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: 0.9767
-- eval_accuracy: 0.7947
-- eval_runtime: 155.0819
-- eval_samples_per_second: 97.69
-- eval_steps_per_second: 6.106
-- epoch: 12.67
-- step: 12000
+- eval_loss: 0.8917
+- eval_accuracy: 0.7984
+- eval_runtime: 154.0819
+- eval_samples_per_second: 98.324
+- eval_steps_per_second: 6.146
+- epoch: 13.73
+- step: 13000
 
 ## Model description
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0c30cd551dee8b88adf1abb5ea9088a9c56b772faeac0a11e9a6f54d0fbf35b
+oid sha256:38c00bea494555b77c31d5ff39fbce3082b261588112621b13bdc1ffc0680430
 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:f9c454d2313ba3956712f0130a2223945517cc8f21e0eb033697f07c35c2f430
-size 15061
+oid sha256:20bd81f6196a2bdeca02fece8451c1ecd2302bd7d710989015d82660d29ffa15
+size 15541

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.7831683168316832,
-  "best_model_checkpoint": "food-image-classification/checkpoint-11000",
-  "epoch": 12.671594508975712,
+  "best_metric": 0.7946534653465347,
+  "best_model_checkpoint": "food-image-classification/checkpoint-12000",
+  "epoch": 13.727560718057022,
   "eval_steps": 1000,
-  "global_step": 12000,
+  "global_step": 13000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -187,6 +187,21 @@
       "eval_samples_per_second": 97.69,
       "eval_steps_per_second": 6.106,
       "step": 12000
+    },
+    {
+      "epoch": 13.73,
+      "learning_rate": 1.3727560718057022e-05,
+      "loss": 0.7817,
+      "step": 13000
+    },
+    {
+      "epoch": 13.73,
+      "eval_accuracy": 0.7984158415841585,
+      "eval_loss": 0.8917332887649536,
+      "eval_runtime": 154.0819,
+      "eval_samples_per_second": 98.324,
+      "eval_steps_per_second": 6.146,
+      "step": 13000
     }
   ],
   "logging_steps": 1000,
@@ -194,7 +209,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 1000,
-  "total_flos": 5.955920916788183e+19,
+  "total_flos": 6.452247659853865e+19,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null