End of training

README.md

@@ -5,6 +5,8 @@ license: apache-2.0
 metrics:
 - accuracy
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 model-index:
 - name: only-lora-beans-vit-base-patch16-224-in21k
@@ -16,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # only-lora-beans-vit-base-patch16-224-in21k
 
-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: 1.0774
-- Accuracy: 0.4135
+- Loss: 1.8956
+- Accuracy: 0.3308
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.5413533834586466,
-    "eval_loss": 0.9381032586097717,
-    "eval_runtime": 0.9636,
-    "eval_samples_per_second": 138.022,
-    "eval_steps_per_second": 17.642,
+    "eval_accuracy": 0.3308270676691729,
+    "eval_loss": 1.895585536956787,
+    "eval_runtime": 0.9592,
+    "eval_samples_per_second": 138.655,
+    "eval_steps_per_second": 17.723,
     "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.9952024547870343,
-    "train_runtime": 174.8367,
-    "train_samples_per_second": 59.141,
-    "train_steps_per_second": 7.436
+    "train_loss": 0.7308059575007512,
+    "train_runtime": 175.3451,
+    "train_samples_per_second": 58.969,
+    "train_steps_per_second": 7.414
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.5413533834586466,
-    "eval_loss": 0.9381032586097717,
-    "eval_runtime": 0.9636,
-    "eval_samples_per_second": 138.022,
-    "eval_steps_per_second": 17.642
+    "eval_accuracy": 0.3308270676691729,
+    "eval_loss": 1.895585536956787,
+    "eval_runtime": 0.9592,
+    "eval_samples_per_second": 138.655,
+    "eval_steps_per_second": 17.723
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.9952024547870343,
-    "train_runtime": 174.8367,
-    "train_samples_per_second": 59.141,
-    "train_steps_per_second": 7.436
+    "train_loss": 0.7308059575007512,
+    "train_runtime": 175.3451,
+    "train_samples_per_second": 58.969,
+    "train_steps_per_second": 7.414
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9381032586097717,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-1300",
+  "best_metric": 0.3044010102748871,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-650",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 1300,
@@ -10,193 +10,193 @@
   "log_history": [
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 2.513921022415161,
-      "learning_rate": 4.615384615384616e-05,
-      "loss": 1.1019,
+      "grad_norm": 2.5466179847717285,
+      "learning_rate": 5e-05,
+      "loss": 1.0208,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.3007518796992481,
-      "eval_loss": 1.0955818891525269,
-      "eval_runtime": 0.9308,
-      "eval_samples_per_second": 142.882,
-      "eval_steps_per_second": 18.263,
+      "eval_accuracy": 0.8045112781954887,
+      "eval_loss": 0.8930914998054504,
+      "eval_runtime": 0.9307,
+      "eval_samples_per_second": 142.898,
+      "eval_steps_per_second": 18.265,
       "step": 130
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 0.6020777225494385,
-      "learning_rate": 4.230769230769231e-05,
-      "loss": 1.0934,
+      "grad_norm": 1.4668034315109253,
+      "learning_rate": 5e-05,
+      "loss": 0.8683,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5413533834586466,
-      "eval_loss": 1.0218424797058105,
-      "eval_runtime": 0.9394,
-      "eval_samples_per_second": 141.582,
-      "eval_steps_per_second": 18.097,
+      "eval_accuracy": 0.8796992481203008,
+      "eval_loss": 0.6855143308639526,
+      "eval_runtime": 0.9382,
+      "eval_samples_per_second": 141.765,
+      "eval_steps_per_second": 18.12,
       "step": 260
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.3475981950759888,
-      "learning_rate": 3.846153846153846e-05,
-      "loss": 1.0541,
+      "grad_norm": 1.5885818004608154,
+      "learning_rate": 5e-05,
+      "loss": 0.7297,
       "step": 300
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5037593984962406,
-      "eval_loss": 0.9863384366035461,
-      "eval_runtime": 0.961,
-      "eval_samples_per_second": 138.393,
-      "eval_steps_per_second": 17.689,
+      "eval_accuracy": 0.9022556390977443,
+      "eval_loss": 0.5008501410484314,
+      "eval_runtime": 0.9428,
+      "eval_samples_per_second": 141.075,
+      "eval_steps_per_second": 18.032,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 1.3711971044540405,
-      "learning_rate": 3.461538461538462e-05,
-      "loss": 0.9991,
+      "grad_norm": 2.4792442321777344,
+      "learning_rate": 5e-05,
+      "loss": 0.6084,
       "step": 400
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.6074744462966919,
-      "learning_rate": 3.0769230769230774e-05,
-      "loss": 0.9911,
+      "grad_norm": 1.0500737428665161,
+      "learning_rate": 5e-05,
+      "loss": 0.4881,
       "step": 500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.5338345864661654,
-      "eval_loss": 0.9627529382705688,
-      "eval_runtime": 0.9661,
-      "eval_samples_per_second": 137.661,
-      "eval_steps_per_second": 17.596,
+      "eval_accuracy": 0.9323308270676691,
+      "eval_loss": 0.3685053288936615,
+      "eval_runtime": 0.9444,
+      "eval_samples_per_second": 140.829,
+      "eval_steps_per_second": 18.001,
       "step": 520
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 0.4636833667755127,
-      "learning_rate": 2.6923076923076923e-05,
-      "loss": 0.9776,
+      "grad_norm": 0.9487807154655457,
+      "learning_rate": 5e-05,
+      "loss": 0.3733,
       "step": 600
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.5413533834586466,
-      "eval_loss": 0.9637588858604431,
-      "eval_runtime": 0.96,
-      "eval_samples_per_second": 138.544,
-      "eval_steps_per_second": 17.709,
+      "eval_accuracy": 0.8872180451127819,
+      "eval_loss": 0.3044010102748871,
+      "eval_runtime": 0.9975,
+      "eval_samples_per_second": 133.327,
+      "eval_steps_per_second": 17.042,
       "step": 650
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 1.6218713521957397,
-      "learning_rate": 2.307692307692308e-05,
-      "loss": 0.9852,
+      "grad_norm": 2.811962604522705,
+      "learning_rate": 5e-05,
+      "loss": 0.3229,
       "step": 700
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.5413533834586466,
-      "eval_loss": 0.9598925113677979,
-      "eval_runtime": 0.956,
-      "eval_samples_per_second": 139.127,
-      "eval_steps_per_second": 17.783,
+      "eval_accuracy": 0.8646616541353384,
+      "eval_loss": 0.3873802721500397,
+      "eval_runtime": 0.9366,
+      "eval_samples_per_second": 142.002,
+      "eval_steps_per_second": 18.151,
       "step": 780
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 1.7121293544769287,
-      "learning_rate": 1.923076923076923e-05,
-      "loss": 0.9753,
+      "grad_norm": 2.758183717727661,
+      "learning_rate": 5e-05,
+      "loss": 0.4189,
       "step": 800
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 1.5765596628189087,
-      "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.9666,
+      "grad_norm": 5.297445297241211,
+      "learning_rate": 5e-05,
+      "loss": 0.5796,
       "step": 900
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.5639097744360902,
-      "eval_loss": 0.9482960104942322,
-      "eval_runtime": 0.9859,
-      "eval_samples_per_second": 134.905,
-      "eval_steps_per_second": 17.244,
+      "eval_accuracy": 0.5939849624060151,
+      "eval_loss": 0.7966009974479675,
+      "eval_runtime": 0.9577,
+      "eval_samples_per_second": 138.872,
+      "eval_steps_per_second": 17.75,
       "step": 910
     },
     {
       "epoch": 7.6923076923076925,
-      "grad_norm": 1.4977715015411377,
-      "learning_rate": 1.153846153846154e-05,
-      "loss": 0.9528,
+      "grad_norm": 28.03814697265625,
+      "learning_rate": 5e-05,
+      "loss": 0.765,
       "step": 1000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.5413533834586466,
-      "eval_loss": 0.9446954727172852,
-      "eval_runtime": 0.9431,
-      "eval_samples_per_second": 141.029,
-      "eval_steps_per_second": 18.026,
+      "eval_accuracy": 0.5263157894736842,
+      "eval_loss": 0.9421801567077637,
+      "eval_runtime": 0.9477,
+      "eval_samples_per_second": 140.335,
+      "eval_steps_per_second": 17.938,
       "step": 1040
     },
     {
       "epoch": 8.461538461538462,
-      "grad_norm": 1.6241708993911743,
-      "learning_rate": 7.692307692307694e-06,
-      "loss": 0.9583,
+      "grad_norm": 5.901334285736084,
+      "learning_rate": 5e-05,
+      "loss": 1.0732,
       "step": 1100
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.5338345864661654,
-      "eval_loss": 0.9394896030426025,
-      "eval_runtime": 0.9745,
-      "eval_samples_per_second": 136.479,
-      "eval_steps_per_second": 17.445,
+      "eval_accuracy": 0.3308270676691729,
+      "eval_loss": 1.0903689861297607,
+      "eval_runtime": 0.9358,
+      "eval_samples_per_second": 142.119,
+      "eval_steps_per_second": 18.166,
       "step": 1170
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 0.5392513871192932,
-      "learning_rate": 3.846153846153847e-06,
-      "loss": 0.9421,
+      "grad_norm": 1.3560993671417236,
+      "learning_rate": 5e-05,
+      "loss": 1.0879,
       "step": 1200
     },
     {
       "epoch": 10.0,
-      "grad_norm": 4.028888702392578,
-      "learning_rate": 0.0,
-      "loss": 0.9401,
+      "grad_norm": 37.35905456542969,
+      "learning_rate": 5e-05,
+      "loss": 1.1643,
       "step": 1300
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.5413533834586466,
-      "eval_loss": 0.9381032586097717,
-      "eval_runtime": 0.9652,
-      "eval_samples_per_second": 137.798,
-      "eval_steps_per_second": 17.613,
+      "eval_accuracy": 0.41353383458646614,
+      "eval_loss": 1.0773515701293945,
+      "eval_runtime": 0.9946,
+      "eval_samples_per_second": 133.719,
+      "eval_steps_per_second": 17.092,
       "step": 1300
     },
     {
       "epoch": 10.0,
       "step": 1300,
       "total_flos": 8.772706474360013e+17,
-      "train_loss": 0.9952024547870343,
-      "train_runtime": 174.8367,
-      "train_samples_per_second": 59.141,
-      "train_steps_per_second": 7.436
+      "train_loss": 0.7308059575007512,
+      "train_runtime": 175.3451,
+      "train_samples_per_second": 58.969,
+      "train_steps_per_second": 7.414
     }
   ],
   "logging_steps": 100,