End of training

README.md

@@ -3,6 +3,8 @@ library_name: transformers
 license: apache-2.0
 base_model: timm/resnet18.a1_in1k
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 metrics:
 - accuracy
@@ -16,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-beans
 
-This model is a fine-tuned version of [timm/resnet18.a1_in1k](https://huggingface.co/timm/resnet18.a1_in1k) on an unknown dataset.
+This model is a fine-tuned version of [timm/resnet18.a1_in1k](https://huggingface.co/timm/resnet18.a1_in1k) on the beans dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6875
 - Accuracy: 0.8647

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 15.0,
-    "eval_accuracy": 0.8120300751879699,
-    "eval_loss": 0.7412300109863281,
-    "eval_runtime": 0.7631,
-    "eval_samples_per_second": 174.289,
-    "eval_steps_per_second": 22.278,
-    "total_flos": 1.5658365504595968e+17,
-    "train_loss": 0.9236146088135548,
-    "train_runtime": 142.953,
-    "train_samples_per_second": 108.497,
-    "train_steps_per_second": 13.641
+    "epoch": 20.0,
+    "eval_accuracy": 0.8646616541353384,
+    "eval_loss": 0.6874601244926453,
+    "eval_runtime": 0.7676,
+    "eval_samples_per_second": 173.27,
+    "eval_steps_per_second": 22.147,
+    "total_flos": 2.0877820672794624e+17,
+    "train_loss": 0.19350949709232038,
+    "train_runtime": 49.806,
+    "train_samples_per_second": 415.211,
+    "train_steps_per_second": 52.203
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 15.0,
-    "eval_accuracy": 0.8120300751879699,
-    "eval_loss": 0.7412300109863281,
-    "eval_runtime": 0.7631,
-    "eval_samples_per_second": 174.289,
-    "eval_steps_per_second": 22.278
+    "epoch": 20.0,
+    "eval_accuracy": 0.8646616541353384,
+    "eval_loss": 0.6874601244926453,
+    "eval_runtime": 0.7676,
+    "eval_samples_per_second": 173.27,
+    "eval_steps_per_second": 22.147
 }

runs/Nov19_22-52-51_ip-10-90-1-182/events.out.tfevents.1732056828.ip-10-90-1-182.3356820.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc3dc72508c8ab2a05a0211d3b7e188a486645e03e73915f4ba266fcee25a663
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 15.0,
-    "total_flos": 1.5658365504595968e+17,
-    "train_loss": 0.9236146088135548,
-    "train_runtime": 142.953,
-    "train_samples_per_second": 108.497,
-    "train_steps_per_second": 13.641
+    "epoch": 20.0,
+    "total_flos": 2.0877820672794624e+17,
+    "train_loss": 0.19350949709232038,
+    "train_runtime": 49.806,
+    "train_samples_per_second": 415.211,
+    "train_steps_per_second": 52.203
 }

trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.7412300109863281,
-  "best_model_checkpoint": "./beans_outputs/checkpoint-1950",
-  "epoch": 15.0,
+  "best_metric": 0.6874601244926453,
+  "best_model_checkpoint": "./beans_outputs/checkpoint-2600",
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 1950,
+  "global_step": 2600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1509,19 +1509,519 @@
       "step": 1950
     },
     {
-      "epoch": 15.0,
-      "step": 1950,
-      "total_flos": 1.5658365504595968e+17,
-      "train_loss": 0.9236146088135548,
-      "train_runtime": 142.953,
-      "train_samples_per_second": 108.497,
-      "train_steps_per_second": 13.641
+      "epoch": 15.076923076923077,
+      "grad_norm": 3.3559834957122803,
+      "learning_rate": 4.923076923076924e-06,
+      "loss": 0.7535,
+      "step": 1960
+    },
+    {
+      "epoch": 15.153846153846153,
+      "grad_norm": 1.8613739013671875,
+      "learning_rate": 4.8461538461538465e-06,
+      "loss": 0.7581,
+      "step": 1970
+    },
+    {
+      "epoch": 15.23076923076923,
+      "grad_norm": 2.3707966804504395,
+      "learning_rate": 4.76923076923077e-06,
+      "loss": 0.7836,
+      "step": 1980
+    },
+    {
+      "epoch": 15.307692307692308,
+      "grad_norm": 2.6265199184417725,
+      "learning_rate": 4.692307692307693e-06,
+      "loss": 0.8334,
+      "step": 1990
+    },
+    {
+      "epoch": 15.384615384615385,
+      "grad_norm": 2.078848123550415,
+      "learning_rate": 4.615384615384616e-06,
+      "loss": 0.7772,
+      "step": 2000
+    },
+    {
+      "epoch": 15.461538461538462,
+      "grad_norm": 2.6433162689208984,
+      "learning_rate": 4.538461538461539e-06,
+      "loss": 0.7955,
+      "step": 2010
+    },
+    {
+      "epoch": 15.538461538461538,
+      "grad_norm": 3.458962917327881,
+      "learning_rate": 4.461538461538462e-06,
+      "loss": 0.787,
+      "step": 2020
+    },
+    {
+      "epoch": 15.615384615384615,
+      "grad_norm": 5.090147495269775,
+      "learning_rate": 4.384615384615385e-06,
+      "loss": 0.7875,
+      "step": 2030
+    },
+    {
+      "epoch": 15.692307692307692,
+      "grad_norm": 1.9066407680511475,
+      "learning_rate": 4.307692307692308e-06,
+      "loss": 0.7764,
+      "step": 2040
+    },
+    {
+      "epoch": 15.76923076923077,
+      "grad_norm": 3.097341299057007,
+      "learning_rate": 4.230769230769231e-06,
+      "loss": 0.7335,
+      "step": 2050
+    },
+    {
+      "epoch": 15.846153846153847,
+      "grad_norm": 2.7201600074768066,
+      "learning_rate": 4.1538461538461545e-06,
+      "loss": 0.7747,
+      "step": 2060
+    },
+    {
+      "epoch": 15.923076923076923,
+      "grad_norm": 2.303032398223877,
+      "learning_rate": 4.076923076923077e-06,
+      "loss": 0.7738,
+      "step": 2070
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 4.420492172241211,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 0.774,
+      "step": 2080
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.849624060150376,
+      "eval_loss": 0.7370420694351196,
+      "eval_runtime": 0.776,
+      "eval_samples_per_second": 171.384,
+      "eval_steps_per_second": 21.906,
+      "step": 2080
+    },
+    {
+      "epoch": 16.076923076923077,
+      "grad_norm": 3.9969003200531006,
+      "learning_rate": 3.923076923076923e-06,
+      "loss": 0.8316,
+      "step": 2090
+    },
+    {
+      "epoch": 16.153846153846153,
+      "grad_norm": 2.3731822967529297,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 0.8162,
+      "step": 2100
+    },
+    {
+      "epoch": 16.23076923076923,
+      "grad_norm": 2.232074737548828,
+      "learning_rate": 3.7692307692307694e-06,
+      "loss": 0.8138,
+      "step": 2110
+    },
+    {
+      "epoch": 16.307692307692307,
+      "grad_norm": 2.8799118995666504,
+      "learning_rate": 3.692307692307693e-06,
+      "loss": 0.8434,
+      "step": 2120
+    },
+    {
+      "epoch": 16.384615384615383,
+      "grad_norm": 2.2093818187713623,
+      "learning_rate": 3.6153846153846156e-06,
+      "loss": 0.7886,
+      "step": 2130
+    },
+    {
+      "epoch": 16.46153846153846,
+      "grad_norm": 1.984840750694275,
+      "learning_rate": 3.538461538461539e-06,
+      "loss": 0.7682,
+      "step": 2140
+    },
+    {
+      "epoch": 16.53846153846154,
+      "grad_norm": 2.711601495742798,
+      "learning_rate": 3.4615384615384617e-06,
+      "loss": 0.7471,
+      "step": 2150
+    },
+    {
+      "epoch": 16.615384615384617,
+      "grad_norm": 2.130311965942383,
+      "learning_rate": 3.384615384615385e-06,
+      "loss": 0.7535,
+      "step": 2160
+    },
+    {
+      "epoch": 16.692307692307693,
+      "grad_norm": 2.327207565307617,
+      "learning_rate": 3.307692307692308e-06,
+      "loss": 0.718,
+      "step": 2170
+    },
+    {
+      "epoch": 16.76923076923077,
+      "grad_norm": 2.198944091796875,
+      "learning_rate": 3.2307692307692313e-06,
+      "loss": 0.8146,
+      "step": 2180
+    },
+    {
+      "epoch": 16.846153846153847,
+      "grad_norm": 2.388453483581543,
+      "learning_rate": 3.153846153846154e-06,
+      "loss": 0.8368,
+      "step": 2190
+    },
+    {
+      "epoch": 16.923076923076923,
+      "grad_norm": 2.2575690746307373,
+      "learning_rate": 3.0769230769230774e-06,
+      "loss": 0.749,
+      "step": 2200
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 6.020498275756836,
+      "learning_rate": 3e-06,
+      "loss": 0.7613,
+      "step": 2210
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.849624060150376,
+      "eval_loss": 0.7059224247932434,
+      "eval_runtime": 0.7496,
+      "eval_samples_per_second": 177.439,
+      "eval_steps_per_second": 22.68,
+      "step": 2210
+    },
+    {
+      "epoch": 17.076923076923077,
+      "grad_norm": 3.134481430053711,
+      "learning_rate": 2.9230769230769236e-06,
+      "loss": 0.7609,
+      "step": 2220
+    },
+    {
+      "epoch": 17.153846153846153,
+      "grad_norm": 2.0070559978485107,
+      "learning_rate": 2.846153846153846e-06,
+      "loss": 0.7483,
+      "step": 2230
+    },
+    {
+      "epoch": 17.23076923076923,
+      "grad_norm": 3.491682291030884,
+      "learning_rate": 2.7692307692307697e-06,
+      "loss": 0.7696,
+      "step": 2240
+    },
+    {
+      "epoch": 17.307692307692307,
+      "grad_norm": 1.9866397380828857,
+      "learning_rate": 2.6923076923076923e-06,
+      "loss": 0.7609,
+      "step": 2250
+    },
+    {
+      "epoch": 17.384615384615383,
+      "grad_norm": 3.458582878112793,
+      "learning_rate": 2.615384615384616e-06,
+      "loss": 0.7813,
+      "step": 2260
+    },
+    {
+      "epoch": 17.46153846153846,
+      "grad_norm": 2.1126835346221924,
+      "learning_rate": 2.5384615384615385e-06,
+      "loss": 0.7003,
+      "step": 2270
+    },
+    {
+      "epoch": 17.53846153846154,
+      "grad_norm": 3.5276880264282227,
+      "learning_rate": 2.461538461538462e-06,
+      "loss": 0.8305,
+      "step": 2280
+    },
+    {
+      "epoch": 17.615384615384617,
+      "grad_norm": 2.3967173099517822,
+      "learning_rate": 2.384615384615385e-06,
+      "loss": 0.7627,
+      "step": 2290
+    },
+    {
+      "epoch": 17.692307692307693,
+      "grad_norm": 4.473978042602539,
+      "learning_rate": 2.307692307692308e-06,
+      "loss": 0.7332,
+      "step": 2300
+    },
+    {
+      "epoch": 17.76923076923077,
+      "grad_norm": 2.1642568111419678,
+      "learning_rate": 2.230769230769231e-06,
+      "loss": 0.7678,
+      "step": 2310
+    },
+    {
+      "epoch": 17.846153846153847,
+      "grad_norm": 3.03192138671875,
+      "learning_rate": 2.153846153846154e-06,
+      "loss": 0.7565,
+      "step": 2320
+    },
+    {
+      "epoch": 17.923076923076923,
+      "grad_norm": 2.9610419273376465,
+      "learning_rate": 2.0769230769230773e-06,
+      "loss": 0.7651,
+      "step": 2330
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 4.160178184509277,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 0.7778,
+      "step": 2340
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.8270676691729323,
+      "eval_loss": 0.6930322647094727,
+      "eval_runtime": 0.7854,
+      "eval_samples_per_second": 169.332,
+      "eval_steps_per_second": 21.644,
+      "step": 2340
+    },
+    {
+      "epoch": 18.076923076923077,
+      "grad_norm": 2.168921947479248,
+      "learning_rate": 1.9230769230769234e-06,
+      "loss": 0.7234,
+      "step": 2350
+    },
+    {
+      "epoch": 18.153846153846153,
+      "grad_norm": 3.935608386993408,
+      "learning_rate": 1.8461538461538465e-06,
+      "loss": 0.8192,
+      "step": 2360
+    },
+    {
+      "epoch": 18.23076923076923,
+      "grad_norm": 1.8215328454971313,
+      "learning_rate": 1.7692307692307695e-06,
+      "loss": 0.7271,
+      "step": 2370
+    },
+    {
+      "epoch": 18.307692307692307,
+      "grad_norm": 2.687016010284424,
+      "learning_rate": 1.6923076923076926e-06,
+      "loss": 0.8063,
+      "step": 2380
+    },
+    {
+      "epoch": 18.384615384615383,
+      "grad_norm": 2.3364577293395996,
+      "learning_rate": 1.6153846153846157e-06,
+      "loss": 0.7699,
+      "step": 2390
+    },
+    {
+      "epoch": 18.46153846153846,
+      "grad_norm": 2.7465319633483887,
+      "learning_rate": 1.5384615384615387e-06,
+      "loss": 0.8214,
+      "step": 2400
+    },
+    {
+      "epoch": 18.53846153846154,
+      "grad_norm": 3.3499436378479004,
+      "learning_rate": 1.4615384615384618e-06,
+      "loss": 0.7432,
+      "step": 2410
+    },
+    {
+      "epoch": 18.615384615384617,
+      "grad_norm": 3.7266149520874023,
+      "learning_rate": 1.3846153846153848e-06,
+      "loss": 0.797,
+      "step": 2420
+    },
+    {
+      "epoch": 18.692307692307693,
+      "grad_norm": 2.661741256713867,
+      "learning_rate": 1.307692307692308e-06,
+      "loss": 0.7404,
+      "step": 2430
+    },
+    {
+      "epoch": 18.76923076923077,
+      "grad_norm": 3.166747808456421,
+      "learning_rate": 1.230769230769231e-06,
+      "loss": 0.8197,
+      "step": 2440
+    },
+    {
+      "epoch": 18.846153846153847,
+      "grad_norm": 3.200448989868164,
+      "learning_rate": 1.153846153846154e-06,
+      "loss": 0.8068,
+      "step": 2450
+    },
+    {
+      "epoch": 18.923076923076923,
+      "grad_norm": 2.4404191970825195,
+      "learning_rate": 1.076923076923077e-06,
+      "loss": 0.788,
+      "step": 2460
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 3.8639049530029297,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 0.8081,
+      "step": 2470
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.8646616541353384,
+      "eval_loss": 0.6890266537666321,
+      "eval_runtime": 0.7797,
+      "eval_samples_per_second": 170.576,
+      "eval_steps_per_second": 21.803,
+      "step": 2470
+    },
+    {
+      "epoch": 19.076923076923077,
+      "grad_norm": 1.7245137691497803,
+      "learning_rate": 9.230769230769232e-07,
+      "loss": 0.7929,
+      "step": 2480
+    },
+    {
+      "epoch": 19.153846153846153,
+      "grad_norm": 3.7959182262420654,
+      "learning_rate": 8.461538461538463e-07,
+      "loss": 0.7397,
+      "step": 2490
+    },
+    {
+      "epoch": 19.23076923076923,
+      "grad_norm": 2.798788070678711,
+      "learning_rate": 7.692307692307694e-07,
+      "loss": 0.7928,
+      "step": 2500
+    },
+    {
+      "epoch": 19.307692307692307,
+      "grad_norm": 2.1275336742401123,
+      "learning_rate": 6.923076923076924e-07,
+      "loss": 0.7672,
+      "step": 2510
+    },
+    {
+      "epoch": 19.384615384615383,
+      "grad_norm": 2.9216866493225098,
+      "learning_rate": 6.153846153846155e-07,
+      "loss": 0.7918,
+      "step": 2520
+    },
+    {
+      "epoch": 19.46153846153846,
+      "grad_norm": 2.3012797832489014,
+      "learning_rate": 5.384615384615386e-07,
+      "loss": 0.7418,
+      "step": 2530
+    },
+    {
+      "epoch": 19.53846153846154,
+      "grad_norm": 2.5353312492370605,
+      "learning_rate": 4.615384615384616e-07,
+      "loss": 0.8115,
+      "step": 2540
+    },
+    {
+      "epoch": 19.615384615384617,
+      "grad_norm": 3.469372510910034,
+      "learning_rate": 3.846153846153847e-07,
+      "loss": 0.7698,
+      "step": 2550
+    },
+    {
+      "epoch": 19.692307692307693,
+      "grad_norm": 2.3621013164520264,
+      "learning_rate": 3.0769230769230774e-07,
+      "loss": 0.6997,
+      "step": 2560
+    },
+    {
+      "epoch": 19.76923076923077,
+      "grad_norm": 1.7231149673461914,
+      "learning_rate": 2.307692307692308e-07,
+      "loss": 0.7207,
+      "step": 2570
+    },
+    {
+      "epoch": 19.846153846153847,
+      "grad_norm": 5.3792924880981445,
+      "learning_rate": 1.5384615384615387e-07,
+      "loss": 0.7656,
+      "step": 2580
+    },
+    {
+      "epoch": 19.923076923076923,
+      "grad_norm": 1.9618691205978394,
+      "learning_rate": 7.692307692307694e-08,
+      "loss": 0.6919,
+      "step": 2590
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 4.051193714141846,
+      "learning_rate": 0.0,
+      "loss": 0.7916,
+      "step": 2600
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.8646616541353384,
+      "eval_loss": 0.6874601244926453,
+      "eval_runtime": 0.8096,
+      "eval_samples_per_second": 164.271,
+      "eval_steps_per_second": 20.997,
+      "step": 2600
+    },
+    {
+      "epoch": 20.0,
+      "step": 2600,
+      "total_flos": 2.0877820672794624e+17,
+      "train_loss": 0.19350949709232038,
+      "train_runtime": 49.806,
+      "train_samples_per_second": 415.211,
+      "train_steps_per_second": 52.203
     }
   ],
   "logging_steps": 10,
-  "max_steps": 1950,
+  "max_steps": 2600,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 15,
+  "num_train_epochs": 20,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -1535,7 +2035,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1.5658365504595968e+17,
+  "total_flos": 2.0877820672794624e+17,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null