SeungYon

README.md

@@ -1,4 +1,5 @@
 ---
+library_name: transformers
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
@@ -19,8 +20,8 @@ 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 beans dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1428
-- Accuracy: 0.9531
+- Loss: 0.1349
+- Accuracy: 0.9688
 
 ## Model description
 
@@ -52,13 +53,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Accuracy |
 |:-------------:|:------:|:----:|:---------------:|:--------:|
-| 0.0402        | 1.5385 | 100  | 0.0243          | 1.0      |
-| 0.0432        | 3.0769 | 200  | 0.0460          | 0.9850   |
+| 0.0721        | 1.5385 | 100  | 0.0832          | 0.9699   |
+| 0.0117        | 3.0769 | 200  | 0.0280          | 0.9925   |
 
 
 ### Framework versions
 
-- Transformers 4.41.2
-- Pytorch 2.3.0+cu121
-- Datasets 2.20.0
+- Transformers 4.44.2
+- Pytorch 2.4.0+cu121
+- Datasets 3.0.0
 - Tokenizers 0.19.1

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.953125,
-    "eval_loss": 0.14284944534301758,
-    "eval_runtime": 1.9368,
-    "eval_samples_per_second": 66.09,
-    "eval_steps_per_second": 8.261,
+    "eval_accuracy": 0.96875,
+    "eval_loss": 0.13487988710403442,
+    "eval_runtime": 1.5292,
+    "eval_samples_per_second": 83.705,
+    "eval_steps_per_second": 10.463,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.10438386141680754,
-    "train_runtime": 106.8925,
-    "train_samples_per_second": 38.693,
-    "train_steps_per_second": 2.432
+    "train_loss": 0.11813208366242739,
+    "train_runtime": 92.1542,
+    "train_samples_per_second": 44.881,
+    "train_steps_per_second": 2.821
 }

config.json

@@ -30,5 +30,5 @@
   "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.41.2"
+  "transformers_version": "4.44.2"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3191f9c1cd7d129c94cfb88d62fef73bc68e43903656182ea13655c8627c1c43
+oid sha256:b9bc52c9fd99c203eaff06a385304eb30e4cc0db86b56e47aeccbc3026affc3e
 size 343227052

preprocessor_config.json

@@ -1,18 +1,4 @@
 {
-  "_valid_processor_keys": [
-    "images",
-    "do_resize",
-    "size",
-    "resample",
-    "do_rescale",
-    "rescale_factor",
-    "do_normalize",
-    "image_mean",
-    "image_std",
-    "return_tensors",
-    "data_format",
-    "input_data_format"
-  ],
   "do_normalize": true,
   "do_rescale": true,
   "do_resize": true,

runs/Sep14_22-40-14_b937d197a4d6/events.out.tfevents.1726353797.b937d197a4d6.320.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fa89e299ca442de11f65a7d7420fa2031307eebacb28ecba3a246aed7a40739
+size 11419

runs/Sep14_22-40-14_b937d197a4d6/events.out.tfevents.1726354019.b937d197a4d6.320.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0d8b792f945c88198b93270c0b0bdfb7e5eed042fda70e8abbee1473eb72810
+size 411

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.953125,
-    "eval_loss": 0.14284944534301758,
-    "eval_runtime": 1.9368,
-    "eval_samples_per_second": 66.09,
-    "eval_steps_per_second": 8.261
+    "eval_accuracy": 0.96875,
+    "eval_loss": 0.13487988710403442,
+    "eval_runtime": 1.5292,
+    "eval_samples_per_second": 83.705,
+    "eval_steps_per_second": 10.463
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.10438386141680754,
-    "train_runtime": 106.8925,
-    "train_samples_per_second": 38.693,
-    "train_steps_per_second": 2.432
+    "train_loss": 0.11813208366242739,
+    "train_runtime": 92.1542,
+    "train_samples_per_second": 44.881,
+    "train_steps_per_second": 2.821
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.024255966767668724,
-  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-100",
+  "best_metric": 0.027972575277090073,
+  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
   "epoch": 4.0,
   "eval_steps": 100,
   "global_step": 260,
@@ -10,212 +10,212 @@
   "log_history": [
     {
       "epoch": 0.15384615384615385,
-      "grad_norm": 1.332680106163025,
+      "grad_norm": 1.7359464168548584,
       "learning_rate": 0.00019230769230769233,
-      "loss": 0.8087,
+      "loss": 0.9068,
       "step": 10
     },
     {
       "epoch": 0.3076923076923077,
-      "grad_norm": 1.3357136249542236,
+      "grad_norm": 1.7515268325805664,
       "learning_rate": 0.00018461538461538463,
-      "loss": 0.2963,
+      "loss": 0.3442,
       "step": 20
     },
     {
       "epoch": 0.46153846153846156,
-      "grad_norm": 2.7216763496398926,
+      "grad_norm": 3.3439853191375732,
       "learning_rate": 0.00017692307692307693,
-      "loss": 0.1797,
+      "loss": 0.1684,
       "step": 30
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 0.6214920282363892,
+      "grad_norm": 2.0178005695343018,
       "learning_rate": 0.00016923076923076923,
-      "loss": 0.1874,
+      "loss": 0.3005,
       "step": 40
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.34344300627708435,
+      "grad_norm": 1.4199612140655518,
       "learning_rate": 0.00016153846153846155,
-      "loss": 0.1499,
+      "loss": 0.1432,
       "step": 50
     },
     {
       "epoch": 0.9230769230769231,
-      "grad_norm": 0.2999103367328644,
+      "grad_norm": 3.5320632457733154,
       "learning_rate": 0.00015384615384615385,
-      "loss": 0.1821,
+      "loss": 0.3094,
       "step": 60
     },
     {
       "epoch": 1.0769230769230769,
-      "grad_norm": 1.7142621278762817,
+      "grad_norm": 3.943850040435791,
       "learning_rate": 0.00014615384615384615,
-      "loss": 0.1567,
+      "loss": 0.164,
       "step": 70
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 0.15415795147418976,
+      "grad_norm": 0.42932555079460144,
       "learning_rate": 0.00013846153846153847,
-      "loss": 0.1337,
+      "loss": 0.0936,
       "step": 80
     },
     {
       "epoch": 1.3846153846153846,
-      "grad_norm": 0.12295465916395187,
+      "grad_norm": 0.2571287155151367,
       "learning_rate": 0.00013076923076923077,
-      "loss": 0.0351,
+      "loss": 0.0563,
       "step": 90
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.2825920581817627,
+      "grad_norm": 0.2423134446144104,
       "learning_rate": 0.0001230769230769231,
-      "loss": 0.0402,
+      "loss": 0.0721,
       "step": 100
     },
     {
       "epoch": 1.5384615384615383,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.024255966767668724,
-      "eval_runtime": 1.8405,
-      "eval_samples_per_second": 72.264,
-      "eval_steps_per_second": 9.237,
+      "eval_accuracy": 0.9699248120300752,
+      "eval_loss": 0.08323061466217041,
+      "eval_runtime": 1.2671,
+      "eval_samples_per_second": 104.964,
+      "eval_steps_per_second": 13.416,
       "step": 100
     },
     {
       "epoch": 1.6923076923076923,
-      "grad_norm": 0.6791179776191711,
+      "grad_norm": 3.0321273803710938,
       "learning_rate": 0.00011538461538461538,
-      "loss": 0.0963,
+      "loss": 0.1154,
       "step": 110
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 5.622032642364502,
+      "grad_norm": 0.13413630425930023,
       "learning_rate": 0.0001076923076923077,
-      "loss": 0.1162,
+      "loss": 0.0794,
       "step": 120
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.0946173444390297,
+      "grad_norm": 0.14623944461345673,
       "learning_rate": 0.0001,
-      "loss": 0.0762,
+      "loss": 0.0543,
       "step": 130
     },
     {
       "epoch": 2.1538461538461537,
-      "grad_norm": 0.10486400872468948,
+      "grad_norm": 0.12642432749271393,
       "learning_rate": 9.230769230769232e-05,
-      "loss": 0.0352,
+      "loss": 0.0683,
       "step": 140
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 0.10110525041818619,
+      "grad_norm": 0.09180299937725067,
       "learning_rate": 8.461538461538461e-05,
-      "loss": 0.02,
+      "loss": 0.0452,
       "step": 150
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 1.2142821550369263,
+      "grad_norm": 3.4162721633911133,
       "learning_rate": 7.692307692307693e-05,
-      "loss": 0.0167,
+      "loss": 0.0202,
       "step": 160
     },
     {
       "epoch": 2.6153846153846154,
-      "grad_norm": 0.10432332754135132,
+      "grad_norm": 0.06922739744186401,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.0155,
+      "loss": 0.0191,
       "step": 170
     },
     {
       "epoch": 2.769230769230769,
-      "grad_norm": 0.06912260502576828,
+      "grad_norm": 0.26483821868896484,
       "learning_rate": 6.153846153846155e-05,
-      "loss": 0.0135,
+      "loss": 0.0165,
       "step": 180
     },
     {
       "epoch": 2.9230769230769234,
-      "grad_norm": 0.06139219552278519,
+      "grad_norm": 0.0765213593840599,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.0126,
+      "loss": 0.0188,
       "step": 190
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 0.061153631657361984,
+      "grad_norm": 0.06417355686426163,
       "learning_rate": 4.615384615384616e-05,
-      "loss": 0.0432,
+      "loss": 0.0117,
       "step": 200
     },
     {
       "epoch": 3.076923076923077,
-      "eval_accuracy": 0.9849624060150376,
-      "eval_loss": 0.04604126885533333,
-      "eval_runtime": 2.5635,
-      "eval_samples_per_second": 51.883,
-      "eval_steps_per_second": 6.632,
+      "eval_accuracy": 0.9924812030075187,
+      "eval_loss": 0.027972575277090073,
+      "eval_runtime": 1.8447,
+      "eval_samples_per_second": 72.097,
+      "eval_steps_per_second": 9.215,
       "step": 200
     },
     {
       "epoch": 3.230769230769231,
-      "grad_norm": 0.05933067575097084,
+      "grad_norm": 0.05795031413435936,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.0237,
+      "loss": 0.011,
       "step": 210
     },
     {
       "epoch": 3.3846153846153846,
-      "grad_norm": 0.05889052525162697,
+      "grad_norm": 0.055242665112018585,
       "learning_rate": 3.0769230769230774e-05,
-      "loss": 0.0294,
+      "loss": 0.0108,
       "step": 220
     },
     {
       "epoch": 3.5384615384615383,
-      "grad_norm": 0.058969203382730484,
+      "grad_norm": 0.057339392602443695,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.0113,
+      "loss": 0.0118,
       "step": 230
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 0.058085132390260696,
+      "grad_norm": 0.05471523106098175,
       "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.0119,
+      "loss": 0.0101,
       "step": 240
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.057358331978321075,
+      "grad_norm": 0.05334159731864929,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.011,
+      "loss": 0.0101,
       "step": 250
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.05577138066291809,
+      "grad_norm": 0.06180350109934807,
       "learning_rate": 0.0,
-      "loss": 0.0113,
+      "loss": 0.0103,
       "step": 260
     },
     {
       "epoch": 4.0,
       "step": 260,
       "total_flos": 3.205097416476426e+17,
-      "train_loss": 0.10438386141680754,
-      "train_runtime": 106.8925,
-      "train_samples_per_second": 38.693,
-      "train_steps_per_second": 2.432
+      "train_loss": 0.11813208366242739,
+      "train_runtime": 92.1542,
+      "train_samples_per_second": 44.881,
+      "train_steps_per_second": 2.821
     }
   ],
   "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:a29690e7f0bcd0acf9515499315855c09ae67e6d00ed43d1b8c5b7bad9f1442f
-size 5112
+oid sha256:3d8e8a284a4f3063ff4700fbd9d3d2e33bc687bd358e2d0101b73cf0d00d4e82
+size 5176