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: 0.7512
 - Accuracy: 0.6541

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "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_accuracy": 0.6541353383458647,
+    "eval_loss": 0.751246988773346,
+    "eval_runtime": 0.982,
+    "eval_samples_per_second": 135.444,
+    "eval_steps_per_second": 17.312,
+    "total_flos": 8.772706474360013e+17,
+    "train_loss": 0.9225140674297626,
+    "train_runtime": 175.6098,
+    "train_samples_per_second": 58.881,
+    "train_steps_per_second": 7.403
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.9624060150375939,
-    "eval_loss": 0.0752067118883133,
-    "eval_runtime": 0.7899,
-    "eval_samples_per_second": 168.383,
-    "eval_steps_per_second": 21.523
+    "eval_accuracy": 0.6541353383458647,
+    "eval_loss": 0.751246988773346,
+    "eval_runtime": 0.982,
+    "eval_samples_per_second": 135.444,
+    "eval_steps_per_second": 17.312
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "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
+    "total_flos": 8.772706474360013e+17,
+    "train_loss": 0.9225140674297626,
+    "train_runtime": 175.6098,
+    "train_samples_per_second": 58.881,
+    "train_steps_per_second": 7.403
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.0752067118883133,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-130",
+  "best_metric": 0.751246988773346,
+  "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": 4.380956172943115,
-      "learning_rate": 0.005,
-      "loss": 0.4846,
+      "grad_norm": 2.51617431640625,
+      "learning_rate": 5e-05,
+      "loss": 1.1019,
       "step": 100
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.0752067118883133,
-      "eval_runtime": 0.7531,
-      "eval_samples_per_second": 176.6,
-      "eval_steps_per_second": 22.573,
+      "eval_accuracy": 0.41353383458646614,
+      "eval_loss": 1.094665288925171,
+      "eval_runtime": 0.9302,
+      "eval_samples_per_second": 142.976,
+      "eval_steps_per_second": 18.275,
       "step": 130
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 0.7832342982292175,
-      "learning_rate": 0.005,
-      "loss": 0.3473,
+      "grad_norm": 0.6315650343894958,
+      "learning_rate": 5e-05,
+      "loss": 1.091,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9548872180451128,
-      "eval_loss": 0.15986023843288422,
-      "eval_runtime": 0.7821,
-      "eval_samples_per_second": 170.05,
-      "eval_steps_per_second": 21.736,
+      "eval_accuracy": 0.518796992481203,
+      "eval_loss": 1.002743124961853,
+      "eval_runtime": 0.9515,
+      "eval_samples_per_second": 139.773,
+      "eval_steps_per_second": 17.866,
       "step": 260
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.9846404790878296,
-      "learning_rate": 0.005,
-      "loss": 0.2953,
+      "grad_norm": 1.4020596742630005,
+      "learning_rate": 5e-05,
+      "loss": 1.0433,
       "step": 300
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9548872180451128,
-      "eval_loss": 0.1191997155547142,
-      "eval_runtime": 0.8019,
-      "eval_samples_per_second": 165.864,
-      "eval_steps_per_second": 21.201,
+      "eval_accuracy": 0.518796992481203,
+      "eval_loss": 0.9840062856674194,
+      "eval_runtime": 0.9456,
+      "eval_samples_per_second": 140.656,
+      "eval_steps_per_second": 17.979,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 1.4397724866867065,
-      "learning_rate": 0.005,
-      "loss": 0.2891,
+      "grad_norm": 1.387670636177063,
+      "learning_rate": 5e-05,
+      "loss": 0.9915,
       "step": 400
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 3.0908043384552,
-      "learning_rate": 0.005,
-      "loss": 0.2653,
+      "grad_norm": 0.7062584161758423,
+      "learning_rate": 5e-05,
+      "loss": 0.9823,
       "step": 500
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9398496240601504,
-      "eval_loss": 0.13930319249629974,
-      "eval_runtime": 0.7871,
-      "eval_samples_per_second": 168.974,
-      "eval_steps_per_second": 21.598,
+      "eval_accuracy": 0.5639097744360902,
+      "eval_loss": 0.9470232129096985,
+      "eval_runtime": 0.9624,
+      "eval_samples_per_second": 138.194,
+      "eval_steps_per_second": 17.664,
       "step": 520
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 2.720701217651367,
-      "learning_rate": 0.005,
-      "loss": 0.2344,
+      "grad_norm": 0.5388019680976868,
+      "learning_rate": 5e-05,
+      "loss": 0.9523,
       "step": 600
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.10011590272188187,
-      "eval_runtime": 0.8095,
-      "eval_samples_per_second": 164.306,
-      "eval_steps_per_second": 21.001,
+      "eval_accuracy": 0.5789473684210527,
+      "eval_loss": 0.924637496471405,
+      "eval_runtime": 0.9596,
+      "eval_samples_per_second": 138.602,
+      "eval_steps_per_second": 17.716,
       "step": 650
     },
     {
       "epoch": 5.384615384615385,
-      "grad_norm": 0.17460593581199646,
-      "learning_rate": 0.005,
-      "loss": 0.21,
+      "grad_norm": 1.5363554954528809,
+      "learning_rate": 5e-05,
+      "loss": 0.9431,
       "step": 700
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.0893179401755333,
-      "eval_runtime": 0.7824,
-      "eval_samples_per_second": 169.993,
-      "eval_steps_per_second": 21.728,
+      "eval_accuracy": 0.556390977443609,
+      "eval_loss": 0.890453577041626,
+      "eval_runtime": 0.9858,
+      "eval_samples_per_second": 134.909,
+      "eval_steps_per_second": 17.244,
       "step": 780
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 0.6050938963890076,
-      "learning_rate": 0.005,
-      "loss": 0.2397,
+      "grad_norm": 1.7380107641220093,
+      "learning_rate": 5e-05,
+      "loss": 0.8877,
       "step": 800
     },
     {
       "epoch": 6.923076923076923,
-      "grad_norm": 1.7308460474014282,
-      "learning_rate": 0.005,
-      "loss": 0.3117,
+      "grad_norm": 2.0791220664978027,
+      "learning_rate": 5e-05,
+      "loss": 0.8587,
       "step": 900
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.924812030075188,
-      "eval_loss": 0.1933087259531021,
-      "eval_runtime": 0.7988,
-      "eval_samples_per_second": 166.494,
-      "eval_steps_per_second": 21.281,
+      "eval_accuracy": 0.5939849624060151,
+      "eval_loss": 0.8433966636657715,
+      "eval_runtime": 0.9653,
+      "eval_samples_per_second": 137.788,
+      "eval_steps_per_second": 17.612,
       "step": 910
     },
     {
       "epoch": 7.6923076923076925,
-      "grad_norm": 1.327934741973877,
-      "learning_rate": 0.005,
-      "loss": 0.2459,
+      "grad_norm": 1.4432713985443115,
+      "learning_rate": 5e-05,
+      "loss": 0.834,
       "step": 1000
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.924812030075188,
-      "eval_loss": 0.19011461734771729,
-      "eval_runtime": 0.7881,
-      "eval_samples_per_second": 168.75,
-      "eval_steps_per_second": 21.57,
+      "eval_accuracy": 0.5714285714285714,
+      "eval_loss": 0.8318859934806824,
+      "eval_runtime": 0.9617,
+      "eval_samples_per_second": 138.302,
+      "eval_steps_per_second": 17.678,
       "step": 1040
     },
     {
       "epoch": 8.461538461538462,
-      "grad_norm": 0.8504745364189148,
-      "learning_rate": 0.005,
-      "loss": 0.25,
+      "grad_norm": 2.059706449508667,
+      "learning_rate": 5e-05,
+      "loss": 0.8149,
       "step": 1100
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.08675242960453033,
-      "eval_runtime": 0.7998,
-      "eval_samples_per_second": 166.287,
-      "eval_steps_per_second": 21.255,
+      "eval_accuracy": 0.6766917293233082,
+      "eval_loss": 0.7831493616104126,
+      "eval_runtime": 0.9673,
+      "eval_samples_per_second": 137.496,
+      "eval_steps_per_second": 17.575,
       "step": 1170
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 3.838754177093506,
-      "learning_rate": 0.005,
-      "loss": 0.1834,
+      "grad_norm": 1.22969651222229,
+      "learning_rate": 5e-05,
+      "loss": 0.7628,
       "step": 1200
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.03880231827497482,
-      "learning_rate": 0.005,
-      "loss": 0.2038,
+      "grad_norm": 3.152574300765991,
+      "learning_rate": 5e-05,
+      "loss": 0.729,
       "step": 1300
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.9473684210526315,
-      "eval_loss": 0.15278470516204834,
-      "eval_runtime": 0.8065,
-      "eval_samples_per_second": 164.915,
-      "eval_steps_per_second": 21.079,
+      "eval_accuracy": 0.6541353383458647,
+      "eval_loss": 0.751246988773346,
+      "eval_runtime": 0.9753,
+      "eval_samples_per_second": 136.369,
+      "eval_steps_per_second": 17.431,
       "step": 1300
     },
     {
       "epoch": 10.0,
       "step": 1300,
-      "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
+      "total_flos": 8.772706474360013e+17,
+      "train_loss": 0.9225140674297626,
+      "train_runtime": 175.6098,
+      "train_samples_per_second": 58.881,
+      "train_steps_per_second": 7.403
     }
   ],
   "logging_steps": 100,
@@ -216,7 +216,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.508699063477043e+17,
+  "total_flos": 8.772706474360013e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null