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: 0.1528
-- Accuracy: 0.9474
+- Loss: 0.0752
+- Accuracy: 0.9624
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.3383458646616541,
-    "eval_loss": 1.098726749420166,
-    "eval_runtime": 0.9888,
-    "eval_samples_per_second": 134.507,
-    "eval_steps_per_second": 17.193,
-    "total_flos": 8.772706474360013e+17,
-    "train_loss": 1.1044638824462891,
-    "train_runtime": 173.1626,
-    "train_samples_per_second": 59.713,
-    "train_steps_per_second": 7.507
+    "eval_accuracy": 0.9624060150375939,
+    "eval_loss": 0.0752067118883133,
+    "eval_runtime": 0.7899,
+    "eval_samples_per_second": 168.383,
+    "eval_steps_per_second": 21.523,
+    "total_flos": 8.508699063477043e+17,
+    "train_loss": 0.2738876885634202,
+    "train_runtime": 131.5423,
+    "train_samples_per_second": 78.606,
+    "train_steps_per_second": 9.883
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.3383458646616541,
-    "eval_loss": 1.098726749420166,
-    "eval_runtime": 0.9888,
-    "eval_samples_per_second": 134.507,
-    "eval_steps_per_second": 17.193
+    "eval_accuracy": 0.9624060150375939,
+    "eval_loss": 0.0752067118883133,
+    "eval_runtime": 0.7899,
+    "eval_samples_per_second": 168.383,
+    "eval_steps_per_second": 21.523
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "total_flos": 8.772706474360013e+17,
-    "train_loss": 1.1044638824462891,
-    "train_runtime": 173.1626,
-    "train_samples_per_second": 59.713,
-    "train_steps_per_second": 7.507
+    "total_flos": 8.508699063477043e+17,
+    "train_loss": 0.2738876885634202,
+    "train_runtime": 131.5423,
+    "train_samples_per_second": 78.606,
+    "train_steps_per_second": 9.883
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 1.098726749420166,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-1300",
+  "best_metric": 0.0752067118883133,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-130",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 1300,
@@ -10,193 +10,193 @@
   "log_history": [
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 3.964278221130371,
+      "grad_norm": 4.380956172943115,
       "learning_rate": 0.005,
-      "loss": 0.9625,
+      "loss": 0.4846,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.3383458646616541,
-      "eval_loss": 1.1997095346450806,
-      "eval_runtime": 0.9373,
-      "eval_samples_per_second": 141.894,
-      "eval_steps_per_second": 18.137,
+      "eval_accuracy": 0.9624060150375939,
+      "eval_loss": 0.0752067118883133,
+      "eval_runtime": 0.7531,
+      "eval_samples_per_second": 176.6,
+      "eval_steps_per_second": 22.573,
       "step": 130
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 0.5754929184913635,
+      "grad_norm": 0.7832342982292175,
       "learning_rate": 0.005,
-      "loss": 1.2026,
+      "loss": 0.3473,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.1204776763916016,
-      "eval_runtime": 0.9634,
-      "eval_samples_per_second": 138.047,
-      "eval_steps_per_second": 17.645,
+      "eval_accuracy": 0.9548872180451128,
+      "eval_loss": 0.15986023843288422,
+      "eval_runtime": 0.7821,
+      "eval_samples_per_second": 170.05,
+      "eval_steps_per_second": 21.736,
       "step": 260
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 0.5505106449127197,
+      "grad_norm": 1.9846404790878296,
       "learning_rate": 0.005,
-      "loss": 1.1354,
+      "loss": 0.2953,
       "step": 300
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.099075198173523,
-      "eval_runtime": 0.9285,
-      "eval_samples_per_second": 143.241,
-      "eval_steps_per_second": 18.309,
+      "eval_accuracy": 0.9548872180451128,
+      "eval_loss": 0.1191997155547142,
+      "eval_runtime": 0.8019,
+      "eval_samples_per_second": 165.864,
+      "eval_steps_per_second": 21.201,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 0.4573482573032379,
+      "grad_norm": 1.4397724866867065,
       "learning_rate": 0.005,
-      "loss": 1.1074,
+      "loss": 0.2891,
       "step": 400
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.1797952502965927,
+      "grad_norm": 3.0908043384552,
       "learning_rate": 0.005,
-      "loss": 1.1071,
+      "loss": 0.2653,
       "step": 500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.101277232170105,
-      "eval_runtime": 0.9468,
-      "eval_samples_per_second": 140.474,
-      "eval_steps_per_second": 17.955,
+      "eval_accuracy": 0.9398496240601504,
+      "eval_loss": 0.13930319249629974,
+      "eval_runtime": 0.7871,
+      "eval_samples_per_second": 168.974,
+      "eval_steps_per_second": 21.598,
       "step": 520
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 0.0814356580376625,
+      "grad_norm": 2.720701217651367,
       "learning_rate": 0.005,
-      "loss": 1.102,
+      "loss": 0.2344,
       "step": 600
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.0996899604797363,
-      "eval_runtime": 0.994,
-      "eval_samples_per_second": 133.802,
-      "eval_steps_per_second": 17.103,
+      "eval_accuracy": 0.9624060150375939,
+      "eval_loss": 0.10011590272188187,
+      "eval_runtime": 0.8095,
+      "eval_samples_per_second": 164.306,
+      "eval_steps_per_second": 21.001,
       "step": 650
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 0.49848759174346924,
+      "grad_norm": 0.17460593581199646,
       "learning_rate": 0.005,
-      "loss": 1.1064,
+      "loss": 0.21,
       "step": 700
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.3383458646616541,
-      "eval_loss": 1.0997720956802368,
-      "eval_runtime": 0.9552,
-      "eval_samples_per_second": 139.237,
-      "eval_steps_per_second": 17.797,
+      "eval_accuracy": 0.9624060150375939,
+      "eval_loss": 0.0893179401755333,
+      "eval_runtime": 0.7824,
+      "eval_samples_per_second": 169.993,
+      "eval_steps_per_second": 21.728,
       "step": 780
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 0.38272586464881897,
+      "grad_norm": 0.6050938963890076,
       "learning_rate": 0.005,
-      "loss": 1.1083,
+      "loss": 0.2397,
       "step": 800
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 0.37760409712791443,
+      "grad_norm": 1.7308460474014282,
       "learning_rate": 0.005,
-      "loss": 1.1008,
+      "loss": 0.3117,
       "step": 900
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.099931001663208,
-      "eval_runtime": 0.9388,
-      "eval_samples_per_second": 141.67,
-      "eval_steps_per_second": 18.108,
+      "eval_accuracy": 0.924812030075188,
+      "eval_loss": 0.1933087259531021,
+      "eval_runtime": 0.7988,
+      "eval_samples_per_second": 166.494,
+      "eval_steps_per_second": 21.281,
       "step": 910
     },
     {
       "epoch": 7.6923076923076925,
-      "grad_norm": 0.26804277300834656,
+      "grad_norm": 1.327934741973877,
       "learning_rate": 0.005,
-      "loss": 1.1015,
+      "loss": 0.2459,
       "step": 1000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.0987836122512817,
-      "eval_runtime": 0.9412,
-      "eval_samples_per_second": 141.315,
-      "eval_steps_per_second": 18.063,
+      "eval_accuracy": 0.924812030075188,
+      "eval_loss": 0.19011461734771729,
+      "eval_runtime": 0.7881,
+      "eval_samples_per_second": 168.75,
+      "eval_steps_per_second": 21.57,
       "step": 1040
     },
     {
       "epoch": 8.461538461538462,
-      "grad_norm": 0.3317204713821411,
+      "grad_norm": 0.8504745364189148,
       "learning_rate": 0.005,
-      "loss": 1.0998,
+      "loss": 0.25,
       "step": 1100
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.1042646169662476,
-      "eval_runtime": 0.9609,
-      "eval_samples_per_second": 138.415,
-      "eval_steps_per_second": 17.692,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.08675242960453033,
+      "eval_runtime": 0.7998,
+      "eval_samples_per_second": 166.287,
+      "eval_steps_per_second": 21.255,
       "step": 1170
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 0.2213129699230194,
+      "grad_norm": 3.838754177093506,
       "learning_rate": 0.005,
-      "loss": 1.1156,
+      "loss": 0.1834,
       "step": 1200
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.4260609745979309,
+      "grad_norm": 0.03880231827497482,
       "learning_rate": 0.005,
-      "loss": 1.1085,
+      "loss": 0.2038,
       "step": 1300
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.3383458646616541,
-      "eval_loss": 1.098726749420166,
-      "eval_runtime": 0.9647,
-      "eval_samples_per_second": 137.865,
-      "eval_steps_per_second": 17.622,
+      "eval_accuracy": 0.9473684210526315,
+      "eval_loss": 0.15278470516204834,
+      "eval_runtime": 0.8065,
+      "eval_samples_per_second": 164.915,
+      "eval_steps_per_second": 21.079,
       "step": 1300
     },
     {
       "epoch": 10.0,
       "step": 1300,
-      "total_flos": 8.772706474360013e+17,
-      "train_loss": 1.1044638824462891,
-      "train_runtime": 173.1626,
-      "train_samples_per_second": 59.713,
-      "train_steps_per_second": 7.507
+      "total_flos": 8.508699063477043e+17,
+      "train_loss": 0.2738876885634202,
+      "train_runtime": 131.5423,
+      "train_samples_per_second": 78.606,
+      "train_steps_per_second": 9.883
     }
   ],
   "logging_steps": 100,
@@ -216,7 +216,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.772706474360013e+17,
+  "total_flos": 8.508699063477043e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null