vit-image-classification-demo-001

README.md

@@ -3,6 +3,7 @@ library_name: transformers
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -16,7 +17,7 @@ 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 an unknown dataset.
+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: 0.0147
 - Accuracy: 1.0

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.029698995873332024,
-    "eval_runtime": 1.285,
-    "eval_samples_per_second": 103.498,
-    "eval_steps_per_second": 13.229,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.014676429331302643,
+    "eval_runtime": 1.2739,
+    "eval_samples_per_second": 104.405,
+    "eval_steps_per_second": 13.345,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.11197870703270803,
-    "train_runtime": 48.716,
-    "train_samples_per_second": 84.9,
-    "train_steps_per_second": 5.337
+    "train_loss": 0.11465750181904207,
+    "train_runtime": 55.3297,
+    "train_samples_per_second": 74.752,
+    "train_steps_per_second": 4.699
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.029698995873332024,
-    "eval_runtime": 1.285,
-    "eval_samples_per_second": 103.498,
-    "eval_steps_per_second": 13.229
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.014676429331302643,
+    "eval_runtime": 1.2739,
+    "eval_samples_per_second": 104.405,
+    "eval_steps_per_second": 13.345
 }

runs/Dec27_05-43-52_mlxlabo4wdppro676a46b9-20241224052929-2obftk-9a0zen-worker/events.out.tfevents.1735278295.mlxlabo4wdppro676a46b9-20241224052929-2obftk-9a0zen-worker

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29b4aff7615ac2b71a90810f7d547a570f7b1b0fc295513bd40a5e2d7b849e28
+size 40

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.11197870703270803,
-    "train_runtime": 48.716,
-    "train_samples_per_second": 84.9,
-    "train_steps_per_second": 5.337
+    "train_loss": 0.11465750181904207,
+    "train_runtime": 55.3297,
+    "train_samples_per_second": 74.752,
+    "train_steps_per_second": 4.699
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.029698995873332024,
-  "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-100",
+  "best_metric": 0.014676429331302643,
+  "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": 1.9058457612991333,
+      "grad_norm": 1.7548131942749023,
       "learning_rate": 0.00019230769230769233,
-      "loss": 0.8049,
+      "loss": 0.8278,
       "step": 10
     },
     {
       "epoch": 0.3076923076923077,
-      "grad_norm": 2.1133065223693848,
+      "grad_norm": 3.528895378112793,
       "learning_rate": 0.00018461538461538463,
-      "loss": 0.3098,
+      "loss": 0.3232,
       "step": 20
     },
     {
       "epoch": 0.46153846153846156,
-      "grad_norm": 0.7872273325920105,
+      "grad_norm": 4.442903518676758,
       "learning_rate": 0.00017692307692307693,
-      "loss": 0.1921,
+      "loss": 0.2161,
       "step": 30
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 1.595115065574646,
+      "grad_norm": 0.861210286617279,
       "learning_rate": 0.00016923076923076923,
-      "loss": 0.1852,
+      "loss": 0.2552,
       "step": 40
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.2894037067890167,
+      "grad_norm": 8.467580795288086,
       "learning_rate": 0.00016153846153846155,
-      "loss": 0.1718,
+      "loss": 0.2294,
       "step": 50
     },
     {
       "epoch": 0.9230769230769231,
-      "grad_norm": 2.315903663635254,
+      "grad_norm": 2.317192316055298,
       "learning_rate": 0.00015384615384615385,
-      "loss": 0.3319,
+      "loss": 0.2226,
       "step": 60
     },
     {
       "epoch": 1.0769230769230769,
-      "grad_norm": 1.0175923109054565,
+      "grad_norm": 0.34912338852882385,
       "learning_rate": 0.00014615384615384615,
-      "loss": 0.1841,
+      "loss": 0.0866,
       "step": 70
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 0.1650422215461731,
+      "grad_norm": 0.17672394216060638,
       "learning_rate": 0.00013846153846153847,
-      "loss": 0.0914,
+      "loss": 0.0817,
       "step": 80
     },
     {
       "epoch": 1.3846153846153846,
-      "grad_norm": 0.11763770133256912,
+      "grad_norm": 0.1419631987810135,
       "learning_rate": 0.00013076923076923077,
-      "loss": 0.0511,
+      "loss": 0.0779,
       "step": 90
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.347074270248413,
+      "grad_norm": 2.7969260215759277,
       "learning_rate": 0.0001230769230769231,
-      "loss": 0.1013,
+      "loss": 0.0541,
       "step": 100
     },
     {
       "epoch": 1.5384615384615383,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.029698995873332024,
-      "eval_runtime": 1.3324,
-      "eval_samples_per_second": 99.817,
-      "eval_steps_per_second": 12.759,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.024188851937651634,
+      "eval_runtime": 1.4458,
+      "eval_samples_per_second": 91.994,
+      "eval_steps_per_second": 11.759,
       "step": 100
     },
     {
       "epoch": 1.6923076923076923,
-      "grad_norm": 2.937368154525757,
+      "grad_norm": 0.6725256443023682,
       "learning_rate": 0.00011538461538461538,
-      "loss": 0.0494,
+      "loss": 0.071,
       "step": 110
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 0.644463300704956,
+      "grad_norm": 3.316457509994507,
       "learning_rate": 0.0001076923076923077,
-      "loss": 0.0978,
+      "loss": 0.1595,
       "step": 120
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.965244472026825,
+      "grad_norm": 0.14894653856754303,
       "learning_rate": 0.0001,
-      "loss": 0.0929,
+      "loss": 0.0907,
       "step": 130
     },
     {
       "epoch": 2.1538461538461537,
-      "grad_norm": 0.10053160786628723,
+      "grad_norm": 0.10909508168697357,
       "learning_rate": 9.230769230769232e-05,
-      "loss": 0.0333,
+      "loss": 0.0613,
       "step": 140
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 0.08449896425008774,
+      "grad_norm": 0.13060985505580902,
       "learning_rate": 8.461538461538461e-05,
-      "loss": 0.0264,
+      "loss": 0.0203,
       "step": 150
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 0.11529306322336197,
+      "grad_norm": 0.08566407859325409,
       "learning_rate": 7.692307692307693e-05,
-      "loss": 0.035,
+      "loss": 0.0326,
       "step": 160
     },
     {
       "epoch": 2.6153846153846154,
-      "grad_norm": 0.06701377034187317,
+      "grad_norm": 0.07223280519247055,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.0138,
+      "loss": 0.0182,
       "step": 170
     },
     {
       "epoch": 2.769230769230769,
-      "grad_norm": 0.08366049826145172,
+      "grad_norm": 0.0757821574807167,
       "learning_rate": 6.153846153846155e-05,
-      "loss": 0.0521,
+      "loss": 0.0357,
       "step": 180
     },
     {
       "epoch": 2.9230769230769234,
-      "grad_norm": 0.089511439204216,
+      "grad_norm": 3.932257652282715,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.0133,
+      "loss": 0.0174,
       "step": 190
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 0.05841778218746185,
+      "grad_norm": 0.06355487555265427,
       "learning_rate": 4.615384615384616e-05,
-      "loss": 0.0118,
+      "loss": 0.014,
       "step": 200
     },
     {
       "epoch": 3.076923076923077,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.030333496630191803,
-      "eval_runtime": 1.3214,
-      "eval_samples_per_second": 100.65,
-      "eval_steps_per_second": 12.865,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.014676429331302643,
+      "eval_runtime": 1.3175,
+      "eval_samples_per_second": 100.95,
+      "eval_steps_per_second": 12.903,
       "step": 200
     },
     {
       "epoch": 3.230769230769231,
-      "grad_norm": 0.05673932656645775,
+      "grad_norm": 0.06115660443902016,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.0112,
+      "loss": 0.028,
       "step": 210
     },
     {
       "epoch": 3.3846153846153846,
-      "grad_norm": 0.056057512760162354,
+      "grad_norm": 0.06409073621034622,
       "learning_rate": 3.0769230769230774e-05,
-      "loss": 0.0104,
+      "loss": 0.0116,
       "step": 220
     },
     {
       "epoch": 3.5384615384615383,
-      "grad_norm": 0.06086534634232521,
+      "grad_norm": 0.05863337591290474,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.0104,
+      "loss": 0.0117,
       "step": 230
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 0.05317115783691406,
+      "grad_norm": 0.059554990381002426,
       "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.0101,
+      "loss": 0.0113,
       "step": 240
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.05058254674077034,
+      "grad_norm": 0.05857035890221596,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.0097,
+      "loss": 0.011,
       "step": 250
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.05316939577460289,
+      "grad_norm": 0.06412912160158157,
       "learning_rate": 0.0,
-      "loss": 0.0103,
+      "loss": 0.012,
       "step": 260
     },
     {
       "epoch": 4.0,
       "step": 260,
       "total_flos": 3.205097416476426e+17,
-      "train_loss": 0.11197870703270803,
-      "train_runtime": 48.716,
-      "train_samples_per_second": 84.9,
-      "train_steps_per_second": 5.337
+      "train_loss": 0.11465750181904207,
+      "train_runtime": 55.3297,
+      "train_samples_per_second": 74.752,
+      "train_steps_per_second": 4.699
     }
   ],
   "logging_steps": 10,