cheers again

README.md

@@ -2,6 +2,7 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
 - generated_from_trainer
 metrics:
 - accuracy
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-beans-tutorial
 
-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: 0.1436
-- Accuracy: 0.9850
+- Loss: 0.3187
+- Accuracy: 0.9609
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
-    "eval_accuracy": 0.953125,
-    "eval_loss": 0.2592126727104187,
-    "eval_runtime": 1.9731,
-    "eval_samples_per_second": 64.873,
-    "eval_steps_per_second": 8.109,
+    "eval_accuracy": 0.9924812030075187,
+    "eval_loss": 0.07803888618946075,
+    "eval_runtime": 2.0235,
+    "eval_samples_per_second": 65.728,
+    "eval_steps_per_second": 8.401,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.000326924599134005,
-    "train_runtime": 192.0253,
-    "train_samples_per_second": 26.924,
-    "train_steps_per_second": 3.385
+    "train_loss": 2.7056336402893067e-05,
+    "train_runtime": 203.0893,
+    "train_samples_per_second": 25.457,
+    "train_steps_per_second": 3.201
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.021117208525538445,
-    "eval_runtime": 2.0833,
-    "eval_samples_per_second": 63.842,
-    "eval_steps_per_second": 8.16
+    "eval_loss": 0.07803888618946075,
+    "eval_runtime": 2.0235,
+    "eval_samples_per_second": 65.728,
+    "eval_steps_per_second": 8.401
 }

runs/Jan24_01-26-08_579caf070ae1/events.out.tfevents.1706059797.579caf070ae1.3225.10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47598497725d9005b232d1deaaba60d8d8bf5c6970a8e3b46876060b5809c662
+size 734

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.000326924599134005,
-    "train_runtime": 192.0253,
-    "train_samples_per_second": 26.924,
-    "train_steps_per_second": 3.385
+    "train_loss": 2.7056336402893067e-05,
+    "train_runtime": 203.0893,
+    "train_samples_per_second": 25.457,
+    "train_steps_per_second": 3.201
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.021117208525538445,
+  "best_metric": 0.07803888618946075,
   "best_model_checkpoint": "./vit-base-beans-tutorial/checkpoint-100",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,456 +10,456 @@
   "log_history": [
     {
       "epoch": 0.08,
-      "learning_rate": 1.9692307692307696e-05,
-      "loss": 0.0007,
+      "learning_rate": 0.00019692307692307696,
+      "loss": 0.0004,
       "step": 10
     },
     {
       "epoch": 0.15,
-      "learning_rate": 1.9384615384615386e-05,
-      "loss": 0.0007,
+      "learning_rate": 0.00019384615384615385,
+      "loss": 0.0002,
       "step": 20
     },
     {
       "epoch": 0.23,
-      "learning_rate": 1.907692307692308e-05,
-      "loss": 0.0006,
+      "learning_rate": 0.0001907692307692308,
+      "loss": 0.0002,
       "step": 30
     },
     {
       "epoch": 0.31,
-      "learning_rate": 1.876923076923077e-05,
-      "loss": 0.0006,
+      "learning_rate": 0.0001876923076923077,
+      "loss": 0.0001,
       "step": 40
     },
     {
       "epoch": 0.38,
-      "learning_rate": 1.8461538461538465e-05,
-      "loss": 0.001,
+      "learning_rate": 0.00018461538461538463,
+      "loss": 0.0001,
       "step": 50
     },
     {
       "epoch": 0.46,
-      "learning_rate": 1.8153846153846155e-05,
-      "loss": 0.0005,
+      "learning_rate": 0.00018153846153846155,
+      "loss": 0.0001,
       "step": 60
     },
     {
       "epoch": 0.54,
-      "learning_rate": 1.784615384615385e-05,
-      "loss": 0.0005,
+      "learning_rate": 0.00017846153846153847,
+      "loss": 0.0001,
       "step": 70
     },
     {
       "epoch": 0.62,
-      "learning_rate": 1.753846153846154e-05,
-      "loss": 0.0005,
+      "learning_rate": 0.0001753846153846154,
+      "loss": 0.0,
       "step": 80
     },
     {
       "epoch": 0.69,
-      "learning_rate": 1.7230769230769234e-05,
-      "loss": 0.0005,
+      "learning_rate": 0.00017230769230769234,
+      "loss": 0.0,
       "step": 90
     },
     {
       "epoch": 0.77,
-      "learning_rate": 1.6923076923076924e-05,
-      "loss": 0.0005,
+      "learning_rate": 0.00016923076923076923,
+      "loss": 0.0,
       "step": 100
     },
     {
       "epoch": 0.77,
       "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.021117208525538445,
-      "eval_runtime": 1.8062,
-      "eval_samples_per_second": 73.637,
-      "eval_steps_per_second": 9.412,
+      "eval_loss": 0.07803888618946075,
+      "eval_runtime": 2.3471,
+      "eval_samples_per_second": 56.666,
+      "eval_steps_per_second": 7.243,
       "step": 100
     },
     {
       "epoch": 0.85,
-      "learning_rate": 1.6615384615384618e-05,
-      "loss": 0.0005,
+      "learning_rate": 0.00016615384615384617,
+      "loss": 0.0,
       "step": 110
     },
     {
       "epoch": 0.92,
-      "learning_rate": 1.630769230769231e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.0001630769230769231,
+      "loss": 0.0,
       "step": 120
     },
     {
       "epoch": 1.0,
-      "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00016,
+      "loss": 0.0,
       "step": 130
     },
     {
       "epoch": 1.08,
-      "learning_rate": 1.5692307692307693e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00015692307692307693,
+      "loss": 0.0,
       "step": 140
     },
     {
       "epoch": 1.15,
-      "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00015384615384615385,
+      "loss": 0.0,
       "step": 150
     },
     {
       "epoch": 1.23,
-      "learning_rate": 1.5076923076923078e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00015076923076923077,
+      "loss": 0.0,
       "step": 160
     },
     {
       "epoch": 1.31,
-      "learning_rate": 1.4769230769230772e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00014769230769230772,
+      "loss": 0.0,
       "step": 170
     },
     {
       "epoch": 1.38,
-      "learning_rate": 1.4461538461538462e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.0001446153846153846,
+      "loss": 0.0,
       "step": 180
     },
     {
       "epoch": 1.46,
-      "learning_rate": 1.4153846153846156e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00014153846153846156,
+      "loss": 0.0,
       "step": 190
     },
     {
       "epoch": 1.54,
-      "learning_rate": 1.3846153846153847e-05,
-      "loss": 0.0004,
+      "learning_rate": 0.00013846153846153847,
+      "loss": 0.0,
       "step": 200
     },
     {
       "epoch": 1.54,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.032906562089920044,
-      "eval_runtime": 1.8659,
-      "eval_samples_per_second": 71.281,
-      "eval_steps_per_second": 9.111,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.1820073425769806,
+      "eval_runtime": 1.9046,
+      "eval_samples_per_second": 69.83,
+      "eval_steps_per_second": 8.926,
       "step": 200
     },
     {
       "epoch": 1.62,
-      "learning_rate": 1.353846153846154e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.0001353846153846154,
+      "loss": 0.0,
       "step": 210
     },
     {
       "epoch": 1.69,
-      "learning_rate": 1.3230769230769231e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.0001323076923076923,
+      "loss": 0.0,
       "step": 220
     },
     {
       "epoch": 1.77,
-      "learning_rate": 1.2923076923076925e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00012923076923076923,
+      "loss": 0.0,
       "step": 230
     },
     {
       "epoch": 1.85,
-      "learning_rate": 1.2615384615384616e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00012615384615384615,
+      "loss": 0.0,
       "step": 240
     },
     {
       "epoch": 1.92,
-      "learning_rate": 1.230769230769231e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.0001230769230769231,
+      "loss": 0.0,
       "step": 250
     },
     {
       "epoch": 2.0,
-      "learning_rate": 1.2e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00012,
+      "loss": 0.0,
       "step": 260
     },
     {
       "epoch": 2.08,
-      "learning_rate": 1.1692307692307694e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00011692307692307694,
+      "loss": 0.0,
       "step": 270
     },
     {
       "epoch": 2.15,
-      "learning_rate": 1.1384615384615385e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00011384615384615384,
+      "loss": 0.0,
       "step": 280
     },
     {
       "epoch": 2.23,
-      "learning_rate": 1.1076923076923079e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00011076923076923077,
+      "loss": 0.0,
       "step": 290
     },
     {
       "epoch": 2.31,
-      "learning_rate": 1.076923076923077e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.0001076923076923077,
+      "loss": 0.0,
       "step": 300
     },
     {
       "epoch": 2.31,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.03912555053830147,
-      "eval_runtime": 1.8293,
-      "eval_samples_per_second": 72.704,
-      "eval_steps_per_second": 9.293,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.1845720261335373,
+      "eval_runtime": 1.7918,
+      "eval_samples_per_second": 74.226,
+      "eval_steps_per_second": 9.488,
       "step": 300
     },
     {
       "epoch": 2.38,
-      "learning_rate": 1.0461538461538463e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00010461538461538463,
+      "loss": 0.0,
       "step": 310
     },
     {
       "epoch": 2.46,
-      "learning_rate": 1.0153846153846154e-05,
-      "loss": 0.0003,
+      "learning_rate": 0.00010153846153846153,
+      "loss": 0.0,
       "step": 320
     },
     {
       "epoch": 2.54,
-      "learning_rate": 9.846153846153848e-06,
-      "loss": 0.0003,
+      "learning_rate": 9.846153846153848e-05,
+      "loss": 0.0,
       "step": 330
     },
     {
       "epoch": 2.62,
-      "learning_rate": 9.53846153846154e-06,
-      "loss": 0.0003,
+      "learning_rate": 9.53846153846154e-05,
+      "loss": 0.0,
       "step": 340
     },
     {
       "epoch": 2.69,
-      "learning_rate": 9.230769230769232e-06,
-      "loss": 0.0003,
+      "learning_rate": 9.230769230769232e-05,
+      "loss": 0.0,
       "step": 350
     },
     {
       "epoch": 2.77,
-      "learning_rate": 8.923076923076925e-06,
-      "loss": 0.0003,
+      "learning_rate": 8.923076923076924e-05,
+      "loss": 0.0,
       "step": 360
     },
     {
       "epoch": 2.85,
-      "learning_rate": 8.615384615384617e-06,
-      "loss": 0.0003,
+      "learning_rate": 8.615384615384617e-05,
+      "loss": 0.0,
       "step": 370
     },
     {
       "epoch": 2.92,
-      "learning_rate": 8.307692307692309e-06,
-      "loss": 0.0002,
+      "learning_rate": 8.307692307692309e-05,
+      "loss": 0.0,
       "step": 380
     },
     {
       "epoch": 3.0,
-      "learning_rate": 8.000000000000001e-06,
-      "loss": 0.0002,
+      "learning_rate": 8e-05,
+      "loss": 0.0,
       "step": 390
     },
     {
       "epoch": 3.08,
-      "learning_rate": 7.692307692307694e-06,
-      "loss": 0.0002,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 0.0,
       "step": 400
     },
     {
       "epoch": 3.08,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.043412912636995316,
-      "eval_runtime": 2.0859,
-      "eval_samples_per_second": 63.761,
-      "eval_steps_per_second": 8.15,
+      "eval_accuracy": 0.9774436090225563,
+      "eval_loss": 0.1543329805135727,
+      "eval_runtime": 1.7915,
+      "eval_samples_per_second": 74.24,
+      "eval_steps_per_second": 9.489,
       "step": 400
     },
     {
       "epoch": 3.15,
-      "learning_rate": 7.384615384615386e-06,
-      "loss": 0.0002,
+      "learning_rate": 7.384615384615386e-05,
+      "loss": 0.0,
       "step": 410
     },
     {
       "epoch": 3.23,
-      "learning_rate": 7.076923076923078e-06,
-      "loss": 0.0002,
+      "learning_rate": 7.076923076923078e-05,
+      "loss": 0.0,
       "step": 420
     },
     {
       "epoch": 3.31,
-      "learning_rate": 6.76923076923077e-06,
-      "loss": 0.0002,
+      "learning_rate": 6.76923076923077e-05,
+      "loss": 0.0,
       "step": 430
     },
     {
       "epoch": 3.38,
-      "learning_rate": 6.461538461538463e-06,
-      "loss": 0.0002,
+      "learning_rate": 6.461538461538462e-05,
+      "loss": 0.0,
       "step": 440
     },
     {
       "epoch": 3.46,
-      "learning_rate": 6.153846153846155e-06,
-      "loss": 0.0002,
+      "learning_rate": 6.153846153846155e-05,
+      "loss": 0.0,
       "step": 450
     },
     {
       "epoch": 3.54,
-      "learning_rate": 5.846153846153847e-06,
-      "loss": 0.0002,
+      "learning_rate": 5.846153846153847e-05,
+      "loss": 0.0,
       "step": 460
     },
     {
       "epoch": 3.62,
-      "learning_rate": 5.538461538461539e-06,
-      "loss": 0.0002,
+      "learning_rate": 5.538461538461539e-05,
+      "loss": 0.0,
       "step": 470
     },
     {
       "epoch": 3.69,
-      "learning_rate": 5.230769230769232e-06,
-      "loss": 0.0002,
+      "learning_rate": 5.230769230769231e-05,
+      "loss": 0.0,
       "step": 480
     },
     {
       "epoch": 3.77,
-      "learning_rate": 4.923076923076924e-06,
-      "loss": 0.0002,
+      "learning_rate": 4.923076923076924e-05,
+      "loss": 0.0,
       "step": 490
     },
     {
       "epoch": 3.85,
-      "learning_rate": 4.615384615384616e-06,
-      "loss": 0.0002,
+      "learning_rate": 4.615384615384616e-05,
+      "loss": 0.0,
       "step": 500
     },
     {
       "epoch": 3.85,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.045747168362140656,
-      "eval_runtime": 1.754,
-      "eval_samples_per_second": 75.825,
-      "eval_steps_per_second": 9.692,
+      "eval_accuracy": 0.9849624060150376,
+      "eval_loss": 0.13960087299346924,
+      "eval_runtime": 1.797,
+      "eval_samples_per_second": 74.011,
+      "eval_steps_per_second": 9.46,
       "step": 500
     },
     {
       "epoch": 3.92,
-      "learning_rate": 4.307692307692308e-06,
-      "loss": 0.0002,
+      "learning_rate": 4.3076923076923084e-05,
+      "loss": 0.0,
       "step": 510
     },
     {
       "epoch": 4.0,
-      "learning_rate": 4.000000000000001e-06,
-      "loss": 0.0002,
+      "learning_rate": 4e-05,
+      "loss": 0.0,
       "step": 520
     },
     {
       "epoch": 4.08,
-      "learning_rate": 3.692307692307693e-06,
-      "loss": 0.0002,
+      "learning_rate": 3.692307692307693e-05,
+      "loss": 0.0,
       "step": 530
     },
     {
       "epoch": 4.15,
-      "learning_rate": 3.384615384615385e-06,
-      "loss": 0.0002,
+      "learning_rate": 3.384615384615385e-05,
+      "loss": 0.0,
       "step": 540
     },
     {
       "epoch": 4.23,
-      "learning_rate": 3.0769230769230774e-06,
-      "loss": 0.0002,
+      "learning_rate": 3.0769230769230774e-05,
+      "loss": 0.0,
       "step": 550
     },
     {
       "epoch": 4.31,
-      "learning_rate": 2.7692307692307697e-06,
-      "loss": 0.0002,
+      "learning_rate": 2.7692307692307694e-05,
+      "loss": 0.0,
       "step": 560
     },
     {
       "epoch": 4.38,
-      "learning_rate": 2.461538461538462e-06,
-      "loss": 0.0002,
+      "learning_rate": 2.461538461538462e-05,
+      "loss": 0.0,
       "step": 570
     },
     {
       "epoch": 4.46,
-      "learning_rate": 2.153846153846154e-06,
-      "loss": 0.0002,
+      "learning_rate": 2.1538461538461542e-05,
+      "loss": 0.0,
       "step": 580
     },
     {
       "epoch": 4.54,
-      "learning_rate": 1.8461538461538465e-06,
-      "loss": 0.0002,
+      "learning_rate": 1.8461538461538465e-05,
+      "loss": 0.0,
       "step": 590
     },
     {
       "epoch": 4.62,
-      "learning_rate": 1.5384615384615387e-06,
-      "loss": 0.0002,
+      "learning_rate": 1.5384615384615387e-05,
+      "loss": 0.0,
       "step": 600
     },
     {
       "epoch": 4.62,
       "eval_accuracy": 0.9849624060150376,
-      "eval_loss": 0.04757217317819595,
-      "eval_runtime": 1.773,
-      "eval_samples_per_second": 75.014,
-      "eval_steps_per_second": 9.588,
+      "eval_loss": 0.14361034333705902,
+      "eval_runtime": 1.8005,
+      "eval_samples_per_second": 73.867,
+      "eval_steps_per_second": 9.442,
       "step": 600
     },
     {
       "epoch": 4.69,
-      "learning_rate": 1.230769230769231e-06,
-      "loss": 0.0002,
+      "learning_rate": 1.230769230769231e-05,
+      "loss": 0.0,
       "step": 610
     },
     {
       "epoch": 4.77,
-      "learning_rate": 9.230769230769232e-07,
-      "loss": 0.0002,
+      "learning_rate": 9.230769230769232e-06,
+      "loss": 0.0,
       "step": 620
     },
     {
       "epoch": 4.85,
-      "learning_rate": 6.153846153846155e-07,
-      "loss": 0.0002,
+      "learning_rate": 6.153846153846155e-06,
+      "loss": 0.0,
       "step": 630
     },
     {
       "epoch": 4.92,
-      "learning_rate": 3.0769230769230774e-07,
-      "loss": 0.0002,
+      "learning_rate": 3.0769230769230774e-06,
+      "loss": 0.0,
       "step": 640
     },
     {
       "epoch": 5.0,
       "learning_rate": 0.0,
-      "loss": 0.0002,
+      "loss": 0.0,
       "step": 650
     },
     {
       "epoch": 5.0,
       "step": 650,
       "total_flos": 4.006371770595533e+17,
-      "train_loss": 0.000326924599134005,
-      "train_runtime": 192.0253,
-      "train_samples_per_second": 26.924,
-      "train_steps_per_second": 3.385
+      "train_loss": 2.7056336402893067e-05,
+      "train_runtime": 203.0893,
+      "train_samples_per_second": 25.457,
+      "train_steps_per_second": 3.201
     }
   ],
   "logging_steps": 10,