Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 16.0,
+    "epoch": 2.77,
     "eval_accuracy": 0.6666666666666666,
     "eval_loss": 1.2180020809173584,
     "eval_runtime": 1.2421,
     "eval_samples_per_second": 70.045,
     "eval_steps_per_second": 2.415,
-    "total_flos": 9.609265079672832e+17,
-    "train_loss": 0.6730460596084594,
-    "train_runtime": 461.0529,
-    "train_samples_per_second": 27.763,
-    "train_steps_per_second": 0.217
+    "total_flos": 1.6971202035873792e+17,
+    "train_loss": 0.5219763120015463,
+    "train_runtime": 81.1004,
+    "train_samples_per_second": 28.668,
+    "train_steps_per_second": 0.111
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe78d8e5e8a5ef1fd3d07d22474d8cdf28a614904289f4c6dcf3af93b68e0388
+oid sha256:b46a90adc2bd7ed767376aeecdd7a4a416ec698cc199c5c81a4b67c711faf988
 size 343233204

runs/Mar13_07-29-11_be0be0d60671/events.out.tfevents.1710314968.be0be0d60671.34.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b888a062b27ebf4f978f93c58d4c9f5fe0e4b7c909a8d69f1338eff32688e1b7
+size 5235

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 16.0,
-    "total_flos": 9.609265079672832e+17,
-    "train_loss": 0.6730460596084594,
-    "train_runtime": 461.0529,
-    "train_samples_per_second": 27.763,
-    "train_steps_per_second": 0.217
+    "epoch": 2.77,
+    "total_flos": 1.6971202035873792e+17,
+    "train_loss": 0.5219763120015463,
+    "train_runtime": 81.1004,
+    "train_samples_per_second": 28.668,
+    "train_steps_per_second": 0.111
 }

trainer_state.json

@@ -1,244 +1,57 @@
 {
-  "best_metric": 1.2180020809173584,
-  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetunedmangodisease/checkpoint-87",
-  "epoch": 16.0,
+  "best_metric": 0.6666666666666666,
+  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetunedmangodisease/checkpoint-9",
+  "epoch": 2.769230769230769,
   "eval_steps": 500,
-  "global_step": 100,
+  "global_step": 9,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.96,
-      "eval_accuracy": 0.6551724137931034,
-      "eval_loss": 1.3800822496414185,
-      "eval_runtime": 0.8949,
-      "eval_samples_per_second": 97.223,
-      "eval_steps_per_second": 3.353,
-      "step": 6
-    },
-    {
-      "epoch": 1.6,
-      "grad_norm": 1.6758902072906494,
-      "learning_rate": 5e-05,
-      "loss": 0.8237,
-      "step": 10
-    },
-    {
-      "epoch": 1.92,
-      "eval_accuracy": 0.6781609195402298,
-      "eval_loss": 1.349981665611267,
-      "eval_runtime": 0.8658,
-      "eval_samples_per_second": 100.488,
-      "eval_steps_per_second": 3.465,
-      "step": 12
-    },
-    {
-      "epoch": 2.88,
-      "eval_accuracy": 0.6896551724137931,
-      "eval_loss": 1.3429534435272217,
-      "eval_runtime": 0.8664,
-      "eval_samples_per_second": 100.415,
-      "eval_steps_per_second": 3.463,
-      "step": 18
-    },
-    {
-      "epoch": 3.2,
-      "grad_norm": 1.6055872440338135,
-      "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.7777,
-      "step": 20
-    },
-    {
-      "epoch": 4.0,
-      "eval_accuracy": 0.6781609195402298,
-      "eval_loss": 1.3140147924423218,
-      "eval_runtime": 1.2457,
-      "eval_samples_per_second": 69.841,
-      "eval_steps_per_second": 2.408,
-      "step": 25
-    },
-    {
-      "epoch": 4.8,
-      "grad_norm": 1.475545048713684,
-      "learning_rate": 3.888888888888889e-05,
-      "loss": 0.7066,
-      "step": 30
+      "epoch": 0.92,
+      "eval_accuracy": 0.6091954022988506,
+      "eval_loss": 1.2354542016983032,
+      "eval_runtime": 0.8662,
+      "eval_samples_per_second": 100.435,
+      "eval_steps_per_second": 2.309,
+      "step": 3
     },
     {
-      "epoch": 4.96,
-      "eval_accuracy": 0.6896551724137931,
-      "eval_loss": 1.2853386402130127,
-      "eval_runtime": 0.8536,
-      "eval_samples_per_second": 101.919,
-      "eval_steps_per_second": 3.514,
-      "step": 31
-    },
-    {
-      "epoch": 5.92,
-      "eval_accuracy": 0.6781609195402298,
-      "eval_loss": 1.2957433462142944,
-      "eval_runtime": 0.8698,
-      "eval_samples_per_second": 100.02,
-      "eval_steps_per_second": 3.449,
-      "step": 37
-    },
-    {
-      "epoch": 6.4,
-      "grad_norm": 1.3556888103485107,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.6696,
-      "step": 40
-    },
-    {
-      "epoch": 6.88,
-      "eval_accuracy": 0.6551724137931034,
-      "eval_loss": 1.2818764448165894,
-      "eval_runtime": 0.8627,
-      "eval_samples_per_second": 100.841,
-      "eval_steps_per_second": 3.477,
-      "step": 43
-    },
-    {
-      "epoch": 8.0,
-      "grad_norm": 4.025577545166016,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.673,
-      "step": 50
-    },
-    {
-      "epoch": 8.0,
-      "eval_accuracy": 0.6781609195402298,
-      "eval_loss": 1.285282850265503,
-      "eval_runtime": 1.3346,
-      "eval_samples_per_second": 65.186,
-      "eval_steps_per_second": 2.248,
-      "step": 50
-    },
-    {
-      "epoch": 8.96,
+      "epoch": 1.85,
       "eval_accuracy": 0.6551724137931034,
-      "eval_loss": 1.2612943649291992,
-      "eval_runtime": 1.3828,
-      "eval_samples_per_second": 62.915,
-      "eval_steps_per_second": 2.169,
-      "step": 56
-    },
-    {
-      "epoch": 9.6,
-      "grad_norm": 1.7103294134140015,
-      "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.6311,
-      "step": 60
-    },
-    {
-      "epoch": 9.92,
-      "eval_accuracy": 0.6436781609195402,
-      "eval_loss": 1.2490617036819458,
-      "eval_runtime": 1.1569,
-      "eval_samples_per_second": 75.2,
-      "eval_steps_per_second": 2.593,
-      "step": 62
-    },
-    {
-      "epoch": 10.88,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.2408325672149658,
-      "eval_runtime": 1.7627,
-      "eval_samples_per_second": 49.356,
-      "eval_steps_per_second": 1.702,
-      "step": 68
-    },
-    {
-      "epoch": 11.2,
-      "grad_norm": 1.8741282224655151,
-      "learning_rate": 1.6666666666666667e-05,
-      "loss": 0.6086,
-      "step": 70
-    },
-    {
-      "epoch": 12.0,
-      "eval_accuracy": 0.632183908045977,
-      "eval_loss": 1.2268433570861816,
-      "eval_runtime": 1.4569,
-      "eval_samples_per_second": 59.717,
-      "eval_steps_per_second": 2.059,
-      "step": 75
-    },
-    {
-      "epoch": 12.8,
-      "grad_norm": 2.322943925857544,
-      "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.6374,
-      "step": 80
-    },
-    {
-      "epoch": 12.96,
-      "eval_accuracy": 0.632183908045977,
-      "eval_loss": 1.2218347787857056,
-      "eval_runtime": 0.8909,
-      "eval_samples_per_second": 97.657,
-      "eval_steps_per_second": 3.367,
-      "step": 81
-    },
-    {
-      "epoch": 13.92,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.2180020809173584,
-      "eval_runtime": 0.8761,
-      "eval_samples_per_second": 99.299,
-      "eval_steps_per_second": 3.424,
-      "step": 87
-    },
-    {
-      "epoch": 14.4,
-      "grad_norm": 1.7907931804656982,
-      "learning_rate": 5.555555555555556e-06,
-      "loss": 0.599,
-      "step": 90
-    },
-    {
-      "epoch": 14.88,
-      "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.2180161476135254,
-      "eval_runtime": 1.267,
-      "eval_samples_per_second": 68.668,
-      "eval_steps_per_second": 2.368,
-      "step": 93
-    },
-    {
-      "epoch": 16.0,
-      "grad_norm": 3.0324525833129883,
-      "learning_rate": 0.0,
-      "loss": 0.6037,
-      "step": 100
+      "eval_loss": 1.2132805585861206,
+      "eval_runtime": 0.8407,
+      "eval_samples_per_second": 103.485,
+      "eval_steps_per_second": 2.379,
+      "step": 6
     },
     {
-      "epoch": 16.0,
+      "epoch": 2.77,
       "eval_accuracy": 0.6666666666666666,
-      "eval_loss": 1.219612717628479,
-      "eval_runtime": 0.8638,
-      "eval_samples_per_second": 100.717,
-      "eval_steps_per_second": 3.473,
-      "step": 100
-    },
-    {
-      "epoch": 16.0,
-      "step": 100,
-      "total_flos": 9.609265079672832e+17,
-      "train_loss": 0.6730460596084594,
-      "train_runtime": 461.0529,
-      "train_samples_per_second": 27.763,
-      "train_steps_per_second": 0.217
+      "eval_loss": 1.2124366760253906,
+      "eval_runtime": 0.8725,
+      "eval_samples_per_second": 99.71,
+      "eval_steps_per_second": 2.292,
+      "step": 9
+    },
+    {
+      "epoch": 2.77,
+      "step": 9,
+      "total_flos": 1.6971202035873792e+17,
+      "train_loss": 0.5219763120015463,
+      "train_runtime": 81.1004,
+      "train_samples_per_second": 28.668,
+      "train_steps_per_second": 0.111
     }
   ],
   "logging_steps": 10,
-  "max_steps": 100,
+  "max_steps": 9,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 17,
+  "num_train_epochs": 3,
   "save_steps": 500,
-  "total_flos": 9.609265079672832e+17,
-  "train_batch_size": 32,
+  "total_flos": 1.6971202035873792e+17,
+  "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fcec20e8d1d9a83e6122676b0bb12b093be8b940a355d6ffc12e6ec4ef3fac6
+oid sha256:19977306281ec5975f4a5592337d380cadb23a0310c5469b74c8574e55d27625
 size 4984