🍻 cheers

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. -->
 
 # fish_disease_datasets
 
-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 fish_disease_datasets dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.0704
 - Accuracy: 0.9810

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9782608695652174,
-    "eval_loss": 0.08242087066173553,
-    "eval_runtime": 1.5232,
-    "eval_samples_per_second": 241.595,
-    "eval_steps_per_second": 30.199,
+    "eval_accuracy": 0.9809782608695652,
+    "eval_loss": 0.07041697204113007,
+    "eval_runtime": 1.7197,
+    "eval_samples_per_second": 213.985,
+    "eval_steps_per_second": 26.748,
     "total_flos": 6.45382209997357e+17,
-    "train_loss": 0.26906703446657604,
-    "train_runtime": 211.2966,
-    "train_samples_per_second": 39.414,
-    "train_steps_per_second": 2.48
+    "train_loss": 0.27801662728986665,
+    "train_runtime": 166.0361,
+    "train_samples_per_second": 50.158,
+    "train_steps_per_second": 3.156
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9782608695652174,
-    "eval_loss": 0.08242087066173553,
-    "eval_runtime": 1.5232,
-    "eval_samples_per_second": 241.595,
-    "eval_steps_per_second": 30.199
+    "eval_accuracy": 0.9809782608695652,
+    "eval_loss": 0.07041697204113007,
+    "eval_runtime": 1.7197,
+    "eval_samples_per_second": 213.985,
+    "eval_steps_per_second": 26.748
 }

runs/May23_06-49-46_1fcbf54ed8fc/events.out.tfevents.1747983261.1fcbf54ed8fc.6395.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e90236846232d71a5c41e7420c83215a8162ffdc4437660722c63ba0c8c3117f
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 6.45382209997357e+17,
-    "train_loss": 0.26906703446657604,
-    "train_runtime": 211.2966,
-    "train_samples_per_second": 39.414,
-    "train_steps_per_second": 2.48
+    "train_loss": 0.27801662728986665,
+    "train_runtime": 166.0361,
+    "train_samples_per_second": 50.158,
+    "train_steps_per_second": 3.156
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 500,
-  "best_metric": 0.08242087066173553,
+  "best_metric": 0.07041697204113007,
   "best_model_checkpoint": "fish_disease_datasets/checkpoint-500",
   "epoch": 4.0,
   "eval_steps": 100,
@@ -11,421 +11,421 @@
   "log_history": [
     {
       "epoch": 0.07633587786259542,
-      "grad_norm": 2.050231695175171,
+      "grad_norm": 1.8386459350585938,
       "learning_rate": 0.00019656488549618322,
-      "loss": 1.8355,
+      "loss": 1.8727,
       "step": 10
     },
     {
       "epoch": 0.15267175572519084,
-      "grad_norm": 2.2453360557556152,
+      "grad_norm": 2.027068614959717,
       "learning_rate": 0.00019274809160305345,
-      "loss": 1.5538,
+      "loss": 1.6646,
       "step": 20
     },
     {
       "epoch": 0.22900763358778625,
-      "grad_norm": 1.8408141136169434,
+      "grad_norm": 1.9189614057540894,
       "learning_rate": 0.00018893129770992367,
-      "loss": 1.354,
+      "loss": 1.4059,
       "step": 30
     },
     {
       "epoch": 0.3053435114503817,
-      "grad_norm": 2.0277440547943115,
+      "grad_norm": 2.0967509746551514,
       "learning_rate": 0.0001851145038167939,
-      "loss": 1.035,
+      "loss": 1.0754,
       "step": 40
     },
     {
       "epoch": 0.3816793893129771,
-      "grad_norm": 2.367299795150757,
+      "grad_norm": 2.0452075004577637,
       "learning_rate": 0.00018129770992366412,
-      "loss": 0.9501,
+      "loss": 0.9438,
       "step": 50
     },
     {
       "epoch": 0.4580152671755725,
-      "grad_norm": 3.103821277618408,
+      "grad_norm": 2.0183463096618652,
       "learning_rate": 0.00017748091603053437,
-      "loss": 0.6932,
+      "loss": 0.8577,
       "step": 60
     },
     {
       "epoch": 0.5343511450381679,
-      "grad_norm": 2.1959078311920166,
+      "grad_norm": 3.3413853645324707,
       "learning_rate": 0.0001736641221374046,
-      "loss": 0.5865,
+      "loss": 0.8169,
       "step": 70
     },
     {
       "epoch": 0.6106870229007634,
-      "grad_norm": 3.5142126083374023,
+      "grad_norm": 2.440781593322754,
       "learning_rate": 0.00016984732824427482,
-      "loss": 0.5495,
+      "loss": 0.6525,
       "step": 80
     },
     {
       "epoch": 0.6870229007633588,
-      "grad_norm": 3.4135642051696777,
+      "grad_norm": 3.064451217651367,
       "learning_rate": 0.00016603053435114505,
-      "loss": 0.5116,
+      "loss": 0.4708,
       "step": 90
     },
     {
       "epoch": 0.7633587786259542,
-      "grad_norm": 1.4583014249801636,
+      "grad_norm": 2.395552396774292,
       "learning_rate": 0.00016221374045801527,
-      "loss": 0.4914,
+      "loss": 0.4357,
       "step": 100
     },
     {
       "epoch": 0.7633587786259542,
-      "eval_accuracy": 0.9130434782608695,
-      "eval_loss": 0.37967830896377563,
-      "eval_runtime": 1.6211,
-      "eval_samples_per_second": 227.007,
-      "eval_steps_per_second": 28.376,
+      "eval_accuracy": 0.9375,
+      "eval_loss": 0.32088735699653625,
+      "eval_runtime": 1.5155,
+      "eval_samples_per_second": 242.823,
+      "eval_steps_per_second": 30.353,
       "step": 100
     },
     {
       "epoch": 0.8396946564885496,
-      "grad_norm": 3.2363483905792236,
+      "grad_norm": 4.234215259552002,
       "learning_rate": 0.0001583969465648855,
-      "loss": 0.3807,
+      "loss": 0.3343,
       "step": 110
     },
     {
       "epoch": 0.916030534351145,
-      "grad_norm": 2.342712879180908,
+      "grad_norm": 2.090794086456299,
       "learning_rate": 0.00015458015267175574,
-      "loss": 0.4461,
+      "loss": 0.3294,
       "step": 120
     },
     {
       "epoch": 0.9923664122137404,
-      "grad_norm": 3.4489521980285645,
+      "grad_norm": 1.4856677055358887,
       "learning_rate": 0.00015076335877862594,
-      "loss": 0.3225,
+      "loss": 0.3227,
       "step": 130
     },
     {
       "epoch": 1.0687022900763359,
-      "grad_norm": 1.5788570642471313,
+      "grad_norm": 0.4898907542228699,
       "learning_rate": 0.0001469465648854962,
-      "loss": 0.2013,
+      "loss": 0.1935,
       "step": 140
     },
     {
       "epoch": 1.1450381679389312,
-      "grad_norm": 5.834712982177734,
+      "grad_norm": 1.27410089969635,
       "learning_rate": 0.00014312977099236642,
-      "loss": 0.2111,
+      "loss": 0.2564,
       "step": 150
     },
     {
       "epoch": 1.2213740458015268,
-      "grad_norm": 0.4217844009399414,
+      "grad_norm": 1.4080382585525513,
       "learning_rate": 0.00013931297709923664,
-      "loss": 0.1577,
+      "loss": 0.1208,
       "step": 160
     },
     {
       "epoch": 1.297709923664122,
-      "grad_norm": 0.20651423931121826,
+      "grad_norm": 4.060358047485352,
       "learning_rate": 0.0001354961832061069,
-      "loss": 0.1269,
+      "loss": 0.1377,
       "step": 170
     },
     {
       "epoch": 1.3740458015267176,
-      "grad_norm": 4.183597087860107,
+      "grad_norm": 1.9385284185409546,
       "learning_rate": 0.0001316793893129771,
-      "loss": 0.1539,
+      "loss": 0.1576,
       "step": 180
     },
     {
       "epoch": 1.450381679389313,
-      "grad_norm": 3.6589787006378174,
+      "grad_norm": 1.01824152469635,
       "learning_rate": 0.00012786259541984734,
-      "loss": 0.1826,
+      "loss": 0.1459,
       "step": 190
     },
     {
       "epoch": 1.5267175572519083,
-      "grad_norm": 0.8545241951942444,
+      "grad_norm": 3.3385653495788574,
       "learning_rate": 0.00012404580152671757,
-      "loss": 0.0893,
+      "loss": 0.1859,
       "step": 200
     },
     {
       "epoch": 1.5267175572519083,
-      "eval_accuracy": 0.9565217391304348,
-      "eval_loss": 0.17774879932403564,
-      "eval_runtime": 1.3742,
-      "eval_samples_per_second": 267.782,
-      "eval_steps_per_second": 33.473,
+      "eval_accuracy": 0.970108695652174,
+      "eval_loss": 0.13519759476184845,
+      "eval_runtime": 1.5011,
+      "eval_samples_per_second": 245.146,
+      "eval_steps_per_second": 30.643,
       "step": 200
     },
     {
       "epoch": 1.6030534351145038,
-      "grad_norm": 6.76841926574707,
+      "grad_norm": 1.7121399641036987,
       "learning_rate": 0.0001202290076335878,
-      "loss": 0.1503,
+      "loss": 0.2204,
       "step": 210
     },
     {
       "epoch": 1.6793893129770994,
-      "grad_norm": 3.40751314163208,
+      "grad_norm": 0.9915915727615356,
       "learning_rate": 0.00011641221374045803,
-      "loss": 0.1792,
+      "loss": 0.1224,
       "step": 220
     },
     {
       "epoch": 1.7557251908396947,
-      "grad_norm": 0.26832714676856995,
+      "grad_norm": 0.21471183001995087,
       "learning_rate": 0.00011259541984732824,
-      "loss": 0.164,
+      "loss": 0.0638,
       "step": 230
     },
     {
       "epoch": 1.83206106870229,
-      "grad_norm": 0.2684068977832794,
+      "grad_norm": 0.16611893475055695,
       "learning_rate": 0.00010877862595419848,
-      "loss": 0.1942,
+      "loss": 0.1206,
       "step": 240
     },
     {
       "epoch": 1.9083969465648853,
-      "grad_norm": 1.6055045127868652,
+      "grad_norm": 1.0316003561019897,
       "learning_rate": 0.00010496183206106871,
-      "loss": 0.0789,
+      "loss": 0.1723,
       "step": 250
     },
     {
       "epoch": 1.984732824427481,
-      "grad_norm": 1.8824135065078735,
+      "grad_norm": 5.562504291534424,
       "learning_rate": 0.00010114503816793894,
-      "loss": 0.1053,
+      "loss": 0.1555,
       "step": 260
     },
     {
       "epoch": 2.0610687022900764,
-      "grad_norm": 0.2628338932991028,
+      "grad_norm": 0.12873513996601105,
       "learning_rate": 9.732824427480916e-05,
-      "loss": 0.0604,
+      "loss": 0.0993,
       "step": 270
     },
     {
       "epoch": 2.1374045801526718,
-      "grad_norm": 4.945899486541748,
+      "grad_norm": 0.1394704133272171,
       "learning_rate": 9.351145038167939e-05,
-      "loss": 0.0896,
+      "loss": 0.0892,
       "step": 280
     },
     {
       "epoch": 2.213740458015267,
-      "grad_norm": 0.10353419184684753,
+      "grad_norm": 0.514519989490509,
       "learning_rate": 8.969465648854962e-05,
-      "loss": 0.1232,
+      "loss": 0.061,
       "step": 290
     },
     {
       "epoch": 2.2900763358778624,
-      "grad_norm": 0.1373486965894699,
+      "grad_norm": 0.5804300904273987,
       "learning_rate": 8.587786259541986e-05,
-      "loss": 0.0733,
+      "loss": 0.0641,
       "step": 300
     },
     {
       "epoch": 2.2900763358778624,
-      "eval_accuracy": 0.9619565217391305,
-      "eval_loss": 0.1245102658867836,
-      "eval_runtime": 1.3915,
-      "eval_samples_per_second": 264.465,
-      "eval_steps_per_second": 33.058,
+      "eval_accuracy": 0.9782608695652174,
+      "eval_loss": 0.09063434600830078,
+      "eval_runtime": 1.5297,
+      "eval_samples_per_second": 240.566,
+      "eval_steps_per_second": 30.071,
       "step": 300
     },
     {
       "epoch": 2.366412213740458,
-      "grad_norm": 0.08756294846534729,
+      "grad_norm": 0.3890454173088074,
       "learning_rate": 8.206106870229007e-05,
-      "loss": 0.0652,
+      "loss": 0.1259,
       "step": 310
     },
     {
       "epoch": 2.4427480916030535,
-      "grad_norm": 0.12176764756441116,
+      "grad_norm": 1.0797548294067383,
       "learning_rate": 7.824427480916031e-05,
-      "loss": 0.0502,
+      "loss": 0.0578,
       "step": 320
     },
     {
       "epoch": 2.519083969465649,
-      "grad_norm": 0.09179558604955673,
+      "grad_norm": 3.639617443084717,
       "learning_rate": 7.442748091603053e-05,
-      "loss": 0.0686,
+      "loss": 0.0914,
       "step": 330
     },
     {
       "epoch": 2.595419847328244,
-      "grad_norm": 0.6511502861976624,
+      "grad_norm": 0.27799326181411743,
       "learning_rate": 7.061068702290077e-05,
-      "loss": 0.0435,
+      "loss": 0.0622,
       "step": 340
     },
     {
       "epoch": 2.67175572519084,
-      "grad_norm": 0.38830748200416565,
+      "grad_norm": 0.09738999605178833,
       "learning_rate": 6.6793893129771e-05,
-      "loss": 0.0416,
+      "loss": 0.0516,
       "step": 350
     },
     {
       "epoch": 2.7480916030534353,
-      "grad_norm": 0.0798097625374794,
+      "grad_norm": 0.08531022071838379,
       "learning_rate": 6.297709923664122e-05,
-      "loss": 0.1125,
+      "loss": 0.0362,
       "step": 360
     },
     {
       "epoch": 2.8244274809160306,
-      "grad_norm": 2.4349348545074463,
+      "grad_norm": 0.08216961473226547,
       "learning_rate": 5.916030534351146e-05,
-      "loss": 0.0514,
+      "loss": 0.0482,
       "step": 370
     },
     {
       "epoch": 2.900763358778626,
-      "grad_norm": 0.07260897010564804,
+      "grad_norm": 0.07119940966367722,
       "learning_rate": 5.534351145038168e-05,
-      "loss": 0.0704,
+      "loss": 0.0288,
       "step": 380
     },
     {
       "epoch": 2.9770992366412212,
-      "grad_norm": 0.08120035380125046,
+      "grad_norm": 0.06432362645864487,
       "learning_rate": 5.152671755725191e-05,
-      "loss": 0.0562,
+      "loss": 0.0445,
       "step": 390
     },
     {
       "epoch": 3.053435114503817,
-      "grad_norm": 0.06785257160663605,
+      "grad_norm": 0.09170796722173691,
       "learning_rate": 4.7709923664122144e-05,
-      "loss": 0.0384,
+      "loss": 0.0202,
       "step": 400
     },
     {
       "epoch": 3.053435114503817,
-      "eval_accuracy": 0.970108695652174,
-      "eval_loss": 0.10201143473386765,
-      "eval_runtime": 1.5213,
-      "eval_samples_per_second": 241.899,
-      "eval_steps_per_second": 30.237,
+      "eval_accuracy": 0.9755434782608695,
+      "eval_loss": 0.10268648713827133,
+      "eval_runtime": 1.5318,
+      "eval_samples_per_second": 240.238,
+      "eval_steps_per_second": 30.03,
       "step": 400
     },
     {
       "epoch": 3.1297709923664123,
-      "grad_norm": 0.6524462103843689,
+      "grad_norm": 0.5211504697799683,
       "learning_rate": 4.389312977099237e-05,
-      "loss": 0.0515,
+      "loss": 0.0262,
       "step": 410
     },
     {
       "epoch": 3.2061068702290076,
-      "grad_norm": 0.07161343842744827,
+      "grad_norm": 0.23740150034427643,
       "learning_rate": 4.00763358778626e-05,
-      "loss": 0.0377,
+      "loss": 0.0276,
       "step": 420
     },
     {
       "epoch": 3.282442748091603,
-      "grad_norm": 0.06215129420161247,
+      "grad_norm": 0.28564441204071045,
       "learning_rate": 3.625954198473282e-05,
-      "loss": 0.0259,
+      "loss": 0.0215,
       "step": 430
     },
     {
       "epoch": 3.3587786259541983,
-      "grad_norm": 0.05841224268078804,
+      "grad_norm": 0.06320058554410934,
       "learning_rate": 3.2442748091603054e-05,
-      "loss": 0.0221,
+      "loss": 0.0306,
       "step": 440
     },
     {
       "epoch": 3.435114503816794,
-      "grad_norm": 1.0506515502929688,
+      "grad_norm": 0.0660184696316719,
       "learning_rate": 2.862595419847328e-05,
-      "loss": 0.0561,
+      "loss": 0.0321,
       "step": 450
     },
     {
       "epoch": 3.5114503816793894,
-      "grad_norm": 6.415824890136719,
+      "grad_norm": 0.7349413633346558,
       "learning_rate": 2.4809160305343512e-05,
-      "loss": 0.058,
+      "loss": 0.0331,
       "step": 460
     },
     {
       "epoch": 3.5877862595419847,
-      "grad_norm": 0.056393466889858246,
+      "grad_norm": 0.057535506784915924,
       "learning_rate": 2.099236641221374e-05,
-      "loss": 0.0336,
+      "loss": 0.0188,
       "step": 470
     },
     {
       "epoch": 3.66412213740458,
-      "grad_norm": 0.2498386651277542,
+      "grad_norm": 0.4652552902698517,
       "learning_rate": 1.717557251908397e-05,
-      "loss": 0.0231,
+      "loss": 0.0179,
       "step": 480
     },
     {
       "epoch": 3.7404580152671754,
-      "grad_norm": 0.06095988303422928,
+      "grad_norm": 0.057850148528814316,
       "learning_rate": 1.3358778625954198e-05,
-      "loss": 0.0184,
+      "loss": 0.0539,
       "step": 490
     },
     {
       "epoch": 3.816793893129771,
-      "grad_norm": 0.06443587690591812,
+      "grad_norm": 1.1574023962020874,
       "learning_rate": 9.541984732824428e-06,
-      "loss": 0.0189,
+      "loss": 0.0352,
       "step": 500
     },
     {
       "epoch": 3.816793893129771,
-      "eval_accuracy": 0.9782608695652174,
-      "eval_loss": 0.08242087066173553,
-      "eval_runtime": 1.3838,
-      "eval_samples_per_second": 265.933,
-      "eval_steps_per_second": 33.242,
+      "eval_accuracy": 0.9809782608695652,
+      "eval_loss": 0.07041697204113007,
+      "eval_runtime": 1.5657,
+      "eval_samples_per_second": 235.036,
+      "eval_steps_per_second": 29.379,
       "step": 500
     },
     {
       "epoch": 3.8931297709923665,
-      "grad_norm": 0.09356739372015,
+      "grad_norm": 0.06465106457471848,
       "learning_rate": 5.725190839694657e-06,
-      "loss": 0.0363,
+      "loss": 0.0173,
       "step": 510
     },
     {
       "epoch": 3.969465648854962,
-      "grad_norm": 0.0693662166595459,
+      "grad_norm": 0.25634443759918213,
       "learning_rate": 1.908396946564886e-06,
-      "loss": 0.0379,
+      "loss": 0.0968,
       "step": 520
     },
     {
       "epoch": 4.0,
       "step": 524,
       "total_flos": 6.45382209997357e+17,
-      "train_loss": 0.26906703446657604,
-      "train_runtime": 211.2966,
-      "train_samples_per_second": 39.414,
-      "train_steps_per_second": 2.48
+      "train_loss": 0.27801662728986665,
+      "train_runtime": 166.0361,
+      "train_samples_per_second": 50.158,
+      "train_steps_per_second": 3.156
     }
   ],
   "logging_steps": 10,