End of training

README.md

@@ -2,6 +2,8 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 metrics:
 - accuracy
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-beans
 
-This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.3436
 - Accuracy: 0.9699

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.3441993296146393,
-    "eval_runtime": 1.5082,
-    "eval_samples_per_second": 88.182,
-    "eval_steps_per_second": 1.989,
+    "eval_loss": 0.34362348914146423,
+    "eval_runtime": 1.5507,
+    "eval_samples_per_second": 85.77,
+    "eval_steps_per_second": 1.935,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.5858797353856704,
-    "train_runtime": 118.8512,
-    "train_samples_per_second": 43.5,
-    "train_steps_per_second": 0.715
+    "train_loss": 0.5853377566618078,
+    "train_runtime": 120.1455,
+    "train_samples_per_second": 43.031,
+    "train_steps_per_second": 0.707
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.3441993296146393,
-    "eval_runtime": 1.5082,
-    "eval_samples_per_second": 88.182,
-    "eval_steps_per_second": 1.989
+    "eval_loss": 0.34362348914146423,
+    "eval_runtime": 1.5507,
+    "eval_samples_per_second": 85.77,
+    "eval_steps_per_second": 1.935
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.5858797353856704,
-    "train_runtime": 118.8512,
-    "train_samples_per_second": 43.5,
-    "train_steps_per_second": 0.715
+    "train_loss": 0.5853377566618078,
+    "train_runtime": 120.1455,
+    "train_samples_per_second": 43.031,
+    "train_steps_per_second": 0.707
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.3441993296146393,
+  "best_metric": 0.34362348914146423,
   "best_model_checkpoint": "./beans_outputs/checkpoint-85",
   "epoch": 5.0,
   "eval_steps": 500,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.339841604232788,
+      "grad_norm": 1.344361662864685,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.9867,
       "step": 10
@@ -18,105 +18,105 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.924812030075188,
-      "eval_loss": 0.756670355796814,
-      "eval_runtime": 1.784,
-      "eval_samples_per_second": 74.551,
-      "eval_steps_per_second": 1.682,
+      "eval_loss": 0.7569556832313538,
+      "eval_runtime": 1.7538,
+      "eval_samples_per_second": 75.835,
+      "eval_steps_per_second": 1.711,
       "step": 17
     },
     {
       "epoch": 1.1764705882352942,
-      "grad_norm": 1.1886072158813477,
+      "grad_norm": 1.187711477279663,
       "learning_rate": 1.5294117647058822e-05,
-      "loss": 0.8132,
+      "loss": 0.8134,
       "step": 20
     },
     {
       "epoch": 1.7647058823529411,
-      "grad_norm": 1.1582683324813843,
+      "grad_norm": 1.2221342325210571,
       "learning_rate": 1.2941176470588238e-05,
-      "loss": 0.6898,
+      "loss": 0.6896,
       "step": 30
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.5538109540939331,
-      "eval_runtime": 1.7431,
-      "eval_samples_per_second": 76.301,
-      "eval_steps_per_second": 1.721,
+      "eval_loss": 0.5536296963691711,
+      "eval_runtime": 1.4943,
+      "eval_samples_per_second": 89.005,
+      "eval_steps_per_second": 2.008,
       "step": 34
     },
     {
       "epoch": 2.3529411764705883,
-      "grad_norm": 1.224738597869873,
+      "grad_norm": 1.2296907901763916,
       "learning_rate": 1.0588235294117648e-05,
-      "loss": 0.576,
+      "loss": 0.5751,
       "step": 40
     },
     {
       "epoch": 2.9411764705882355,
-      "grad_norm": 1.0394960641860962,
+      "grad_norm": 1.0475002527236938,
       "learning_rate": 8.23529411764706e-06,
-      "loss": 0.4986,
+      "loss": 0.4977,
       "step": 50
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.42004239559173584,
-      "eval_runtime": 1.7355,
-      "eval_samples_per_second": 76.635,
-      "eval_steps_per_second": 1.729,
+      "eval_loss": 0.41969558596611023,
+      "eval_runtime": 1.5376,
+      "eval_samples_per_second": 86.499,
+      "eval_steps_per_second": 1.951,
       "step": 51
     },
     {
       "epoch": 3.5294117647058822,
-      "grad_norm": 1.3868175745010376,
+      "grad_norm": 1.2174403667449951,
       "learning_rate": 5.882352941176471e-06,
-      "loss": 0.4546,
+      "loss": 0.4535,
       "step": 60
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.36084505915641785,
-      "eval_runtime": 1.7426,
-      "eval_samples_per_second": 76.323,
-      "eval_steps_per_second": 1.722,
+      "eval_loss": 0.3606758713722229,
+      "eval_runtime": 1.516,
+      "eval_samples_per_second": 87.731,
+      "eval_steps_per_second": 1.979,
       "step": 68
     },
     {
       "epoch": 4.117647058823529,
-      "grad_norm": 1.0660698413848877,
+      "grad_norm": 1.0585979223251343,
       "learning_rate": 3.529411764705883e-06,
-      "loss": 0.391,
+      "loss": 0.3906,
       "step": 70
     },
     {
       "epoch": 4.705882352941177,
-      "grad_norm": 1.1180354356765747,
+      "grad_norm": 1.2112836837768555,
       "learning_rate": 1.1764705882352942e-06,
-      "loss": 0.3928,
+      "loss": 0.3919,
       "step": 80
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.3441993296146393,
-      "eval_runtime": 1.9856,
-      "eval_samples_per_second": 66.981,
-      "eval_steps_per_second": 1.511,
+      "eval_loss": 0.34362348914146423,
+      "eval_runtime": 1.644,
+      "eval_samples_per_second": 80.903,
+      "eval_steps_per_second": 1.825,
       "step": 85
     },
     {
       "epoch": 5.0,
       "step": 85,
       "total_flos": 4.006371770595533e+17,
-      "train_loss": 0.5858797353856704,
-      "train_runtime": 118.8512,
-      "train_samples_per_second": 43.5,
-      "train_steps_per_second": 0.715
+      "train_loss": 0.5853377566618078,
+      "train_runtime": 120.1455,
+      "train_samples_per_second": 43.031,
+      "train_steps_per_second": 0.707
     }
   ],
   "logging_steps": 10,