End of training

README.md

@@ -17,13 +17,13 @@ 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 food101 dataset.
 It achieves the following results on the evaluation set:
-- eval_loss: 3.6342
-- eval_accuracy: 0.6112
-- eval_runtime: 156.2293
-- eval_samples_per_second: 96.973
-- eval_steps_per_second: 6.062
-- epoch: 4.22
-- step: 4000
+- eval_loss: 3.2252
+- eval_accuracy: 0.6515
+- eval_runtime: 154.6911
+- eval_samples_per_second: 97.937
+- eval_steps_per_second: 6.122
+- epoch: 5.28
+- step: 5000
 
 ## Model description
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a574228757dbeb58dda7eb7c5137d9c6fe0952f3ba4faf5b4a1392ad3aa32e3c
+oid sha256:481b50b254fc5d798ca7841a37b28fd52550a3d5a27d8a4fe672d076c3a48678
 size 343528508

runs/Feb05_03-08-58_983d148b451b/events.out.tfevents.1707102539.983d148b451b.26.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bad0bdc4c25c75c7a63f8ccf285a91b4e697e6615fec6b0cbdfeb86d6c9e4fb
-size 11221
+oid sha256:b5123372ad017b6345bb526b80b72debcaf26727d68ee0bad53e6b242b1f331d
+size 11701

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.5498349834983498,
-  "best_model_checkpoint": "food-image-classification/checkpoint-3000",
-  "epoch": 4.223864836325237,
+  "best_metric": 0.6111551155115511,
+  "best_model_checkpoint": "food-image-classification/checkpoint-4000",
+  "epoch": 5.279831045406547,
   "eval_steps": 1000,
-  "global_step": 4000,
+  "global_step": 5000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -67,6 +67,21 @@
       "eval_samples_per_second": 96.973,
       "eval_steps_per_second": 6.062,
       "step": 4000
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 5.279831045406547e-06,
+      "loss": 3.4182,
+      "step": 5000
+    },
+    {
+      "epoch": 5.28,
+      "eval_accuracy": 0.6514851485148515,
+      "eval_loss": 3.225186586380005,
+      "eval_runtime": 154.6911,
+      "eval_samples_per_second": 97.937,
+      "eval_steps_per_second": 6.122,
+      "step": 5000
     }
   ],
   "logging_steps": 1000,
@@ -74,7 +89,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 1000,
-  "total_flos": 1.9853069722627277e+19,
+  "total_flos": 2.4816337153284096e+19,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null