Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.77,
+    "epoch": 4.62,
     "eval_accuracy": 0.632183908045977,
     "eval_loss": 1.3906450271606445,
     "eval_runtime": 0.791,
     "eval_samples_per_second": 109.994,
     "eval_steps_per_second": 2.529,
-    "total_flos": 1.6971202035873792e+17,
-    "train_loss": 1.5557504230075412,
-    "train_runtime": 77.6328,
-    "train_samples_per_second": 29.949,
-    "train_steps_per_second": 0.116
+    "total_flos": 2.799085924820828e+17,
+    "train_loss": 1.5239589055379232,
+    "train_runtime": 123.7366,
+    "train_samples_per_second": 31.317,
+    "train_steps_per_second": 0.121
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e0f886901993d0a4f33eb9dcae38498850bf94579c5925dc1b22064d7776b62
+oid sha256:c5746a6c4e01eab1616d39325dfbd1efeb5f9af88881bbe3dab04f4f5f920d33
 size 343233204

runs/Mar13_08-04-02_83e1a6899153/events.out.tfevents.1710317056.83e1a6899153.35.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8c764a13dbc364d82c1356d0720fd100f1c5cebb0f62754ad2c5bef8fb5798a
+size 5195

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.77,
-    "total_flos": 1.6971202035873792e+17,
-    "train_loss": 1.5557504230075412,
-    "train_runtime": 77.6328,
-    "train_samples_per_second": 29.949,
-    "train_steps_per_second": 0.116
+    "epoch": 4.62,
+    "total_flos": 2.799085924820828e+17,
+    "train_loss": 1.5239589055379232,
+    "train_runtime": 123.7366,
+    "train_samples_per_second": 31.317,
+    "train_steps_per_second": 0.121
 }

trainer_state.json

@@ -1,56 +1,81 @@
 {
-  "best_metric": 0.3103448275862069,
-  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetune/checkpoint-9",
-  "epoch": 2.769230769230769,
+  "best_metric": 0.22988505747126436,
+  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetune/checkpoint-13",
+  "epoch": 4.615384615384615,
   "eval_steps": 500,
-  "global_step": 9,
+  "global_step": 15,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.92,
-      "eval_accuracy": 0.26436781609195403,
-      "eval_loss": 1.6164019107818604,
-      "eval_runtime": 0.759,
-      "eval_samples_per_second": 114.622,
-      "eval_steps_per_second": 2.635,
+      "eval_accuracy": 0.09195402298850575,
+      "eval_loss": 1.6376721858978271,
+      "eval_runtime": 0.7691,
+      "eval_samples_per_second": 113.12,
+      "eval_steps_per_second": 2.6,
       "step": 3
     },
     {
       "epoch": 1.85,
-      "eval_accuracy": 0.2988505747126437,
-      "eval_loss": 1.6093627214431763,
-      "eval_runtime": 0.7415,
-      "eval_samples_per_second": 117.336,
-      "eval_steps_per_second": 2.697,
+      "eval_accuracy": 0.19540229885057472,
+      "eval_loss": 1.6207975149154663,
+      "eval_runtime": 0.7427,
+      "eval_samples_per_second": 117.142,
+      "eval_steps_per_second": 2.693,
       "step": 6
     },
     {
       "epoch": 2.77,
-      "eval_accuracy": 0.3103448275862069,
-      "eval_loss": 1.607041835784912,
-      "eval_runtime": 0.7261,
-      "eval_samples_per_second": 119.824,
-      "eval_steps_per_second": 2.755,
+      "eval_accuracy": 0.21839080459770116,
+      "eval_loss": 1.6119964122772217,
+      "eval_runtime": 0.7397,
+      "eval_samples_per_second": 117.615,
+      "eval_steps_per_second": 2.704,
       "step": 9
     },
     {
-      "epoch": 2.77,
-      "step": 9,
-      "total_flos": 1.6971202035873792e+17,
-      "train_loss": 1.5557504230075412,
-      "train_runtime": 77.6328,
-      "train_samples_per_second": 29.949,
-      "train_steps_per_second": 0.116
+      "epoch": 3.08,
+      "grad_norm": 1.20820951461792,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": 1.5539,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.22988505747126436,
+      "eval_loss": 1.605418086051941,
+      "eval_runtime": 0.7627,
+      "eval_samples_per_second": 114.068,
+      "eval_steps_per_second": 2.622,
+      "step": 13
+    },
+    {
+      "epoch": 4.62,
+      "eval_accuracy": 0.22988505747126436,
+      "eval_loss": 1.6043109893798828,
+      "eval_runtime": 0.734,
+      "eval_samples_per_second": 118.535,
+      "eval_steps_per_second": 2.725,
+      "step": 15
+    },
+    {
+      "epoch": 4.62,
+      "step": 15,
+      "total_flos": 2.799085924820828e+17,
+      "train_loss": 1.5239589055379232,
+      "train_runtime": 123.7366,
+      "train_samples_per_second": 31.317,
+      "train_steps_per_second": 0.121
     }
   ],
   "logging_steps": 10,
-  "max_steps": 9,
+  "max_steps": 15,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 1.6971202035873792e+17,
+  "total_flos": 2.799085924820828e+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:bae9dfdcd4cda7947f23be609e923a6fe7b965c7e7097d6555ee8f493dd72e15
+oid sha256:4a3b87744dff61e3a940687f89984fb7dd6df3c0df815dbab3e887efd71027ec
 size 4984