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: 0.7070
-- Accuracy: 0.7293
+- Loss: 2.6898
+- Accuracy: 0.3383
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.3308270676691729,
-    "eval_loss": 1.1110202074050903,
-    "eval_runtime": 1.0014,
-    "eval_samples_per_second": 132.816,
-    "eval_steps_per_second": 16.976,
+    "eval_accuracy": 0.3383458646616541,
+    "eval_loss": 2.689824104309082,
+    "eval_runtime": 0.9993,
+    "eval_samples_per_second": 133.095,
+    "eval_steps_per_second": 17.012,
     "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.6310529987628644,
-    "train_runtime": 186.1113,
-    "train_samples_per_second": 55.558,
-    "train_steps_per_second": 6.985
+    "train_loss": 0.3764252471923828,
+    "train_runtime": 205.4123,
+    "train_samples_per_second": 50.338,
+    "train_steps_per_second": 6.329
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.3308270676691729,
-    "eval_loss": 1.1110202074050903,
-    "eval_runtime": 1.0014,
-    "eval_samples_per_second": 132.816,
-    "eval_steps_per_second": 16.976
+    "eval_accuracy": 0.3383458646616541,
+    "eval_loss": 2.689824104309082,
+    "eval_runtime": 0.9993,
+    "eval_samples_per_second": 133.095,
+    "eval_steps_per_second": 17.012
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 8.772706474360013e+17,
-    "train_loss": 0.6310529987628644,
-    "train_runtime": 186.1113,
-    "train_samples_per_second": 55.558,
-    "train_steps_per_second": 6.985
+    "train_loss": 0.3764252471923828,
+    "train_runtime": 205.4123,
+    "train_samples_per_second": 50.338,
+    "train_steps_per_second": 6.329
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.17903044819831848,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-520",
+  "best_metric": 0.0367523655295372,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-390",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 1300,
@@ -10,193 +10,193 @@
   "log_history": [
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 2.7811906337738037,
-      "learning_rate": 5e-05,
-      "loss": 0.9748,
+      "grad_norm": 4.539805889129639,
+      "learning_rate": 0.0005,
+      "loss": 0.4713,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.8947368421052632,
-      "eval_loss": 0.7215397357940674,
-      "eval_runtime": 0.9364,
-      "eval_samples_per_second": 142.034,
-      "eval_steps_per_second": 18.155,
+      "eval_accuracy": 0.9548872180451128,
+      "eval_loss": 0.09497568756341934,
+      "eval_runtime": 1.0331,
+      "eval_samples_per_second": 128.734,
+      "eval_steps_per_second": 16.455,
       "step": 130
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.6656582355499268,
-      "learning_rate": 5e-05,
-      "loss": 0.6849,
+      "grad_norm": 0.8214866518974304,
+      "learning_rate": 0.0005,
+      "loss": 0.1717,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9473684210526315,
-      "eval_loss": 0.4107130765914917,
-      "eval_runtime": 0.9628,
-      "eval_samples_per_second": 138.137,
-      "eval_steps_per_second": 17.657,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.08848864585161209,
+      "eval_runtime": 0.9725,
+      "eval_samples_per_second": 136.757,
+      "eval_steps_per_second": 17.48,
       "step": 260
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.416643738746643,
-      "learning_rate": 5e-05,
-      "loss": 0.4838,
+      "grad_norm": 5.174630165100098,
+      "learning_rate": 0.0005,
+      "loss": 0.1941,
       "step": 300
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9473684210526315,
-      "eval_loss": 0.2423216700553894,
-      "eval_runtime": 0.9821,
-      "eval_samples_per_second": 135.426,
-      "eval_steps_per_second": 17.31,
+      "eval_accuracy": 0.9924812030075187,
+      "eval_loss": 0.0367523655295372,
+      "eval_runtime": 1.0358,
+      "eval_samples_per_second": 128.405,
+      "eval_steps_per_second": 16.413,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 2.6065406799316406,
-      "learning_rate": 5e-05,
-      "loss": 0.3533,
+      "grad_norm": 6.90311336517334,
+      "learning_rate": 0.0005,
+      "loss": 0.178,
       "step": 400
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 1.5872589349746704,
-      "learning_rate": 5e-05,
-      "loss": 0.2594,
+      "grad_norm": 0.0417233482003212,
+      "learning_rate": 0.0005,
+      "loss": 0.1484,
       "step": 500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.17903044819831848,
-      "eval_runtime": 0.9895,
-      "eval_samples_per_second": 134.417,
-      "eval_steps_per_second": 17.181,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.10054013133049011,
+      "eval_runtime": 1.0366,
+      "eval_samples_per_second": 128.308,
+      "eval_steps_per_second": 16.4,
       "step": 520
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 2.6757798194885254,
-      "learning_rate": 5e-05,
-      "loss": 0.2082,
+      "grad_norm": 3.222952365875244,
+      "learning_rate": 0.0005,
+      "loss": 0.1574,
       "step": 600
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9323308270676691,
-      "eval_loss": 0.19146375358104706,
-      "eval_runtime": 1.0393,
-      "eval_samples_per_second": 127.973,
-      "eval_steps_per_second": 16.357,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.06399582326412201,
+      "eval_runtime": 1.1232,
+      "eval_samples_per_second": 118.415,
+      "eval_steps_per_second": 15.136,
       "step": 650
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 5.998991012573242,
-      "learning_rate": 5e-05,
-      "loss": 0.2159,
+      "grad_norm": 0.5326167345046997,
+      "learning_rate": 0.0005,
+      "loss": 0.1948,
       "step": 700
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9097744360902256,
-      "eval_loss": 0.3304254412651062,
-      "eval_runtime": 0.9409,
-      "eval_samples_per_second": 141.349,
-      "eval_steps_per_second": 18.067,
+      "eval_accuracy": 0.9172932330827067,
+      "eval_loss": 0.18919076025485992,
+      "eval_runtime": 1.1717,
+      "eval_samples_per_second": 113.51,
+      "eval_steps_per_second": 14.509,
       "step": 780
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 4.108273029327393,
-      "learning_rate": 5e-05,
-      "loss": 0.339,
+      "grad_norm": 2.2912113666534424,
+      "learning_rate": 0.0005,
+      "loss": 0.2541,
       "step": 800
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 5.80717134475708,
-      "learning_rate": 5e-05,
-      "loss": 0.5954,
+      "grad_norm": 4.357283592224121,
+      "learning_rate": 0.0005,
+      "loss": 0.4159,
       "step": 900
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.6917293233082706,
-      "eval_loss": 0.6861125230789185,
-      "eval_runtime": 1.0533,
-      "eval_samples_per_second": 126.266,
-      "eval_steps_per_second": 16.139,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.15302829444408417,
+      "eval_runtime": 1.1496,
+      "eval_samples_per_second": 115.692,
+      "eval_steps_per_second": 14.788,
       "step": 910
     },
     {
       "epoch": 7.6923076923076925,
-      "grad_norm": 4.3353142738342285,
-      "learning_rate": 5e-05,
-      "loss": 0.8045,
+      "grad_norm": 2.716820240020752,
+      "learning_rate": 0.0005,
+      "loss": 0.4951,
       "step": 1000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.48872180451127817,
-      "eval_loss": 1.0407782793045044,
-      "eval_runtime": 1.1746,
-      "eval_samples_per_second": 113.234,
-      "eval_steps_per_second": 14.474,
+      "eval_accuracy": 0.6842105263157895,
+      "eval_loss": 0.7176167368888855,
+      "eval_runtime": 0.9764,
+      "eval_samples_per_second": 136.21,
+      "eval_steps_per_second": 17.41,
       "step": 1040
     },
     {
       "epoch": 8.461538461538462,
-      "grad_norm": 1.6110053062438965,
-      "learning_rate": 5e-05,
-      "loss": 1.0506,
+      "grad_norm": 1.9368160963058472,
+      "learning_rate": 0.0005,
+      "loss": 0.6818,
       "step": 1100
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.3308270676691729,
-      "eval_loss": 1.1016151905059814,
-      "eval_runtime": 1.2259,
-      "eval_samples_per_second": 108.493,
-      "eval_steps_per_second": 13.868,
+      "eval_accuracy": 0.42857142857142855,
+      "eval_loss": 1.4557244777679443,
+      "eval_runtime": 1.1951,
+      "eval_samples_per_second": 111.285,
+      "eval_steps_per_second": 14.224,
       "step": 1170
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 0.6255662441253662,
-      "learning_rate": 5e-05,
-      "loss": 1.1038,
+      "grad_norm": 1.1992454528808594,
+      "learning_rate": 0.0005,
+      "loss": 0.7561,
       "step": 1200
     },
     {
       "epoch": 10.0,
-      "grad_norm": 6.890110969543457,
-      "learning_rate": 5e-05,
-      "loss": 1.1299,
+      "grad_norm": 3.8898727893829346,
+      "learning_rate": 0.0005,
+      "loss": 0.7749,
       "step": 1300
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.3458646616541353,
-      "eval_loss": 1.0934619903564453,
-      "eval_runtime": 1.1913,
-      "eval_samples_per_second": 111.642,
-      "eval_steps_per_second": 14.27,
+      "eval_accuracy": 0.7293233082706767,
+      "eval_loss": 0.7070097923278809,
+      "eval_runtime": 1.1878,
+      "eval_samples_per_second": 111.974,
+      "eval_steps_per_second": 14.312,
       "step": 1300
     },
     {
       "epoch": 10.0,
       "step": 1300,
       "total_flos": 8.772706474360013e+17,
-      "train_loss": 0.6310529987628644,
-      "train_runtime": 186.1113,
-      "train_samples_per_second": 55.558,
-      "train_steps_per_second": 6.985
+      "train_loss": 0.3764252471923828,
+      "train_runtime": 205.4123,
+      "train_samples_per_second": 50.338,
+      "train_steps_per_second": 6.329
     }
   ],
   "logging_steps": 100,