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: 0.4312
-- Accuracy: 0.9
+- Loss: 0.6300
+- Accuracy: 0.95
 
 ## Model description
 

all_results.json

@@ -1,8 +1,13 @@
 {
-    "epoch": 2.0,
-    "total_flos": 2.789761632731136e+16,
-    "train_loss": 1.476369063059489,
-    "train_runtime": 26.7457,
-    "train_samples_per_second": 20.19,
-    "train_steps_per_second": 0.112
+    "epoch": 8.0,
+    "eval_accuracy": 0.95,
+    "eval_loss": 0.6299833059310913,
+    "eval_runtime": 0.3394,
+    "eval_samples_per_second": 58.919,
+    "eval_steps_per_second": 2.946,
+    "total_flos": 1.1159046530924544e+17,
+    "train_loss": 0.795779287815094,
+    "train_runtime": 107.6532,
+    "train_samples_per_second": 20.064,
+    "train_steps_per_second": 0.111
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 8.0,
+    "eval_accuracy": 0.95,
+    "eval_loss": 0.6299833059310913,
+    "eval_runtime": 0.3394,
+    "eval_samples_per_second": 58.919,
+    "eval_steps_per_second": 2.946
+}

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 2.789761632731136e+16,
-    "train_loss": 1.476369063059489,
-    "train_runtime": 26.7457,
-    "train_samples_per_second": 20.19,
-    "train_steps_per_second": 0.112
+    "epoch": 8.0,
+    "total_flos": 1.1159046530924544e+17,
+    "train_loss": 0.795779287815094,
+    "train_runtime": 107.6532,
+    "train_samples_per_second": 20.064,
+    "train_steps_per_second": 0.111
 }

trainer_state.json

@@ -1,46 +1,106 @@
 {
-  "best_metric": 0.5,
-  "best_model_checkpoint": "vit-base-patch16-224-MSC-dmae/checkpoint-3",
-  "epoch": 2.0,
+  "best_metric": 0.95,
+  "best_model_checkpoint": "vit-base-patch16-224-MSC-dmae/checkpoint-7",
+  "epoch": 8.0,
   "eval_steps": 500,
-  "global_step": 3,
+  "global_step": 12,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.67,
-      "eval_accuracy": 0.15,
-      "eval_loss": 1.6465644836425781,
-      "eval_runtime": 0.2774,
-      "eval_samples_per_second": 72.11,
-      "eval_steps_per_second": 3.606,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.2257916927337646,
+      "eval_runtime": 0.3201,
+      "eval_samples_per_second": 62.477,
+      "eval_steps_per_second": 3.124,
       "step": 1
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5,
-      "eval_loss": 1.2257916927337646,
-      "eval_runtime": 0.3307,
-      "eval_samples_per_second": 60.471,
-      "eval_steps_per_second": 3.024,
+      "eval_accuracy": 0.7,
+      "eval_loss": 1.0536372661590576,
+      "eval_runtime": 0.3238,
+      "eval_samples_per_second": 61.768,
+      "eval_steps_per_second": 3.088,
       "step": 3
     },
     {
-      "epoch": 2.0,
-      "step": 3,
-      "total_flos": 2.789761632731136e+16,
-      "train_loss": 1.476369063059489,
-      "train_runtime": 26.7457,
-      "train_samples_per_second": 20.19,
-      "train_steps_per_second": 0.112
+      "epoch": 2.67,
+      "eval_accuracy": 0.75,
+      "eval_loss": 0.9143384099006653,
+      "eval_runtime": 0.3026,
+      "eval_samples_per_second": 66.089,
+      "eval_steps_per_second": 3.304,
+      "step": 4
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.6899412870407104,
+      "eval_runtime": 0.3297,
+      "eval_samples_per_second": 60.655,
+      "eval_steps_per_second": 3.033,
+      "step": 6
+    },
+    {
+      "epoch": 4.67,
+      "eval_accuracy": 0.95,
+      "eval_loss": 0.6299833059310913,
+      "eval_runtime": 0.286,
+      "eval_samples_per_second": 69.919,
+      "eval_steps_per_second": 3.496,
+      "step": 7
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.5069305896759033,
+      "eval_runtime": 0.3181,
+      "eval_samples_per_second": 62.873,
+      "eval_steps_per_second": 3.144,
+      "step": 9
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 1e-05,
+      "loss": 0.8554,
+      "step": 10
+    },
+    {
+      "epoch": 6.67,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.46708640456199646,
+      "eval_runtime": 0.317,
+      "eval_samples_per_second": 63.085,
+      "eval_steps_per_second": 3.154,
+      "step": 10
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.43121227622032166,
+      "eval_runtime": 0.3145,
+      "eval_samples_per_second": 63.594,
+      "eval_steps_per_second": 3.18,
+      "step": 12
+    },
+    {
+      "epoch": 8.0,
+      "step": 12,
+      "total_flos": 1.1159046530924544e+17,
+      "train_loss": 0.795779287815094,
+      "train_runtime": 107.6532,
+      "train_samples_per_second": 20.064,
+      "train_steps_per_second": 0.111
     }
   ],
   "logging_steps": 10,
-  "max_steps": 3,
-  "num_train_epochs": 3,
+  "max_steps": 12,
+  "num_train_epochs": 12,
   "save_steps": 500,
-  "total_flos": 2.789761632731136e+16,
+  "total_flos": 1.1159046530924544e+17,
   "trial_name": null,
   "trial_params": null
 }