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
@@ -16,7 +18,7 @@ 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: 1.0987
 - Accuracy: 0.3383

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.3308270676691729,
-    "eval_loss": 2.0399293899536133,
-    "eval_runtime": 0.9627,
-    "eval_samples_per_second": 138.153,
-    "eval_steps_per_second": 17.659,
+    "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": 0.9533355126014122,
-    "train_runtime": 174.5905,
-    "train_samples_per_second": 59.224,
-    "train_steps_per_second": 7.446
+    "train_loss": 1.1044638824462891,
+    "train_runtime": 173.1626,
+    "train_samples_per_second": 59.713,
+    "train_steps_per_second": 7.507
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.3308270676691729,
-    "eval_loss": 2.0399293899536133,
-    "eval_runtime": 0.9627,
-    "eval_samples_per_second": 138.153,
-    "eval_steps_per_second": 17.659
+    "eval_accuracy": 0.3383458646616541,
+    "eval_loss": 1.098726749420166,
+    "eval_runtime": 0.9888,
+    "eval_samples_per_second": 134.507,
+    "eval_steps_per_second": 17.193
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.9533355126014122,
-    "train_runtime": 174.5905,
-    "train_samples_per_second": 59.224,
-    "train_steps_per_second": 7.446
+    "train_loss": 1.1044638824462891,
+    "train_runtime": 173.1626,
+    "train_samples_per_second": 59.713,
+    "train_steps_per_second": 7.507
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.23085123300552368,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-130",
+  "best_metric": 1.098726749420166,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-1300",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 1300,
@@ -10,193 +10,193 @@
   "log_history": [
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 1.3140615224838257,
+      "grad_norm": 3.964278221130371,
       "learning_rate": 0.005,
-      "loss": 0.4537,
+      "loss": 0.9625,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8947368421052632,
-      "eval_loss": 0.23085123300552368,
-      "eval_runtime": 0.9384,
-      "eval_samples_per_second": 141.736,
-      "eval_steps_per_second": 18.117,
+      "eval_accuracy": 0.3383458646616541,
+      "eval_loss": 1.1997095346450806,
+      "eval_runtime": 0.9373,
+      "eval_samples_per_second": 141.894,
+      "eval_steps_per_second": 18.137,
       "step": 130
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.5738377571105957,
+      "grad_norm": 0.5754929184913635,
       "learning_rate": 0.005,
-      "loss": 0.5034,
+      "loss": 1.2026,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.6165413533834586,
-      "eval_loss": 1.0887765884399414,
-      "eval_runtime": 0.9484,
-      "eval_samples_per_second": 140.237,
-      "eval_steps_per_second": 17.925,
+      "eval_accuracy": 0.3308270676691729,
+      "eval_loss": 1.1204776763916016,
+      "eval_runtime": 0.9634,
+      "eval_samples_per_second": 138.047,
+      "eval_steps_per_second": 17.645,
       "step": 260
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 4.012147426605225,
+      "grad_norm": 0.5505106449127197,
       "learning_rate": 0.005,
-      "loss": 0.506,
+      "loss": 1.1354,
       "step": 300
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.5714285714285714,
-      "eval_loss": 0.9555807709693909,
-      "eval_runtime": 0.9238,
-      "eval_samples_per_second": 143.97,
-      "eval_steps_per_second": 18.402,
+      "eval_accuracy": 0.3308270676691729,
+      "eval_loss": 1.099075198173523,
+      "eval_runtime": 0.9285,
+      "eval_samples_per_second": 143.241,
+      "eval_steps_per_second": 18.309,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 2.720919132232666,
+      "grad_norm": 0.4573482573032379,
       "learning_rate": 0.005,
-      "loss": 0.9151,
+      "loss": 1.1074,
       "step": 400
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.1517225205898285,
+      "grad_norm": 0.1797952502965927,
       "learning_rate": 0.005,
-      "loss": 1.1528,
+      "loss": 1.1071,
       "step": 500
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.1074310541152954,
-      "eval_runtime": 0.9649,
-      "eval_samples_per_second": 137.839,
-      "eval_steps_per_second": 17.618,
+      "eval_loss": 1.101277232170105,
+      "eval_runtime": 0.9468,
+      "eval_samples_per_second": 140.474,
+      "eval_steps_per_second": 17.955,
       "step": 520
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 0.09401340782642365,
+      "grad_norm": 0.0814356580376625,
       "learning_rate": 0.005,
-      "loss": 1.1076,
+      "loss": 1.102,
       "step": 600
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.100753664970398,
-      "eval_runtime": 0.982,
-      "eval_samples_per_second": 135.436,
-      "eval_steps_per_second": 17.311,
+      "eval_loss": 1.0996899604797363,
+      "eval_runtime": 0.994,
+      "eval_samples_per_second": 133.802,
+      "eval_steps_per_second": 17.103,
       "step": 650
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 0.5329942107200623,
+      "grad_norm": 0.49848759174346924,
       "learning_rate": 0.005,
-      "loss": 1.1148,
+      "loss": 1.1064,
       "step": 700
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.1018885374069214,
-      "eval_runtime": 0.9388,
-      "eval_samples_per_second": 141.675,
-      "eval_steps_per_second": 18.109,
+      "eval_accuracy": 0.3383458646616541,
+      "eval_loss": 1.0997720956802368,
+      "eval_runtime": 0.9552,
+      "eval_samples_per_second": 139.237,
+      "eval_steps_per_second": 17.797,
       "step": 780
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 0.4093654453754425,
+      "grad_norm": 0.38272586464881897,
       "learning_rate": 0.005,
-      "loss": 1.1095,
+      "loss": 1.1083,
       "step": 800
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 0.4959578216075897,
+      "grad_norm": 0.37760409712791443,
       "learning_rate": 0.005,
-      "loss": 1.1026,
+      "loss": 1.1008,
       "step": 900
     },
     {
       "epoch": 7.0,
       "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.1080141067504883,
-      "eval_runtime": 1.0873,
-      "eval_samples_per_second": 122.317,
-      "eval_steps_per_second": 15.635,
+      "eval_loss": 1.099931001663208,
+      "eval_runtime": 0.9388,
+      "eval_samples_per_second": 141.67,
+      "eval_steps_per_second": 18.108,
       "step": 910
     },
     {
       "epoch": 7.6923076923076925,
-      "grad_norm": 0.2822917699813843,
+      "grad_norm": 0.26804277300834656,
       "learning_rate": 0.005,
-      "loss": 1.1073,
+      "loss": 1.1015,
       "step": 1000
     },
     {
       "epoch": 8.0,
       "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.0986037254333496,
-      "eval_runtime": 0.9744,
-      "eval_samples_per_second": 136.492,
-      "eval_steps_per_second": 17.446,
+      "eval_loss": 1.0987836122512817,
+      "eval_runtime": 0.9412,
+      "eval_samples_per_second": 141.315,
+      "eval_steps_per_second": 18.063,
       "step": 1040
     },
     {
       "epoch": 8.461538461538462,
-      "grad_norm": 0.33672869205474854,
+      "grad_norm": 0.3317204713821411,
       "learning_rate": 0.005,
-      "loss": 1.1009,
+      "loss": 1.0998,
       "step": 1100
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.3383458646616541,
-      "eval_loss": 1.0998892784118652,
-      "eval_runtime": 0.9516,
-      "eval_samples_per_second": 139.767,
-      "eval_steps_per_second": 17.865,
+      "eval_accuracy": 0.3308270676691729,
+      "eval_loss": 1.1042646169662476,
+      "eval_runtime": 0.9609,
+      "eval_samples_per_second": 138.415,
+      "eval_steps_per_second": 17.692,
       "step": 1170
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 0.08797406405210495,
+      "grad_norm": 0.2213129699230194,
       "learning_rate": 0.005,
-      "loss": 1.1118,
+      "loss": 1.1156,
       "step": 1200
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.42109444737434387,
+      "grad_norm": 0.4260609745979309,
       "learning_rate": 0.005,
-      "loss": 1.1081,
+      "loss": 1.1085,
       "step": 1300
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.098858118057251,
-      "eval_runtime": 0.9588,
-      "eval_samples_per_second": 138.708,
-      "eval_steps_per_second": 17.73,
+      "eval_accuracy": 0.3383458646616541,
+      "eval_loss": 1.098726749420166,
+      "eval_runtime": 0.9647,
+      "eval_samples_per_second": 137.865,
+      "eval_steps_per_second": 17.622,
       "step": 1300
     },
     {
       "epoch": 10.0,
       "step": 1300,
       "total_flos": 8.772706474360013e+17,
-      "train_loss": 0.9533355126014122,
-      "train_runtime": 174.5905,
-      "train_samples_per_second": 59.224,
-      "train_steps_per_second": 7.446
+      "train_loss": 1.1044638824462891,
+      "train_runtime": 173.1626,
+      "train_samples_per_second": 59.713,
+      "train_steps_per_second": 7.507
     }
   ],
   "logging_steps": 100,