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: 0.2267
 - Accuracy: 0.9699

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 7.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.2265840619802475,
-    "eval_runtime": 5.8045,
-    "eval_samples_per_second": 22.913,
-    "eval_steps_per_second": 0.517,
+    "eval_loss": 0.22669997811317444,
+    "eval_runtime": 5.4349,
+    "eval_samples_per_second": 24.472,
+    "eval_steps_per_second": 0.552,
     "total_flos": 5.608920478833746e+17,
-    "train_loss": 0.4664890986530721,
-    "train_runtime": 403.8856,
-    "train_samples_per_second": 17.921,
+    "train_loss": 0.4668571989075476,
+    "train_runtime": 403.193,
+    "train_samples_per_second": 17.952,
     "train_steps_per_second": 0.295
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.2265840619802475,
-    "eval_runtime": 5.8045,
-    "eval_samples_per_second": 22.913,
-    "eval_steps_per_second": 0.517
+    "eval_loss": 0.22669997811317444,
+    "eval_runtime": 5.4349,
+    "eval_samples_per_second": 24.472,
+    "eval_steps_per_second": 0.552
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 7.0,
     "total_flos": 5.608920478833746e+17,
-    "train_loss": 0.4664890986530721,
-    "train_runtime": 403.8856,
-    "train_samples_per_second": 17.921,
+    "train_loss": 0.4668571989075476,
+    "train_runtime": 403.193,
+    "train_samples_per_second": 17.952,
     "train_steps_per_second": 0.295
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.2265840619802475,
+  "best_metric": 0.22669997811317444,
   "best_model_checkpoint": "./beans_outputs/checkpoint-119",
   "epoch": 7.0,
   "eval_steps": 500,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.367475986480713,
+      "grad_norm": 1.3734956979751587,
       "learning_rate": 1.831932773109244e-05,
       "loss": 0.9859,
       "step": 10
@@ -18,143 +18,143 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.9323308270676691,
-      "eval_loss": 0.7493126392364502,
-      "eval_runtime": 6.4143,
-      "eval_samples_per_second": 20.735,
-      "eval_steps_per_second": 0.468,
+      "eval_loss": 0.7495129704475403,
+      "eval_runtime": 5.9299,
+      "eval_samples_per_second": 22.429,
+      "eval_steps_per_second": 0.506,
       "step": 17
     },
     {
       "epoch": 1.1764705882352942,
-      "grad_norm": 1.1856489181518555,
+      "grad_norm": 1.1843513250350952,
       "learning_rate": 1.6638655462184877e-05,
-      "loss": 0.8078,
+      "loss": 0.808,
       "step": 20
     },
     {
       "epoch": 1.7647058823529411,
-      "grad_norm": 1.1680834293365479,
+      "grad_norm": 1.209150791168213,
       "learning_rate": 1.4957983193277313e-05,
-      "loss": 0.6762,
+      "loss": 0.6767,
       "step": 30
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.5282919406890869,
-      "eval_runtime": 5.4369,
-      "eval_samples_per_second": 24.462,
-      "eval_steps_per_second": 0.552,
+      "eval_loss": 0.5292961001396179,
+      "eval_runtime": 5.7583,
+      "eval_samples_per_second": 23.097,
+      "eval_steps_per_second": 0.521,
       "step": 34
     },
     {
       "epoch": 2.3529411764705883,
-      "grad_norm": 1.217616319656372,
+      "grad_norm": 1.226790428161621,
       "learning_rate": 1.3277310924369749e-05,
-      "loss": 0.5495,
+      "loss": 0.5504,
       "step": 40
     },
     {
       "epoch": 2.9411764705882355,
-      "grad_norm": 1.0128166675567627,
+      "grad_norm": 1.0332392454147339,
       "learning_rate": 1.1596638655462186e-05,
-      "loss": 0.4605,
+      "loss": 0.4611,
       "step": 50
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.372103214263916,
-      "eval_runtime": 5.8724,
-      "eval_samples_per_second": 22.648,
-      "eval_steps_per_second": 0.511,
+      "eval_loss": 0.3726067841053009,
+      "eval_runtime": 5.9181,
+      "eval_samples_per_second": 22.473,
+      "eval_steps_per_second": 0.507,
       "step": 51
     },
     {
       "epoch": 3.5294117647058822,
-      "grad_norm": 1.1687066555023193,
+      "grad_norm": 1.1689759492874146,
       "learning_rate": 9.915966386554622e-06,
-      "loss": 0.404,
+      "loss": 0.4046,
       "step": 60
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.29797834157943726,
-      "eval_runtime": 5.8268,
-      "eval_samples_per_second": 22.825,
-      "eval_steps_per_second": 0.515,
+      "eval_loss": 0.2977140247821808,
+      "eval_runtime": 5.7869,
+      "eval_samples_per_second": 22.983,
+      "eval_steps_per_second": 0.518,
       "step": 68
     },
     {
       "epoch": 4.117647058823529,
-      "grad_norm": 1.0509178638458252,
+      "grad_norm": 1.0469905138015747,
       "learning_rate": 8.23529411764706e-06,
-      "loss": 0.3263,
+      "loss": 0.3266,
       "step": 70
     },
     {
       "epoch": 4.705882352941177,
-      "grad_norm": 1.413130521774292,
+      "grad_norm": 1.3861141204833984,
       "learning_rate": 6.5546218487394966e-06,
-      "loss": 0.3171,
+      "loss": 0.3173,
       "step": 80
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.25269007682800293,
-      "eval_runtime": 5.1082,
-      "eval_samples_per_second": 26.037,
-      "eval_steps_per_second": 0.587,
+      "eval_loss": 0.2532085180282593,
+      "eval_runtime": 5.8905,
+      "eval_samples_per_second": 22.579,
+      "eval_steps_per_second": 0.509,
       "step": 85
     },
     {
       "epoch": 5.294117647058823,
-      "grad_norm": 0.9891301393508911,
+      "grad_norm": 1.1056371927261353,
       "learning_rate": 4.873949579831933e-06,
       "loss": 0.2773,
       "step": 90
     },
     {
       "epoch": 5.882352941176471,
-      "grad_norm": 0.8509126305580139,
+      "grad_norm": 0.9023756980895996,
       "learning_rate": 3.1932773109243696e-06,
-      "loss": 0.2545,
+      "loss": 0.2547,
       "step": 100
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.9774436090225563,
-      "eval_loss": 0.22762636840343475,
-      "eval_runtime": 5.6199,
-      "eval_samples_per_second": 23.666,
-      "eval_steps_per_second": 0.534,
+      "eval_loss": 0.22800950706005096,
+      "eval_runtime": 5.8041,
+      "eval_samples_per_second": 22.915,
+      "eval_steps_per_second": 0.517,
       "step": 102
     },
     {
       "epoch": 6.470588235294118,
-      "grad_norm": 1.092388391494751,
+      "grad_norm": 1.1000926494598389,
       "learning_rate": 1.5126050420168068e-06,
-      "loss": 0.2626,
+      "loss": 0.2629,
       "step": 110
     },
     {
       "epoch": 7.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.2265840619802475,
-      "eval_runtime": 5.8553,
-      "eval_samples_per_second": 22.715,
-      "eval_steps_per_second": 0.512,
+      "eval_loss": 0.22669997811317444,
+      "eval_runtime": 6.0113,
+      "eval_samples_per_second": 22.125,
+      "eval_steps_per_second": 0.499,
       "step": 119
     },
     {
       "epoch": 7.0,
       "step": 119,
       "total_flos": 5.608920478833746e+17,
-      "train_loss": 0.4664890986530721,
-      "train_runtime": 403.8856,
-      "train_samples_per_second": 17.921,
+      "train_loss": 0.4668571989075476,
+      "train_runtime": 403.193,
+      "train_samples_per_second": 17.952,
       "train_steps_per_second": 0.295
     }
   ],