5.677742412634522e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be924be90e78ca0ad88653bad5661ec7b07026becdcd659a1a958aac29f8ada6
+oid sha256:88b2cbb247aebdd5a51e04499f69332e274d6f451507f0217261a21d606e299f
 size 4792