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: 0.8235
-- eval_accuracy: 0.8044
-- eval_runtime: 153.6348
-- eval_samples_per_second: 98.61
-- eval_steps_per_second: 6.164
-- epoch: 14.78
-- step: 14000
+- eval_loss: 0.7959
+- eval_accuracy: 0.8075
+- eval_runtime: 153.9402
+- eval_samples_per_second: 98.415
+- eval_steps_per_second: 6.152
+- epoch: 15.84
+- step: 15000
 
 ## Model description
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44fb177d54f7d920ebec7ed9d305e79d34fbbd63d2870afcc2ea43abebc25eb9
+oid sha256:c1b09c44bb650673e03d905cbf2cfecd3c70f20f9fc55fff7e74ce2f82877fdb
 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 sha256:45434541e41ee33f030b4689849ab4629e7ae88b07aa47ea7fe666830806a396
-size 16021
+oid sha256:c35e15cb5d3094dc101119d8841eae79ffdf984df25ffddf26551440b87fe5c8
+size 16501

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.7984158415841585,
-  "best_model_checkpoint": "food-image-classification/checkpoint-13000",
-  "epoch": 14.783526927138332,
+  "best_metric": 0.8043564356435644,
+  "best_model_checkpoint": "food-image-classification/checkpoint-14000",
+  "epoch": 15.839493136219641,
   "eval_steps": 1000,
-  "global_step": 14000,
+  "global_step": 15000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -217,6 +217,21 @@
       "eval_samples_per_second": 98.61,
       "eval_steps_per_second": 6.164,
       "step": 14000
+    },
+    {
+      "epoch": 15.84,
+      "learning_rate": 1.583949313621964e-05,
+      "loss": 0.6281,
+      "step": 15000
+    },
+    {
+      "epoch": 15.84,
+      "eval_accuracy": 0.8075247524752476,
+      "eval_loss": 0.7959182858467102,
+      "eval_runtime": 153.9402,
+      "eval_samples_per_second": 98.415,
+      "eval_steps_per_second": 6.152,
+      "step": 15000
     }
   ],
   "logging_steps": 1000,
@@ -224,7 +239,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 1000,
-  "total_flos": 6.948574402919547e+19,
+  "total_flos": 7.444901145985229e+19,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null