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: 0.7243
-- eval_accuracy: 0.8202
-- eval_runtime: 153.8982
-- eval_samples_per_second: 98.442
-- eval_steps_per_second: 6.153
-- epoch: 23.23
-- step: 22000
+- eval_loss: 0.7594
+- eval_accuracy: 0.8104
+- eval_runtime: 155.7313
+- eval_samples_per_second: 97.283
+- eval_steps_per_second: 6.081
+- epoch: 24.29
+- step: 23000
 
 ## Model description
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1dcec261e11534ea2e4e2de550fb5c3ba9a68d71402198f972c441f52648a96
+oid sha256:7dc6f5a578f532ad87106d4cec6c0cdf748aa5b9c3cd35e88f7f8ad52cbb735f
 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 sha256:57a78def9c1f785eb4d7f6bbaa73ab47ee215105a841c610607e72b4365bb855
-size 19915
+oid sha256:0ce6e73591fdb51be189a8d88f82f51132563b067d67c396fb99cde827572455
+size 20404

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.8158415841584158,
-  "best_model_checkpoint": "food-image-classification/checkpoint-16000",
-  "epoch": 23.231256599788807,
+  "best_metric": 0.8201980198019801,
+  "best_model_checkpoint": "food-image-classification/checkpoint-22000",
+  "epoch": 24.287222808870116,
   "eval_steps": 1000,
-  "global_step": 22000,
+  "global_step": 23000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -337,6 +337,21 @@
       "eval_samples_per_second": 98.442,
       "eval_steps_per_second": 6.153,
       "step": 22000
+    },
+    {
+      "epoch": 24.29,
+      "learning_rate": 2.4287222808870115e-05,
+      "loss": 0.3785,
+      "step": 23000
+    },
+    {
+      "epoch": 24.29,
+      "eval_accuracy": 0.8103630363036304,
+      "eval_loss": 0.7593609690666199,
+      "eval_runtime": 155.7313,
+      "eval_samples_per_second": 97.283,
+      "eval_steps_per_second": 6.081,
+      "step": 23000
     }
   ],
   "logging_steps": 1000,
@@ -344,7 +359,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 1000,
-  "total_flos": 1.0919126298846044e+20,
+  "total_flos": 1.1415453041911726e+20,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null