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: 2.3822
-- eval_accuracy: 0.7071
-- eval_runtime: 155.5967
-- eval_samples_per_second: 97.367
-- eval_steps_per_second: 6.086
-- epoch: 7.39
-- step: 7000
+- eval_loss: 1.9754
+- eval_accuracy: 0.7336
+- eval_runtime: 156.8793
+- eval_samples_per_second: 96.571
+- eval_steps_per_second: 6.036
+- epoch: 8.45
+- step: 8000
 
 ## Model description
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:42e59812c0f6edec89f8849798093e690a2f7634ed7bca9a5ca5578ab84fcab2
+oid sha256:b20edeb0fe7d3b161c0a46a58cae188fa221d032d8c7766255b9a02137a73c09
 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 sha256:5cc803c469db4aff184af8460d4a1ebbe418b0f1fdef1522e6e8903678ef5fbb
-size 12661
+oid sha256:a033a1a586ec82c34e1ea9c30d4c0e829e843a9fd4da33e9e4e02374982ce418
+size 13141

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.687062706270627,
-  "best_model_checkpoint": "food-image-classification/checkpoint-6000",
-  "epoch": 7.391763463569166,
+  "best_metric": 0.7070627062706271,
+  "best_model_checkpoint": "food-image-classification/checkpoint-7000",
+  "epoch": 8.447729672650475,
   "eval_steps": 1000,
-  "global_step": 7000,
+  "global_step": 8000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -112,6 +112,21 @@
       "eval_samples_per_second": 97.367,
       "eval_steps_per_second": 6.086,
       "step": 7000
+    },
+    {
+      "epoch": 8.45,
+      "learning_rate": 8.447729672650476e-06,
+      "loss": 2.1397,
+      "step": 8000
+    },
+    {
+      "epoch": 8.45,
+      "eval_accuracy": 0.7335973597359736,
+      "eval_loss": 1.975380539894104,
+      "eval_runtime": 156.8793,
+      "eval_samples_per_second": 96.571,
+      "eval_steps_per_second": 6.036,
+      "step": 8000
     }
   ],
   "logging_steps": 1000,
@@ -119,7 +134,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 500,
   "save_steps": 1000,
-  "total_flos": 3.4742872014597734e+19,
+  "total_flos": 3.970613944525455e+19,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null