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: 0.3441
 - Accuracy: 0.9699

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.3435651957988739,
-    "eval_runtime": 13.8807,
-    "eval_samples_per_second": 9.582,
-    "eval_steps_per_second": 0.216,
+    "eval_loss": 0.34406405687332153,
+    "eval_runtime": 1.5551,
+    "eval_samples_per_second": 85.523,
+    "eval_steps_per_second": 1.929,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.5856990561765782,
-    "train_runtime": 600.0005,
-    "train_samples_per_second": 8.617,
-    "train_steps_per_second": 0.142
+    "train_loss": 0.586445689201355,
+    "train_runtime": 117.1883,
+    "train_samples_per_second": 44.117,
+    "train_steps_per_second": 0.725
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9699248120300752,
-    "eval_loss": 0.3435651957988739,
-    "eval_runtime": 13.8807,
-    "eval_samples_per_second": 9.582,
-    "eval_steps_per_second": 0.216
+    "eval_loss": 0.34406405687332153,
+    "eval_runtime": 1.5551,
+    "eval_samples_per_second": 85.523,
+    "eval_steps_per_second": 1.929
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.5856990561765782,
-    "train_runtime": 600.0005,
-    "train_samples_per_second": 8.617,
-    "train_steps_per_second": 0.142
+    "train_loss": 0.586445689201355,
+    "train_runtime": 117.1883,
+    "train_samples_per_second": 44.117,
+    "train_steps_per_second": 0.725
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.3435651957988739,
+  "best_metric": 0.34406405687332153,
   "best_model_checkpoint": "./beans_outputs/checkpoint-85",
   "epoch": 5.0,
   "eval_steps": 500,
@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 0.5882352941176471,
-      "grad_norm": 1.339490532875061,
+      "grad_norm": 1.3438670635223389,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.9867,
       "step": 10
@@ -18,105 +18,105 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.924812030075188,
-      "eval_loss": 0.7569698095321655,
-      "eval_runtime": 13.9686,
-      "eval_samples_per_second": 9.521,
-      "eval_steps_per_second": 0.215,
+      "eval_loss": 0.7574980854988098,
+      "eval_runtime": 1.5987,
+      "eval_samples_per_second": 83.19,
+      "eval_steps_per_second": 1.876,
       "step": 17
     },
     {
       "epoch": 1.1764705882352942,
-      "grad_norm": 1.1882508993148804,
+      "grad_norm": 1.1910630464553833,
       "learning_rate": 1.5294117647058822e-05,
-      "loss": 0.8133,
+      "loss": 0.8136,
       "step": 20
     },
     {
       "epoch": 1.7647058823529411,
-      "grad_norm": 1.160631775856018,
+      "grad_norm": 1.7142910957336426,
       "learning_rate": 1.2941176470588238e-05,
-      "loss": 0.6897,
+      "loss": 0.6904,
       "step": 30
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.5532798767089844,
-      "eval_runtime": 13.9166,
-      "eval_samples_per_second": 9.557,
-      "eval_steps_per_second": 0.216,
+      "eval_loss": 0.5554725527763367,
+      "eval_runtime": 1.5477,
+      "eval_samples_per_second": 85.934,
+      "eval_steps_per_second": 1.938,
       "step": 34
     },
     {
       "epoch": 2.3529411764705883,
-      "grad_norm": 1.2677371501922607,
+      "grad_norm": 1.2150076627731323,
       "learning_rate": 1.0588235294117648e-05,
-      "loss": 0.5754,
+      "loss": 0.5771,
       "step": 40
     },
     {
       "epoch": 2.9411764705882355,
-      "grad_norm": 1.1117501258850098,
+      "grad_norm": 1.0504060983657837,
       "learning_rate": 8.23529411764706e-06,
-      "loss": 0.4982,
+      "loss": 0.5001,
       "step": 50
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.42039212584495544,
-      "eval_runtime": 13.4157,
-      "eval_samples_per_second": 9.914,
-      "eval_steps_per_second": 0.224,
+      "eval_loss": 0.4214977025985718,
+      "eval_runtime": 1.6146,
+      "eval_samples_per_second": 82.375,
+      "eval_steps_per_second": 1.858,
       "step": 51
     },
     {
       "epoch": 3.5294117647058822,
-      "grad_norm": 1.190975308418274,
+      "grad_norm": 1.2046395540237427,
       "learning_rate": 5.882352941176471e-06,
-      "loss": 0.4543,
+      "loss": 0.4553,
       "step": 60
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.36109158396720886,
-      "eval_runtime": 13.6574,
-      "eval_samples_per_second": 9.738,
-      "eval_steps_per_second": 0.22,
+      "eval_loss": 0.3612525165081024,
+      "eval_runtime": 1.5732,
+      "eval_samples_per_second": 84.543,
+      "eval_steps_per_second": 1.907,
       "step": 68
     },
     {
       "epoch": 4.117647058823529,
-      "grad_norm": 1.0600579977035522,
+      "grad_norm": 1.0449424982070923,
       "learning_rate": 3.529411764705883e-06,
-      "loss": 0.391,
+      "loss": 0.3916,
       "step": 70
     },
     {
       "epoch": 4.705882352941177,
-      "grad_norm": 1.121734857559204,
+      "grad_norm": 1.1166667938232422,
       "learning_rate": 1.1764705882352942e-06,
-      "loss": 0.3928,
+      "loss": 0.3931,
       "step": 80
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.3435651957988739,
-      "eval_runtime": 14.9585,
-      "eval_samples_per_second": 8.891,
-      "eval_steps_per_second": 0.201,
+      "eval_loss": 0.34406405687332153,
+      "eval_runtime": 1.5434,
+      "eval_samples_per_second": 86.174,
+      "eval_steps_per_second": 1.944,
       "step": 85
     },
     {
       "epoch": 5.0,
       "step": 85,
       "total_flos": 4.006371770595533e+17,
-      "train_loss": 0.5856990561765782,
-      "train_runtime": 600.0005,
-      "train_samples_per_second": 8.617,
-      "train_steps_per_second": 0.142
+      "train_loss": 0.586445689201355,
+      "train_runtime": 117.1883,
+      "train_samples_per_second": 44.117,
+      "train_steps_per_second": 0.725
     }
   ],
   "logging_steps": 10,