End of training

README.md

@@ -21,7 +21,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8977272727272727
+      value: 0.9015151515151515
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Tidzo/dit-base-rvlcdip-finetuned-grp-actual](https://huggingface.co/Tidzo/dit-base-rvlcdip-finetuned-grp-actual) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4582
-- Accuracy: 0.8977
+- Loss: 0.4601
+- Accuracy: 0.9015
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 6.72,
-    "eval_accuracy": 0.7575757575757576,
-    "eval_loss": 1.3033273220062256,
-    "eval_runtime": 14.5204,
-    "eval_samples_per_second": 18.181,
-    "eval_steps_per_second": 0.62,
+    "eval_accuracy": 0.9015151515151515,
+    "eval_loss": 0.46005359292030334,
+    "eval_runtime": 17.5331,
+    "eval_samples_per_second": 15.057,
+    "eval_steps_per_second": 0.513,
     "total_flos": 1.2388649195611423e+18,
-    "train_loss": 1.7003454405163962,
-    "train_runtime": 3667.1764,
-    "train_samples_per_second": 4.535,
-    "train_steps_per_second": 0.034
+    "train_loss": 0.7077309669010223,
+    "train_runtime": 1389.7418,
+    "train_samples_per_second": 11.968,
+    "train_steps_per_second": 0.091
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 6.72,
-    "eval_accuracy": 0.7575757575757576,
-    "eval_loss": 1.3033273220062256,
-    "eval_runtime": 14.5204,
-    "eval_samples_per_second": 18.181,
-    "eval_steps_per_second": 0.62
+    "eval_accuracy": 0.9015151515151515,
+    "eval_loss": 0.46005359292030334,
+    "eval_runtime": 17.5331,
+    "eval_samples_per_second": 15.057,
+    "eval_steps_per_second": 0.513
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 6.72,
     "total_flos": 1.2388649195611423e+18,
-    "train_loss": 1.7003454405163962,
-    "train_runtime": 3667.1764,
-    "train_samples_per_second": 4.535,
-    "train_steps_per_second": 0.034
+    "train_loss": 0.7077309669010223,
+    "train_runtime": 1389.7418,
+    "train_samples_per_second": 11.968,
+    "train_steps_per_second": 0.091
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.7575757575757576,
-  "best_model_checkpoint": "dit-base-rvlcdip-finetuned-grp-actual/checkpoint-93",
+  "best_metric": 0.9015151515151515,
+  "best_model_checkpoint": "dit-base-rvlcdip-finetuned-grp-actual/checkpoint-112",
   "epoch": 6.72,
   "eval_steps": 500,
   "global_step": 126,
@@ -11,146 +11,146 @@
     {
       "epoch": 0.53,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 2.3577,
+      "loss": 0.8692,
       "step": 10
     },
     {
       "epoch": 0.96,
-      "eval_accuracy": 0.5113636363636364,
-      "eval_loss": 2.086306571960449,
-      "eval_runtime": 240.3077,
-      "eval_samples_per_second": 1.099,
-      "eval_steps_per_second": 0.037,
+      "eval_accuracy": 0.8560606060606061,
+      "eval_loss": 0.6972441673278809,
+      "eval_runtime": 15.3616,
+      "eval_samples_per_second": 17.186,
+      "eval_steps_per_second": 0.586,
       "step": 18
     },
     {
       "epoch": 1.07,
       "learning_rate": 4.690265486725664e-05,
-      "loss": 2.2163,
+      "loss": 0.7995,
       "step": 20
     },
     {
       "epoch": 1.6,
       "learning_rate": 4.247787610619469e-05,
-      "loss": 2.0601,
+      "loss": 0.7348,
       "step": 30
     },
     {
       "epoch": 1.97,
-      "eval_accuracy": 0.6477272727272727,
-      "eval_loss": 1.8153679370880127,
-      "eval_runtime": 14.6331,
-      "eval_samples_per_second": 18.041,
-      "eval_steps_per_second": 0.615,
+      "eval_accuracy": 0.8598484848484849,
+      "eval_loss": 0.6350475549697876,
+      "eval_runtime": 16.1781,
+      "eval_samples_per_second": 16.318,
+      "eval_steps_per_second": 0.556,
       "step": 37
     },
     {
       "epoch": 2.13,
       "learning_rate": 3.8053097345132744e-05,
-      "loss": 1.943,
+      "loss": 0.6855,
       "step": 40
     },
     {
       "epoch": 2.67,
       "learning_rate": 3.3628318584070804e-05,
-      "loss": 1.8068,
+      "loss": 0.6655,
       "step": 50
     },
     {
       "epoch": 2.99,
-      "eval_accuracy": 0.6704545454545454,
-      "eval_loss": 1.5881296396255493,
-      "eval_runtime": 14.8318,
-      "eval_samples_per_second": 17.8,
-      "eval_steps_per_second": 0.607,
+      "eval_accuracy": 0.8712121212121212,
+      "eval_loss": 0.533867359161377,
+      "eval_runtime": 15.3125,
+      "eval_samples_per_second": 17.241,
+      "eval_steps_per_second": 0.588,
       "step": 56
     },
     {
       "epoch": 3.2,
       "learning_rate": 2.9203539823008852e-05,
-      "loss": 1.66,
+      "loss": 0.7083,
       "step": 60
     },
     {
       "epoch": 3.73,
       "learning_rate": 2.4778761061946905e-05,
-      "loss": 1.5953,
+      "loss": 0.7167,
       "step": 70
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.7159090909090909,
-      "eval_loss": 1.4111517667770386,
-      "eval_runtime": 14.6707,
-      "eval_samples_per_second": 17.995,
-      "eval_steps_per_second": 0.613,
+      "eval_accuracy": 0.8901515151515151,
+      "eval_loss": 0.5045967102050781,
+      "eval_runtime": 15.2445,
+      "eval_samples_per_second": 17.318,
+      "eval_steps_per_second": 0.59,
       "step": 75
     },
     {
       "epoch": 4.27,
       "learning_rate": 2.0353982300884957e-05,
-      "loss": 1.4929,
+      "loss": 0.6926,
       "step": 80
     },
     {
       "epoch": 4.8,
       "learning_rate": 1.592920353982301e-05,
-      "loss": 1.4304,
+      "loss": 0.694,
       "step": 90
     },
     {
       "epoch": 4.96,
-      "eval_accuracy": 0.7575757575757576,
-      "eval_loss": 1.3033273220062256,
-      "eval_runtime": 14.9724,
-      "eval_samples_per_second": 17.632,
-      "eval_steps_per_second": 0.601,
+      "eval_accuracy": 0.8863636363636364,
+      "eval_loss": 0.5026077628135681,
+      "eval_runtime": 15.2119,
+      "eval_samples_per_second": 17.355,
+      "eval_steps_per_second": 0.592,
       "step": 93
     },
     {
       "epoch": 5.33,
       "learning_rate": 1.1504424778761062e-05,
-      "loss": 1.3606,
+      "loss": 0.6366,
       "step": 100
     },
     {
       "epoch": 5.87,
       "learning_rate": 7.079646017699115e-06,
-      "loss": 1.3458,
+      "loss": 0.6638,
       "step": 110
     },
     {
       "epoch": 5.97,
-      "eval_accuracy": 0.75,
-      "eval_loss": 1.2400753498077393,
-      "eval_runtime": 14.8483,
-      "eval_samples_per_second": 17.78,
-      "eval_steps_per_second": 0.606,
+      "eval_accuracy": 0.9015151515151515,
+      "eval_loss": 0.46005359292030334,
+      "eval_runtime": 15.7633,
+      "eval_samples_per_second": 16.748,
+      "eval_steps_per_second": 0.571,
       "step": 112
     },
     {
       "epoch": 6.4,
       "learning_rate": 2.6548672566371683e-06,
-      "loss": 1.3523,
+      "loss": 0.6618,
       "step": 120
     },
     {
       "epoch": 6.72,
-      "eval_accuracy": 0.7575757575757576,
-      "eval_loss": 1.2240339517593384,
-      "eval_runtime": 14.9425,
-      "eval_samples_per_second": 17.668,
-      "eval_steps_per_second": 0.602,
+      "eval_accuracy": 0.8977272727272727,
+      "eval_loss": 0.45817553997039795,
+      "eval_runtime": 17.722,
+      "eval_samples_per_second": 14.897,
+      "eval_steps_per_second": 0.508,
       "step": 126
     },
     {
       "epoch": 6.72,
       "step": 126,
       "total_flos": 1.2388649195611423e+18,
-      "train_loss": 1.7003454405163962,
-      "train_runtime": 3667.1764,
-      "train_samples_per_second": 4.535,
-      "train_steps_per_second": 0.034
+      "train_loss": 0.7077309669010223,
+      "train_runtime": 1389.7418,
+      "train_samples_per_second": 11.968,
+      "train_steps_per_second": 0.091
     }
   ],
   "logging_steps": 10,