End of training

README.md

@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 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 imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6870
+- Loss: 0.6871
 - Accuracy: 0.5306
 
 ## Model description

all_results.json

@@ -1,8 +1,13 @@
 {
     "epoch": 10.0,
-    "total_flos": 6.858183153733632e+17,
-    "train_loss": 0.6843075002942767,
-    "train_runtime": 315.6994,
-    "train_samples_per_second": 27.653,
-    "train_steps_per_second": 0.222
+    "eval_accuracy": 0.5306122448979592,
+    "eval_loss": 0.6871312856674194,
+    "eval_runtime": 2.0705,
+    "eval_samples_per_second": 47.332,
+    "eval_steps_per_second": 1.932,
+    "total_flos": 6.788424757963162e+17,
+    "train_loss": 0.6839384623936244,
+    "train_runtime": 322.9839,
+    "train_samples_per_second": 27.029,
+    "train_steps_per_second": 0.217
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.5306122448979592,
+    "eval_loss": 0.6871312856674194,
+    "eval_runtime": 2.0705,
+    "eval_samples_per_second": 47.332,
+    "eval_steps_per_second": 1.932
+}

runs/Aug15_14-14-13_c03d0dbfda2e/events.out.tfevents.1723731647.c03d0dbfda2e.1885.10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5b54770aacbed5c19a8d5acfd8be42d6ac13e8f501a16b7426f5e060acd5475
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "total_flos": 6.858183153733632e+17,
-    "train_loss": 0.6843075002942767,
-    "train_runtime": 315.6994,
-    "train_samples_per_second": 27.653,
-    "train_steps_per_second": 0.222
+    "total_flos": 6.788424757963162e+17,
+    "train_loss": 0.6839384623936244,
+    "train_runtime": 322.9839,
+    "train_samples_per_second": 27.029,
+    "train_steps_per_second": 0.217
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.5204081632653061,
-  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-qlora-houseplant/checkpoint-21",
+  "best_metric": 0.5306122448979592,
+  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-qlora-houseplant/checkpoint-63",
   "epoch": 10.0,
   "eval_steps": 500,
   "global_step": 70,
@@ -10,151 +10,151 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5102040816326531,
-      "eval_loss": 0.6916877627372742,
-      "eval_runtime": 1.9776,
-      "eval_samples_per_second": 49.555,
-      "eval_steps_per_second": 2.023,
+      "eval_accuracy": 0.5,
+      "eval_loss": 0.6916753053665161,
+      "eval_runtime": 2.0071,
+      "eval_samples_per_second": 48.827,
+      "eval_steps_per_second": 1.993,
       "step": 7
     },
     {
       "epoch": 1.4285714285714286,
-      "grad_norm": 0.5460406541824341,
+      "grad_norm": 0.5469765067100525,
       "learning_rate": 9.523809523809525e-06,
       "loss": 0.688,
       "step": 10
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5102040816326531,
-      "eval_loss": 0.6909428834915161,
-      "eval_runtime": 2.0178,
-      "eval_samples_per_second": 48.568,
-      "eval_steps_per_second": 1.982,
+      "eval_accuracy": 0.5204081632653061,
+      "eval_loss": 0.6907908916473389,
+      "eval_runtime": 1.9735,
+      "eval_samples_per_second": 49.657,
+      "eval_steps_per_second": 2.027,
       "step": 14
     },
     {
       "epoch": 2.857142857142857,
-      "grad_norm": 0.44764864444732666,
+      "grad_norm": 0.44840723276138306,
       "learning_rate": 7.936507936507936e-06,
-      "loss": 0.6845,
+      "loss": 0.6843,
       "step": 20
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.690205454826355,
-      "eval_runtime": 2.828,
-      "eval_samples_per_second": 34.654,
-      "eval_steps_per_second": 1.414,
+      "eval_loss": 0.6899289488792419,
+      "eval_runtime": 2.7667,
+      "eval_samples_per_second": 35.421,
+      "eval_steps_per_second": 1.446,
       "step": 21
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6895502805709839,
-      "eval_runtime": 1.9885,
-      "eval_samples_per_second": 49.284,
-      "eval_steps_per_second": 2.012,
+      "eval_loss": 0.689208984375,
+      "eval_runtime": 1.9903,
+      "eval_samples_per_second": 49.24,
+      "eval_steps_per_second": 2.01,
       "step": 28
     },
     {
       "epoch": 4.285714285714286,
-      "grad_norm": 0.48582252860069275,
+      "grad_norm": 0.486519992351532,
       "learning_rate": 6.349206349206349e-06,
-      "loss": 0.6874,
+      "loss": 0.6871,
       "step": 30
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6889798045158386,
-      "eval_runtime": 1.9771,
-      "eval_samples_per_second": 49.568,
-      "eval_steps_per_second": 2.023,
+      "eval_loss": 0.688566267490387,
+      "eval_runtime": 2.1429,
+      "eval_samples_per_second": 45.733,
+      "eval_steps_per_second": 1.867,
       "step": 35
     },
     {
       "epoch": 5.714285714285714,
-      "grad_norm": 0.4863179922103882,
+      "grad_norm": 0.487440288066864,
       "learning_rate": 4.761904761904762e-06,
-      "loss": 0.6838,
+      "loss": 0.6833,
       "step": 40
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6885014772415161,
-      "eval_runtime": 2.3711,
-      "eval_samples_per_second": 41.331,
-      "eval_steps_per_second": 1.687,
+      "eval_loss": 0.688008189201355,
+      "eval_runtime": 2.0813,
+      "eval_samples_per_second": 47.085,
+      "eval_steps_per_second": 1.922,
       "step": 42
     },
     {
       "epoch": 7.0,
       "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6881576776504517,
-      "eval_runtime": 1.9786,
-      "eval_samples_per_second": 49.529,
-      "eval_steps_per_second": 2.022,
+      "eval_loss": 0.6875921487808228,
+      "eval_runtime": 1.9963,
+      "eval_samples_per_second": 49.091,
+      "eval_steps_per_second": 2.004,
       "step": 49
     },
     {
       "epoch": 7.142857142857143,
-      "grad_norm": 0.5103937983512878,
+      "grad_norm": 0.5109758973121643,
       "learning_rate": 3.1746031746031746e-06,
-      "loss": 0.6817,
+      "loss": 0.6812,
       "step": 50
     },
     {
       "epoch": 8.0,
       "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6879384517669678,
-      "eval_runtime": 1.9723,
-      "eval_samples_per_second": 49.687,
-      "eval_steps_per_second": 2.028,
+      "eval_loss": 0.687288224697113,
+      "eval_runtime": 2.4156,
+      "eval_samples_per_second": 40.57,
+      "eval_steps_per_second": 1.656,
       "step": 56
     },
     {
       "epoch": 8.571428571428571,
-      "grad_norm": 0.45278120040893555,
+      "grad_norm": 0.45402806997299194,
       "learning_rate": 1.5873015873015873e-06,
-      "loss": 0.6813,
+      "loss": 0.6807,
       "step": 60
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6877266764640808,
-      "eval_runtime": 1.9657,
-      "eval_samples_per_second": 49.856,
-      "eval_steps_per_second": 2.035,
+      "eval_accuracy": 0.5306122448979592,
+      "eval_loss": 0.6871312856674194,
+      "eval_runtime": 1.9727,
+      "eval_samples_per_second": 49.679,
+      "eval_steps_per_second": 2.028,
       "step": 63
     },
     {
       "epoch": 10.0,
-      "grad_norm": 0.6086100935935974,
+      "grad_norm": 0.6085575222969055,
       "learning_rate": 0.0,
-      "loss": 0.6835,
+      "loss": 0.6829,
       "step": 70
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.5204081632653061,
-      "eval_loss": 0.6876619458198547,
-      "eval_runtime": 2.0199,
-      "eval_samples_per_second": 48.516,
-      "eval_steps_per_second": 1.98,
+      "eval_accuracy": 0.5306122448979592,
+      "eval_loss": 0.6870341300964355,
+      "eval_runtime": 1.9967,
+      "eval_samples_per_second": 49.082,
+      "eval_steps_per_second": 2.003,
       "step": 70
     },
     {
       "epoch": 10.0,
       "step": 70,
-      "total_flos": 6.858183153733632e+17,
-      "train_loss": 0.6843075002942767,
-      "train_runtime": 315.6994,
-      "train_samples_per_second": 27.653,
-      "train_steps_per_second": 0.222
+      "total_flos": 6.788424757963162e+17,
+      "train_loss": 0.6839384623936244,
+      "train_runtime": 322.9839,
+      "train_samples_per_second": 27.029,
+      "train_steps_per_second": 0.217
     }
   ],
   "logging_steps": 10,
@@ -174,7 +174,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.858183153733632e+17,
+  "total_flos": 6.788424757963162e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null