🍻 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: 0.0824
 - Accuracy: 0.9783

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9809782608695652,
-    "eval_loss": 0.0674990862607956,
-    "eval_runtime": 1.7056,
-    "eval_samples_per_second": 215.763,
-    "eval_steps_per_second": 26.97,
+    "eval_accuracy": 0.9782608695652174,
+    "eval_loss": 0.08242087066173553,
+    "eval_runtime": 1.5232,
+    "eval_samples_per_second": 241.595,
+    "eval_steps_per_second": 30.199,
     "total_flos": 6.45382209997357e+17,
-    "train_loss": 0.13892418100633694,
-    "train_runtime": 142.587,
-    "train_samples_per_second": 58.406,
-    "train_steps_per_second": 3.675
+    "train_loss": 0.26906703446657604,
+    "train_runtime": 211.2966,
+    "train_samples_per_second": 39.414,
+    "train_steps_per_second": 2.48
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9809782608695652,
-    "eval_loss": 0.0674990862607956,
-    "eval_runtime": 1.7056,
-    "eval_samples_per_second": 215.763,
-    "eval_steps_per_second": 26.97
+    "eval_accuracy": 0.9782608695652174,
+    "eval_loss": 0.08242087066173553,
+    "eval_runtime": 1.5232,
+    "eval_samples_per_second": 241.595,
+    "eval_steps_per_second": 30.199
 }

runs/May22_12-46-10_ef8175b0da48/events.out.tfevents.1747918461.ef8175b0da48.11741.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff1df53df78f1c48ff0609e91ae93f3386d665f6aac93793c66eaa0b41ea223c
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 6.45382209997357e+17,
-    "train_loss": 0.13892418100633694,
-    "train_runtime": 142.587,
-    "train_samples_per_second": 58.406,
-    "train_steps_per_second": 3.675
+    "train_loss": 0.26906703446657604,
+    "train_runtime": 211.2966,
+    "train_samples_per_second": 39.414,
+    "train_steps_per_second": 2.48
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_global_step": 500,
-  "best_metric": 0.0674990862607956,
+  "best_metric": 0.08242087066173553,
   "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": 8.140775680541992,
+      "grad_norm": 2.050231695175171,
       "learning_rate": 0.00019656488549618322,
-      "loss": 0.2332,
+      "loss": 1.8355,
       "step": 10
     },
     {
       "epoch": 0.15267175572519084,
-      "grad_norm": 5.731508731842041,
+      "grad_norm": 2.2453360557556152,
       "learning_rate": 0.00019274809160305345,
-      "loss": 0.2397,
+      "loss": 1.5538,
       "step": 20
     },
     {
       "epoch": 0.22900763358778625,
-      "grad_norm": 4.3719563484191895,
+      "grad_norm": 1.8408141136169434,
       "learning_rate": 0.00018893129770992367,
-      "loss": 0.2881,
+      "loss": 1.354,
       "step": 30
     },
     {
       "epoch": 0.3053435114503817,
-      "grad_norm": 2.272369623184204,
+      "grad_norm": 2.0277440547943115,
       "learning_rate": 0.0001851145038167939,
-      "loss": 0.2598,
+      "loss": 1.035,
       "step": 40
     },
     {
       "epoch": 0.3816793893129771,
-      "grad_norm": 3.4309451580047607,
+      "grad_norm": 2.367299795150757,
       "learning_rate": 0.00018129770992366412,
-      "loss": 0.1827,
+      "loss": 0.9501,
       "step": 50
     },
     {
       "epoch": 0.4580152671755725,
-      "grad_norm": 1.1197692155838013,
+      "grad_norm": 3.103821277618408,
       "learning_rate": 0.00017748091603053437,
-      "loss": 0.152,
+      "loss": 0.6932,
       "step": 60
     },
     {
       "epoch": 0.5343511450381679,
-      "grad_norm": 4.275225639343262,
+      "grad_norm": 2.1959078311920166,
       "learning_rate": 0.0001736641221374046,
-      "loss": 0.2208,
+      "loss": 0.5865,
       "step": 70
     },
     {
       "epoch": 0.6106870229007634,
-      "grad_norm": 3.2157912254333496,
+      "grad_norm": 3.5142126083374023,
       "learning_rate": 0.00016984732824427482,
-      "loss": 0.3555,
+      "loss": 0.5495,
       "step": 80
     },
     {
       "epoch": 0.6870229007633588,
-      "grad_norm": 5.242416858673096,
+      "grad_norm": 3.4135642051696777,
       "learning_rate": 0.00016603053435114505,
-      "loss": 0.163,
+      "loss": 0.5116,
       "step": 90
     },
     {
       "epoch": 0.7633587786259542,
-      "grad_norm": 0.5541224479675293,
+      "grad_norm": 1.4583014249801636,
       "learning_rate": 0.00016221374045801527,
-      "loss": 0.2052,
+      "loss": 0.4914,
       "step": 100
     },
     {
       "epoch": 0.7633587786259542,
-      "eval_accuracy": 0.9483695652173914,
-      "eval_loss": 0.1684044897556305,
-      "eval_runtime": 1.6021,
-      "eval_samples_per_second": 229.704,
-      "eval_steps_per_second": 28.713,
+      "eval_accuracy": 0.9130434782608695,
+      "eval_loss": 0.37967830896377563,
+      "eval_runtime": 1.6211,
+      "eval_samples_per_second": 227.007,
+      "eval_steps_per_second": 28.376,
       "step": 100
     },
     {
       "epoch": 0.8396946564885496,
-      "grad_norm": 0.14907874166965485,
+      "grad_norm": 3.2363483905792236,
       "learning_rate": 0.0001583969465648855,
-      "loss": 0.1426,
+      "loss": 0.3807,
       "step": 110
     },
     {
       "epoch": 0.916030534351145,
-      "grad_norm": 1.8745291233062744,
+      "grad_norm": 2.342712879180908,
       "learning_rate": 0.00015458015267175574,
-      "loss": 0.1216,
+      "loss": 0.4461,
       "step": 120
     },
     {
       "epoch": 0.9923664122137404,
-      "grad_norm": 5.384662628173828,
+      "grad_norm": 3.4489521980285645,
       "learning_rate": 0.00015076335877862594,
-      "loss": 0.17,
+      "loss": 0.3225,
       "step": 130
     },
     {
       "epoch": 1.0687022900763359,
-      "grad_norm": 4.790631294250488,
+      "grad_norm": 1.5788570642471313,
       "learning_rate": 0.0001469465648854962,
-      "loss": 0.0845,
+      "loss": 0.2013,
       "step": 140
     },
     {
       "epoch": 1.1450381679389312,
-      "grad_norm": 4.6353230476379395,
+      "grad_norm": 5.834712982177734,
       "learning_rate": 0.00014312977099236642,
-      "loss": 0.1092,
+      "loss": 0.2111,
       "step": 150
     },
     {
       "epoch": 1.2213740458015268,
-      "grad_norm": 0.8423321843147278,
+      "grad_norm": 0.4217844009399414,
       "learning_rate": 0.00013931297709923664,
-      "loss": 0.126,
+      "loss": 0.1577,
       "step": 160
     },
     {
       "epoch": 1.297709923664122,
-      "grad_norm": 0.4985540211200714,
+      "grad_norm": 0.20651423931121826,
       "learning_rate": 0.0001354961832061069,
-      "loss": 0.1103,
+      "loss": 0.1269,
       "step": 170
     },
     {
       "epoch": 1.3740458015267176,
-      "grad_norm": 4.837319850921631,
+      "grad_norm": 4.183597087860107,
       "learning_rate": 0.0001316793893129771,
-      "loss": 0.1849,
+      "loss": 0.1539,
       "step": 180
     },
     {
       "epoch": 1.450381679389313,
-      "grad_norm": 4.57224702835083,
+      "grad_norm": 3.6589787006378174,
       "learning_rate": 0.00012786259541984734,
-      "loss": 0.195,
+      "loss": 0.1826,
       "step": 190
     },
     {
       "epoch": 1.5267175572519083,
-      "grad_norm": 0.5998224020004272,
+      "grad_norm": 0.8545241951942444,
       "learning_rate": 0.00012404580152671757,
-      "loss": 0.05,
+      "loss": 0.0893,
       "step": 200
     },
     {
       "epoch": 1.5267175572519083,
-      "eval_accuracy": 0.9592391304347826,
-      "eval_loss": 0.16003353893756866,
-      "eval_runtime": 1.5924,
-      "eval_samples_per_second": 231.101,
-      "eval_steps_per_second": 28.888,
+      "eval_accuracy": 0.9565217391304348,
+      "eval_loss": 0.17774879932403564,
+      "eval_runtime": 1.3742,
+      "eval_samples_per_second": 267.782,
+      "eval_steps_per_second": 33.473,
       "step": 200
     },
     {
       "epoch": 1.6030534351145038,
-      "grad_norm": 2.8493752479553223,
+      "grad_norm": 6.76841926574707,
       "learning_rate": 0.0001202290076335878,
-      "loss": 0.3001,
+      "loss": 0.1503,
       "step": 210
     },
     {
       "epoch": 1.6793893129770994,
-      "grad_norm": 0.12870310246944427,
+      "grad_norm": 3.40751314163208,
       "learning_rate": 0.00011641221374045803,
-      "loss": 0.0944,
+      "loss": 0.1792,
       "step": 220
     },
     {
       "epoch": 1.7557251908396947,
-      "grad_norm": 6.482563018798828,
+      "grad_norm": 0.26832714676856995,
       "learning_rate": 0.00011259541984732824,
-      "loss": 0.0905,
+      "loss": 0.164,
       "step": 230
     },
     {
       "epoch": 1.83206106870229,
-      "grad_norm": 1.615530252456665,
+      "grad_norm": 0.2684068977832794,
       "learning_rate": 0.00010877862595419848,
-      "loss": 0.1326,
+      "loss": 0.1942,
       "step": 240
     },
     {
       "epoch": 1.9083969465648853,
-      "grad_norm": 2.059842586517334,
+      "grad_norm": 1.6055045127868652,
       "learning_rate": 0.00010496183206106871,
-      "loss": 0.1274,
+      "loss": 0.0789,
       "step": 250
     },
     {
       "epoch": 1.984732824427481,
-      "grad_norm": 2.750730276107788,
+      "grad_norm": 1.8824135065078735,
       "learning_rate": 0.00010114503816793894,
-      "loss": 0.2225,
+      "loss": 0.1053,
       "step": 260
     },
     {
       "epoch": 2.0610687022900764,
-      "grad_norm": 0.04552418366074562,
+      "grad_norm": 0.2628338932991028,
       "learning_rate": 9.732824427480916e-05,
-      "loss": 0.0418,
+      "loss": 0.0604,
       "step": 270
     },
     {
       "epoch": 2.1374045801526718,
-      "grad_norm": 3.4131999015808105,
+      "grad_norm": 4.945899486541748,
       "learning_rate": 9.351145038167939e-05,
-      "loss": 0.1439,
+      "loss": 0.0896,
       "step": 280
     },
     {
       "epoch": 2.213740458015267,
-      "grad_norm": 1.8281826972961426,
+      "grad_norm": 0.10353419184684753,
       "learning_rate": 8.969465648854962e-05,
-      "loss": 0.1084,
+      "loss": 0.1232,
       "step": 290
     },
     {
       "epoch": 2.2900763358778624,
-      "grad_norm": 0.03664281591773033,
+      "grad_norm": 0.1373486965894699,
       "learning_rate": 8.587786259541986e-05,
-      "loss": 0.0579,
+      "loss": 0.0733,
       "step": 300
     },
     {
       "epoch": 2.2900763358778624,
-      "eval_accuracy": 0.9510869565217391,
-      "eval_loss": 0.161894753575325,
-      "eval_runtime": 1.5917,
-      "eval_samples_per_second": 231.199,
-      "eval_steps_per_second": 28.9,
+      "eval_accuracy": 0.9619565217391305,
+      "eval_loss": 0.1245102658867836,
+      "eval_runtime": 1.3915,
+      "eval_samples_per_second": 264.465,
+      "eval_steps_per_second": 33.058,
       "step": 300
     },
     {
       "epoch": 2.366412213740458,
-      "grad_norm": 10.416833877563477,
+      "grad_norm": 0.08756294846534729,
       "learning_rate": 8.206106870229007e-05,
-      "loss": 0.1502,
+      "loss": 0.0652,
       "step": 310
     },
     {
       "epoch": 2.4427480916030535,
-      "grad_norm": 2.426678419113159,
+      "grad_norm": 0.12176764756441116,
       "learning_rate": 7.824427480916031e-05,
-      "loss": 0.2334,
+      "loss": 0.0502,
       "step": 320
     },
     {
       "epoch": 2.519083969465649,
-      "grad_norm": 5.191472053527832,
+      "grad_norm": 0.09179558604955673,
       "learning_rate": 7.442748091603053e-05,
-      "loss": 0.0829,
+      "loss": 0.0686,
       "step": 330
     },
     {
       "epoch": 2.595419847328244,
-      "grad_norm": 2.1241517066955566,
+      "grad_norm": 0.6511502861976624,
       "learning_rate": 7.061068702290077e-05,
-      "loss": 0.066,
+      "loss": 0.0435,
       "step": 340
     },
     {
       "epoch": 2.67175572519084,
-      "grad_norm": 0.10309349745512009,
+      "grad_norm": 0.38830748200416565,
       "learning_rate": 6.6793893129771e-05,
-      "loss": 0.1356,
+      "loss": 0.0416,
       "step": 350
     },
     {
       "epoch": 2.7480916030534353,
-      "grad_norm": 0.8078840374946594,
+      "grad_norm": 0.0798097625374794,
       "learning_rate": 6.297709923664122e-05,
-      "loss": 0.09,
+      "loss": 0.1125,
       "step": 360
     },
     {
       "epoch": 2.8244274809160306,
-      "grad_norm": 0.7653095126152039,
+      "grad_norm": 2.4349348545074463,
       "learning_rate": 5.916030534351146e-05,
-      "loss": 0.0678,
+      "loss": 0.0514,
       "step": 370
     },
     {
       "epoch": 2.900763358778626,
-      "grad_norm": 0.1392490267753601,
+      "grad_norm": 0.07260897010564804,
       "learning_rate": 5.534351145038168e-05,
-      "loss": 0.0532,
+      "loss": 0.0704,
       "step": 380
     },
     {
       "epoch": 2.9770992366412212,
-      "grad_norm": 7.884419918060303,
+      "grad_norm": 0.08120035380125046,
       "learning_rate": 5.152671755725191e-05,
-      "loss": 0.1382,
+      "loss": 0.0562,
       "step": 390
     },
     {
       "epoch": 3.053435114503817,
-      "grad_norm": 10.337115287780762,
+      "grad_norm": 0.06785257160663605,
       "learning_rate": 4.7709923664122144e-05,
-      "loss": 0.1181,
+      "loss": 0.0384,
       "step": 400
     },
     {
       "epoch": 3.053435114503817,
-      "eval_accuracy": 0.9809782608695652,
-      "eval_loss": 0.06854937225580215,
-      "eval_runtime": 2.5796,
-      "eval_samples_per_second": 142.659,
-      "eval_steps_per_second": 17.832,
+      "eval_accuracy": 0.970108695652174,
+      "eval_loss": 0.10201143473386765,
+      "eval_runtime": 1.5213,
+      "eval_samples_per_second": 241.899,
+      "eval_steps_per_second": 30.237,
       "step": 400
     },
     {
       "epoch": 3.1297709923664123,
-      "grad_norm": 0.06403453648090363,
+      "grad_norm": 0.6524462103843689,
       "learning_rate": 4.389312977099237e-05,
-      "loss": 0.1245,
+      "loss": 0.0515,
       "step": 410
     },
     {
       "epoch": 3.2061068702290076,
-      "grad_norm": 3.0858404636383057,
+      "grad_norm": 0.07161343842744827,
       "learning_rate": 4.00763358778626e-05,
-      "loss": 0.0901,
+      "loss": 0.0377,
       "step": 420
     },
     {
       "epoch": 3.282442748091603,
-      "grad_norm": 1.1947516202926636,
+      "grad_norm": 0.06215129420161247,
       "learning_rate": 3.625954198473282e-05,
-      "loss": 0.1319,
+      "loss": 0.0259,
       "step": 430
     },
     {
       "epoch": 3.3587786259541983,
-      "grad_norm": 2.4845542907714844,
+      "grad_norm": 0.05841224268078804,
       "learning_rate": 3.2442748091603054e-05,
-      "loss": 0.0932,
+      "loss": 0.0221,
       "step": 440
     },
     {
       "epoch": 3.435114503816794,
-      "grad_norm": 0.7602748870849609,
+      "grad_norm": 1.0506515502929688,
       "learning_rate": 2.862595419847328e-05,
-      "loss": 0.0749,
+      "loss": 0.0561,
       "step": 450
     },
     {
       "epoch": 3.5114503816793894,
-      "grad_norm": 3.669577121734619,
+      "grad_norm": 6.415824890136719,
       "learning_rate": 2.4809160305343512e-05,
-      "loss": 0.1067,
+      "loss": 0.058,
       "step": 460
     },
     {
       "epoch": 3.5877862595419847,
-      "grad_norm": 0.02125421166419983,
+      "grad_norm": 0.056393466889858246,
       "learning_rate": 2.099236641221374e-05,
-      "loss": 0.0278,
+      "loss": 0.0336,
       "step": 470
     },
     {
       "epoch": 3.66412213740458,
-      "grad_norm": 0.025344278663396835,
+      "grad_norm": 0.2498386651277542,
       "learning_rate": 1.717557251908397e-05,
-      "loss": 0.0717,
+      "loss": 0.0231,
       "step": 480
     },
     {
       "epoch": 3.7404580152671754,
-      "grad_norm": 0.454380601644516,
+      "grad_norm": 0.06095988303422928,
       "learning_rate": 1.3358778625954198e-05,
-      "loss": 0.1241,
+      "loss": 0.0184,
       "step": 490
     },
     {
       "epoch": 3.816793893129771,
-      "grad_norm": 0.49031591415405273,
+      "grad_norm": 0.06443587690591812,
       "learning_rate": 9.541984732824428e-06,
-      "loss": 0.1321,
+      "loss": 0.0189,
       "step": 500
     },
     {
       "epoch": 3.816793893129771,
-      "eval_accuracy": 0.9809782608695652,
-      "eval_loss": 0.0674990862607956,
-      "eval_runtime": 1.6192,
-      "eval_samples_per_second": 227.276,
-      "eval_steps_per_second": 28.409,
+      "eval_accuracy": 0.9782608695652174,
+      "eval_loss": 0.08242087066173553,
+      "eval_runtime": 1.3838,
+      "eval_samples_per_second": 265.933,
+      "eval_steps_per_second": 33.242,
       "step": 500
     },
     {
       "epoch": 3.8931297709923665,
-      "grad_norm": 7.950806617736816,
+      "grad_norm": 0.09356739372015,
       "learning_rate": 5.725190839694657e-06,
-      "loss": 0.1726,
+      "loss": 0.0363,
       "step": 510
     },
     {
       "epoch": 3.969465648854962,
-      "grad_norm": 0.29777708649635315,
+      "grad_norm": 0.0693662166595459,
       "learning_rate": 1.908396946564886e-06,
-      "loss": 0.0653,
+      "loss": 0.0379,
       "step": 520
     },
     {
       "epoch": 4.0,
       "step": 524,
       "total_flos": 6.45382209997357e+17,
-      "train_loss": 0.13892418100633694,
-      "train_runtime": 142.587,
-      "train_samples_per_second": 58.406,
-      "train_steps_per_second": 3.675
+      "train_loss": 0.26906703446657604,
+      "train_runtime": 211.2966,
+      "train_samples_per_second": 39.414,
+      "train_steps_per_second": 2.48
     }
   ],
   "logging_steps": 10,