Bongyun

README.md

@@ -1,10 +1,13 @@
 ---
+library_name: transformers
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
 - image-classification
 - ViT
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: vit-base-beans-demo-v5
   results: []
@@ -16,6 +19,9 @@ should probably proofread and complete it, then remove this comment. -->
 # vit-base-beans-demo-v5
 
 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 beans dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0227
+- Accuracy: 0.9925
 
 ## Model description
 
@@ -38,14 +44,22 @@ The following hyperparameters were used during training:
 - train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - num_epochs: 4
 - mixed_precision_training: Native AMP
 
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
+|:-------------:|:------:|:----:|:---------------:|:--------:|
+| 0.0783        | 1.5385 | 100  | 0.0551          | 0.9850   |
+| 0.0124        | 3.0769 | 200  | 0.0227          | 0.9925   |
+
+
 ### Framework versions
 
-- Transformers 4.42.4
-- Pytorch 2.3.1+cu121
-- Datasets 2.20.0
-- Tokenizers 0.19.1
+- Transformers 4.46.2
+- Pytorch 2.5.0+cu121
+- Datasets 3.1.0
+- Tokenizers 0.20.3

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.11968801841139794,
-    "train_runtime": 105.0874,
-    "train_samples_per_second": 39.358,
-    "train_steps_per_second": 2.474
+    "train_loss": 0.12550590594227498,
+    "train_runtime": 104.2364,
+    "train_samples_per_second": 39.679,
+    "train_steps_per_second": 2.494
 }

config.json

@@ -30,5 +30,5 @@
   "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.42.4"
+  "transformers_version": "4.46.2"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24b062e7e864e24a09178386c6ae7ddeef3e8fe6e8db2dcd796aaf28738a9586
+oid sha256:38b87efad3dccfb9ec1d1621491fd143e65044773c30145f9f6393e94fbe82fd
 size 343227052

preprocessor_config.json

@@ -1,7 +1,7 @@
 {
   "do_normalize": true,
   "do_rescale": true,
-  "do_resize": "google/vit-base-patch16-224-in21k",
+  "do_resize": true,
   "image_mean": [
     0.5,
     0.5,

runs/Nov13_04-02-46_1f52fa223e2d/events.out.tfevents.1731470572.1f52fa223e2d.833.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34c20ca800a31afbfc4e22c87279e968f74d9de614aba2b2e7aec68c9ef55b7c
+size 11559

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.11968801841139794,
-    "train_runtime": 105.0874,
-    "train_samples_per_second": 39.358,
-    "train_steps_per_second": 2.474
+    "train_loss": 0.12550590594227498,
+    "train_runtime": 104.2364,
+    "train_samples_per_second": 39.679,
+    "train_steps_per_second": 2.494
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.05026606097817421,
+  "best_metric": 0.02266957052052021,
   "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
   "epoch": 4.0,
   "eval_steps": 100,
@@ -10,212 +10,212 @@
   "log_history": [
     {
       "epoch": 0.15384615384615385,
-      "grad_norm": 1.2843873500823975,
+      "grad_norm": 1.8716892004013062,
       "learning_rate": 0.00019230769230769233,
-      "loss": 0.8434,
+      "loss": 0.8364,
       "step": 10
     },
     {
       "epoch": 0.3076923076923077,
-      "grad_norm": 3.7151455879211426,
+      "grad_norm": 2.2301363945007324,
       "learning_rate": 0.00018461538461538463,
-      "loss": 0.3085,
+      "loss": 0.3741,
       "step": 20
     },
     {
       "epoch": 0.46153846153846156,
-      "grad_norm": 0.8530910611152649,
+      "grad_norm": 2.373594284057617,
       "learning_rate": 0.00017692307692307693,
-      "loss": 0.248,
+      "loss": 0.2396,
       "step": 30
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 1.1804994344711304,
+      "grad_norm": 0.6037946343421936,
       "learning_rate": 0.00016923076923076923,
-      "loss": 0.2334,
+      "loss": 0.2642,
       "step": 40
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.5084353089332581,
+      "grad_norm": 0.3296580910682678,
       "learning_rate": 0.00016153846153846155,
-      "loss": 0.1626,
+      "loss": 0.2242,
       "step": 50
     },
     {
       "epoch": 0.9230769230769231,
-      "grad_norm": 6.173972129821777,
+      "grad_norm": 3.0283963680267334,
       "learning_rate": 0.00015384615384615385,
-      "loss": 0.2912,
+      "loss": 0.3345,
       "step": 60
     },
     {
       "epoch": 1.0769230769230769,
-      "grad_norm": 2.3577582836151123,
+      "grad_norm": 0.7587921619415283,
       "learning_rate": 0.00014615384615384615,
-      "loss": 0.1226,
+      "loss": 0.1903,
       "step": 70
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 0.14449569582939148,
+      "grad_norm": 0.2090693563222885,
       "learning_rate": 0.00013846153846153847,
-      "loss": 0.0796,
+      "loss": 0.0633,
       "step": 80
     },
     {
       "epoch": 1.3846153846153846,
-      "grad_norm": 4.846245288848877,
+      "grad_norm": 0.17675693333148956,
       "learning_rate": 0.00013076923076923077,
-      "loss": 0.0813,
+      "loss": 0.0942,
       "step": 90
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 3.6284821033477783,
+      "grad_norm": 0.648328423500061,
       "learning_rate": 0.0001230769230769231,
-      "loss": 0.0645,
+      "loss": 0.0783,
       "step": 100
     },
     {
       "epoch": 1.5384615384615383,
-      "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.09509244561195374,
-      "eval_runtime": 1.3146,
-      "eval_samples_per_second": 101.169,
-      "eval_steps_per_second": 12.931,
+      "eval_accuracy": 0.9849624060150376,
+      "eval_loss": 0.055133331567049026,
+      "eval_runtime": 2.1339,
+      "eval_samples_per_second": 62.326,
+      "eval_steps_per_second": 7.966,
       "step": 100
     },
     {
       "epoch": 1.6923076923076923,
-      "grad_norm": 2.209476947784424,
+      "grad_norm": 0.66573166847229,
       "learning_rate": 0.00011538461538461538,
-      "loss": 0.0932,
+      "loss": 0.1371,
       "step": 110
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 0.10745280981063843,
+      "grad_norm": 3.307586908340454,
       "learning_rate": 0.0001076923076923077,
-      "loss": 0.1311,
+      "loss": 0.0776,
       "step": 120
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.1176445409655571,
+      "grad_norm": 0.10508356243371964,
       "learning_rate": 0.0001,
-      "loss": 0.0584,
+      "loss": 0.0821,
       "step": 130
     },
     {
       "epoch": 2.1538461538461537,
-      "grad_norm": 0.11741472780704498,
+      "grad_norm": 0.14109393954277039,
       "learning_rate": 9.230769230769232e-05,
-      "loss": 0.0464,
+      "loss": 0.0622,
       "step": 140
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.5773084163665771,
+      "grad_norm": 0.2330102026462555,
       "learning_rate": 8.461538461538461e-05,
-      "loss": 0.0727,
+      "loss": 0.0204,
       "step": 150
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 0.5665645003318787,
+      "grad_norm": 0.10599979013204575,
       "learning_rate": 7.692307692307693e-05,
-      "loss": 0.0669,
+      "loss": 0.0219,
       "step": 160
     },
     {
       "epoch": 2.6153846153846154,
-      "grad_norm": 0.20813684165477753,
+      "grad_norm": 0.07637907564640045,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.0246,
+      "loss": 0.0149,
       "step": 170
     },
     {
       "epoch": 2.769230769230769,
-      "grad_norm": 0.12770630419254303,
+      "grad_norm": 0.06972856819629669,
       "learning_rate": 6.153846153846155e-05,
-      "loss": 0.0285,
+      "loss": 0.0383,
       "step": 180
     },
     {
       "epoch": 2.9230769230769234,
-      "grad_norm": 0.06980925053358078,
+      "grad_norm": 0.06536921858787537,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.0185,
+      "loss": 0.0135,
       "step": 190
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 0.07084397971630096,
+      "grad_norm": 0.06367834657430649,
       "learning_rate": 4.615384615384616e-05,
-      "loss": 0.0316,
+      "loss": 0.0124,
       "step": 200
     },
     {
       "epoch": 3.076923076923077,
-      "eval_accuracy": 0.9849624060150376,
-      "eval_loss": 0.05026606097817421,
-      "eval_runtime": 2.2826,
-      "eval_samples_per_second": 58.267,
-      "eval_steps_per_second": 7.448,
+      "eval_accuracy": 0.9924812030075187,
+      "eval_loss": 0.02266957052052021,
+      "eval_runtime": 1.2444,
+      "eval_samples_per_second": 106.875,
+      "eval_steps_per_second": 13.661,
       "step": 200
     },
     {
       "epoch": 3.230769230769231,
-      "grad_norm": 0.061542343348264694,
+      "grad_norm": 0.06352599710226059,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.0128,
+      "loss": 0.0133,
       "step": 210
     },
     {
       "epoch": 3.3846153846153846,
-      "grad_norm": 0.4158130884170532,
+      "grad_norm": 0.06076899170875549,
       "learning_rate": 3.0769230769230774e-05,
-      "loss": 0.0173,
+      "loss": 0.0123,
       "step": 220
     },
     {
       "epoch": 3.5384615384615383,
-      "grad_norm": 0.07470071315765381,
+      "grad_norm": 0.06032305210828781,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.0119,
+      "loss": 0.0114,
       "step": 230
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 0.0632583349943161,
+      "grad_norm": 0.05871906504034996,
       "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.0391,
+      "loss": 0.0111,
       "step": 240
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.46634915471076965,
+      "grad_norm": 0.06305041909217834,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.012,
+      "loss": 0.0112,
       "step": 250
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.056492287665605545,
+      "grad_norm": 0.060099028050899506,
       "learning_rate": 0.0,
-      "loss": 0.0116,
+      "loss": 0.0244,
       "step": 260
     },
     {
       "epoch": 4.0,
       "step": 260,
       "total_flos": 3.205097416476426e+17,
-      "train_loss": 0.11968801841139794,
-      "train_runtime": 105.0874,
-      "train_samples_per_second": 39.358,
-      "train_steps_per_second": 2.474
+      "train_loss": 0.12550590594227498,
+      "train_runtime": 104.2364,
+      "train_samples_per_second": 39.679,
+      "train_steps_per_second": 2.494
     }
   ],
   "logging_steps": 10,
@@ -230,7 +230,7 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0678555ad6faa886947d13cb58047927698f16f53c2db26bf31733267fecff14
-size 5112
+oid sha256:83e883a042c6d83197805380350a3ce4a9ce367789ac2974287d6580a562bfd4
+size 5304