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: 0.3439
 - Accuracy: 0.9699

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.34362348914146423,
-    "eval_runtime": 1.5507,
-    "eval_samples_per_second": 85.77,
-    "eval_steps_per_second": 1.935,
+    "eval_loss": 0.34387466311454773,
+    "eval_runtime": 14.0968,
+    "eval_samples_per_second": 9.435,
+    "eval_steps_per_second": 0.213,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.5853377566618078,
-    "train_runtime": 120.1455,
-    "train_samples_per_second": 43.031,
-    "train_steps_per_second": 0.707
+    "train_loss": 0.5856448944877176,
+    "train_runtime": 608.2336,
+    "train_samples_per_second": 8.5,
+    "train_steps_per_second": 0.14
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.34362348914146423,
-    "eval_runtime": 1.5507,
-    "eval_samples_per_second": 85.77,
-    "eval_steps_per_second": 1.935
+    "eval_loss": 0.34387466311454773,
+    "eval_runtime": 14.0968,
+    "eval_samples_per_second": 9.435,
+    "eval_steps_per_second": 0.213
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.5853377566618078,
-    "train_runtime": 120.1455,
-    "train_samples_per_second": 43.031,
-    "train_steps_per_second": 0.707
+    "train_loss": 0.5856448944877176,
+    "train_runtime": 608.2336,
+    "train_samples_per_second": 8.5,
+    "train_steps_per_second": 0.14
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.34362348914146423,
+  "best_metric": 0.34387466311454773,
   "best_model_checkpoint": "./beans_outputs/checkpoint-85",
   "epoch": 5.0,
   "eval_steps": 500,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.344361662864685,
+      "grad_norm": 1.3442401885986328,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.9867,
       "step": 10
@@ -18,22 +18,22 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.924812030075188,
-      "eval_loss": 0.7569556832313538,
-      "eval_runtime": 1.7538,
-      "eval_samples_per_second": 75.835,
-      "eval_steps_per_second": 1.711,
+      "eval_loss": 0.7572791576385498,
+      "eval_runtime": 14.3345,
+      "eval_samples_per_second": 9.278,
+      "eval_steps_per_second": 0.209,
       "step": 17
     },
     {
       "epoch": 1.1764705882352942,
-      "grad_norm": 1.187711477279663,
+      "grad_norm": 1.1912851333618164,
       "learning_rate": 1.5294117647058822e-05,
       "loss": 0.8134,
       "step": 20
     },
     {
       "epoch": 1.7647058823529411,
-      "grad_norm": 1.2221342325210571,
+      "grad_norm": 1.3143070936203003,
       "learning_rate": 1.2941176470588238e-05,
       "loss": 0.6896,
       "step": 30
@@ -41,82 +41,82 @@
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.5536296963691711,
-      "eval_runtime": 1.4943,
-      "eval_samples_per_second": 89.005,
-      "eval_steps_per_second": 2.008,
+      "eval_loss": 0.5535997748374939,
+      "eval_runtime": 13.8121,
+      "eval_samples_per_second": 9.629,
+      "eval_steps_per_second": 0.217,
       "step": 34
     },
     {
       "epoch": 2.3529411764705883,
-      "grad_norm": 1.2296907901763916,
+      "grad_norm": 1.235939860343933,
       "learning_rate": 1.0588235294117648e-05,
-      "loss": 0.5751,
+      "loss": 0.5754,
       "step": 40
     },
     {
       "epoch": 2.9411764705882355,
-      "grad_norm": 1.0475002527236938,
+      "grad_norm": 1.0637030601501465,
       "learning_rate": 8.23529411764706e-06,
-      "loss": 0.4977,
+      "loss": 0.498,
       "step": 50
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.41969558596611023,
-      "eval_runtime": 1.5376,
-      "eval_samples_per_second": 86.499,
-      "eval_steps_per_second": 1.951,
+      "eval_loss": 0.4194970428943634,
+      "eval_runtime": 14.2589,
+      "eval_samples_per_second": 9.328,
+      "eval_steps_per_second": 0.21,
       "step": 51
     },
     {
       "epoch": 3.5294117647058822,
-      "grad_norm": 1.2174403667449951,
+      "grad_norm": 1.221981167793274,
       "learning_rate": 5.882352941176471e-06,
-      "loss": 0.4535,
+      "loss": 0.4537,
       "step": 60
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.3606758713722229,
-      "eval_runtime": 1.516,
-      "eval_samples_per_second": 87.731,
-      "eval_steps_per_second": 1.979,
+      "eval_loss": 0.36039993166923523,
+      "eval_runtime": 14.1856,
+      "eval_samples_per_second": 9.376,
+      "eval_steps_per_second": 0.211,
       "step": 68
     },
     {
       "epoch": 4.117647058823529,
-      "grad_norm": 1.0585979223251343,
+      "grad_norm": 1.0668258666992188,
       "learning_rate": 3.529411764705883e-06,
-      "loss": 0.3906,
+      "loss": 0.3909,
       "step": 70
     },
     {
       "epoch": 4.705882352941177,
-      "grad_norm": 1.2112836837768555,
+      "grad_norm": 1.3315623998641968,
       "learning_rate": 1.1764705882352942e-06,
-      "loss": 0.3919,
+      "loss": 0.3927,
       "step": 80
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.34362348914146423,
-      "eval_runtime": 1.644,
-      "eval_samples_per_second": 80.903,
-      "eval_steps_per_second": 1.825,
+      "eval_loss": 0.34387466311454773,
+      "eval_runtime": 13.7327,
+      "eval_samples_per_second": 9.685,
+      "eval_steps_per_second": 0.218,
       "step": 85
     },
     {
       "epoch": 5.0,
       "step": 85,
       "total_flos": 4.006371770595533e+17,
-      "train_loss": 0.5853377566618078,
-      "train_runtime": 120.1455,
-      "train_samples_per_second": 43.031,
-      "train_steps_per_second": 0.707
+      "train_loss": 0.5856448944877176,
+      "train_runtime": 608.2336,
+      "train_samples_per_second": 8.5,
+      "train_steps_per_second": 0.14
     }
   ],
   "logging_steps": 10,