End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6220
-- Accuracy: 0.75
+- Loss: 0.7514
+- Accuracy: 0.8
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 10.67,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.4407931864261627,
-    "eval_runtime": 0.3405,
-    "eval_samples_per_second": 58.744,
-    "eval_steps_per_second": 2.937,
-    "total_flos": 1.4940723410848973e+17,
-    "train_loss": 0.8236366659402847,
-    "train_runtime": 144.9541,
-    "train_samples_per_second": 19.868,
-    "train_steps_per_second": 0.11
+    "epoch": 9.33,
+    "eval_accuracy": 0.8,
+    "eval_loss": 0.7513991594314575,
+    "eval_runtime": 0.3965,
+    "eval_samples_per_second": 50.447,
+    "eval_steps_per_second": 2.522,
+    "total_flos": 1.3049884970886758e+17,
+    "train_loss": 0.9208662169320243,
+    "train_runtime": 133.2771,
+    "train_samples_per_second": 18.908,
+    "train_steps_per_second": 0.105
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.67,
-    "eval_accuracy": 1.0,
-    "eval_loss": 0.4407931864261627,
-    "eval_runtime": 0.3405,
-    "eval_samples_per_second": 58.744,
-    "eval_steps_per_second": 2.937
+    "epoch": 9.33,
+    "eval_accuracy": 0.8,
+    "eval_loss": 0.7513991594314575,
+    "eval_runtime": 0.3965,
+    "eval_samples_per_second": 50.447,
+    "eval_steps_per_second": 2.522
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.67,
-    "total_flos": 1.4940723410848973e+17,
-    "train_loss": 0.8236366659402847,
-    "train_runtime": 144.9541,
-    "train_samples_per_second": 19.868,
-    "train_steps_per_second": 0.11
+    "epoch": 9.33,
+    "total_flos": 1.3049884970886758e+17,
+    "train_loss": 0.9208662169320243,
+    "train_runtime": 133.2771,
+    "train_samples_per_second": 18.908,
+    "train_steps_per_second": 0.105
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 1.0,
-  "best_model_checkpoint": "vit-base-patch16-224-MSC-ARMD/checkpoint-13",
-  "epoch": 10.666666666666666,
+  "best_metric": 0.8,
+  "best_model_checkpoint": "vit-base-patch16-224-MSC-ARMD/checkpoint-9",
+  "epoch": 9.333333333333334,
   "eval_steps": 500,
-  "global_step": 16,
+  "global_step": 14,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,123 +11,114 @@
     {
       "epoch": 0.67,
       "eval_accuracy": 0.25,
-      "eval_loss": 1.5351471900939941,
-      "eval_runtime": 0.5183,
-      "eval_samples_per_second": 38.59,
-      "eval_steps_per_second": 1.93,
+      "eval_loss": 1.5119588375091553,
+      "eval_runtime": 0.7017,
+      "eval_samples_per_second": 28.501,
+      "eval_steps_per_second": 1.425,
       "step": 1
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.5,
-      "eval_loss": 1.1686809062957764,
-      "eval_runtime": 0.3541,
-      "eval_samples_per_second": 56.482,
-      "eval_steps_per_second": 2.824,
+      "eval_loss": 1.2053601741790771,
+      "eval_runtime": 0.2913,
+      "eval_samples_per_second": 68.655,
+      "eval_steps_per_second": 3.433,
       "step": 3
     },
     {
       "epoch": 2.67,
-      "eval_accuracy": 0.55,
-      "eval_loss": 1.0381157398223877,
-      "eval_runtime": 0.3555,
-      "eval_samples_per_second": 56.253,
-      "eval_steps_per_second": 2.813,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.1144676208496094,
+      "eval_runtime": 0.323,
+      "eval_samples_per_second": 61.913,
+      "eval_steps_per_second": 3.096,
       "step": 4
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9,
-      "eval_loss": 0.7977484464645386,
-      "eval_runtime": 0.292,
-      "eval_samples_per_second": 68.489,
-      "eval_steps_per_second": 3.424,
+      "eval_accuracy": 0.6,
+      "eval_loss": 0.9345596432685852,
+      "eval_runtime": 0.3114,
+      "eval_samples_per_second": 64.231,
+      "eval_steps_per_second": 3.212,
       "step": 6
     },
     {
       "epoch": 4.67,
-      "eval_accuracy": 0.95,
-      "eval_loss": 0.7476352453231812,
-      "eval_runtime": 0.3015,
-      "eval_samples_per_second": 66.344,
-      "eval_steps_per_second": 3.317,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.8657038807868958,
+      "eval_runtime": 0.2854,
+      "eval_samples_per_second": 70.066,
+      "eval_steps_per_second": 3.503,
       "step": 7
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9,
-      "eval_loss": 0.583551287651062,
-      "eval_runtime": 0.3193,
-      "eval_samples_per_second": 62.628,
-      "eval_steps_per_second": 3.131,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.7513991594314575,
+      "eval_runtime": 0.3099,
+      "eval_samples_per_second": 64.545,
+      "eval_steps_per_second": 3.227,
       "step": 9
     },
     {
       "epoch": 6.67,
-      "learning_rate": 2.1428571428571428e-05,
-      "loss": 1.0167,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 1.0524,
       "step": 10
     },
     {
       "epoch": 6.67,
-      "eval_accuracy": 0.85,
-      "eval_loss": 0.5264605283737183,
-      "eval_runtime": 0.288,
-      "eval_samples_per_second": 69.451,
-      "eval_steps_per_second": 3.473,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.7025678753852844,
+      "eval_runtime": 0.2863,
+      "eval_samples_per_second": 69.863,
+      "eval_steps_per_second": 3.493,
       "step": 10
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9,
-      "eval_loss": 0.4598480761051178,
-      "eval_runtime": 0.3258,
-      "eval_samples_per_second": 61.385,
-      "eval_steps_per_second": 3.069,
+      "eval_accuracy": 0.8,
+      "eval_loss": 0.6446658372879028,
+      "eval_runtime": 0.3043,
+      "eval_samples_per_second": 65.714,
+      "eval_steps_per_second": 3.286,
       "step": 12
     },
     {
       "epoch": 8.67,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.4407931864261627,
-      "eval_runtime": 0.3249,
-      "eval_samples_per_second": 61.563,
-      "eval_steps_per_second": 3.078,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.6296584010124207,
+      "eval_runtime": 0.2888,
+      "eval_samples_per_second": 69.243,
+      "eval_steps_per_second": 3.462,
       "step": 13
     },
     {
-      "epoch": 10.0,
-      "eval_accuracy": 0.95,
-      "eval_loss": 0.41913142800331116,
-      "eval_runtime": 0.3169,
-      "eval_samples_per_second": 63.114,
-      "eval_steps_per_second": 3.156,
-      "step": 15
-    },
-    {
-      "epoch": 10.67,
-      "eval_accuracy": 0.95,
-      "eval_loss": 0.41399478912353516,
-      "eval_runtime": 0.3106,
-      "eval_samples_per_second": 64.393,
-      "eval_steps_per_second": 3.22,
-      "step": 16
+      "epoch": 9.33,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.6220413446426392,
+      "eval_runtime": 0.3202,
+      "eval_samples_per_second": 62.47,
+      "eval_steps_per_second": 3.123,
+      "step": 14
     },
     {
-      "epoch": 10.67,
-      "step": 16,
-      "total_flos": 1.4940723410848973e+17,
-      "train_loss": 0.8236366659402847,
-      "train_runtime": 144.9541,
-      "train_samples_per_second": 19.868,
-      "train_steps_per_second": 0.11
+      "epoch": 9.33,
+      "step": 14,
+      "total_flos": 1.3049884970886758e+17,
+      "train_loss": 0.9208662169320243,
+      "train_runtime": 133.2771,
+      "train_samples_per_second": 18.908,
+      "train_steps_per_second": 0.105
     }
   ],
   "logging_steps": 10,
-  "max_steps": 16,
-  "num_train_epochs": 16,
+  "max_steps": 14,
+  "num_train_epochs": 14,
   "save_steps": 500,
-  "total_flos": 1.4940723410848973e+17,
+  "total_flos": 1.3049884970886758e+17,
   "trial_name": null,
   "trial_params": null
 }