Training in progress, epoch 0

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.5316455696202531,
-    "eval_loss": 0.6931473612785339,
-    "eval_runtime": 4.6123,
-    "eval_samples_per_second": 17.128,
-    "eval_steps_per_second": 0.65,
-    "total_flos": 2.898785558237184e+16,
-    "train_loss": 0.6919973691304525,
-    "train_runtime": 76.6563,
-    "train_samples_per_second": 7.24,
-    "train_steps_per_second": 0.039
+    "epoch": 5.0,
+    "eval_accuracy": 0.6075949367088608,
+    "eval_loss": 0.6998201608657837,
+    "eval_runtime": 4.8772,
+    "eval_samples_per_second": 16.198,
+    "eval_steps_per_second": 0.615,
+    "total_flos": 7.24696389559296e+16,
+    "train_loss": 0.6189225832621257,
+    "train_runtime": 239.3324,
+    "train_samples_per_second": 3.865,
+    "train_steps_per_second": 0.063
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "microsoft/swin-base-patch4-window7-224-in22k",
+  "_name_or_path": "microsoft/swin-base-patch4-window7-224",
   "architectures": [
     "SwinForImageClassification"
   ],

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.5316455696202531,
-    "eval_loss": 0.6931473612785339,
-    "eval_runtime": 4.6123,
-    "eval_samples_per_second": 17.128,
-    "eval_steps_per_second": 0.65
+    "epoch": 5.0,
+    "eval_accuracy": 0.6075949367088608,
+    "eval_loss": 0.6998201608657837,
+    "eval_runtime": 4.8772,
+    "eval_samples_per_second": 16.198,
+    "eval_steps_per_second": 0.615
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:363abce521002a977839e8351f62ccd2f073f9b8cf1d343ccff6f7c2289df88d
+oid sha256:df8adc4c51d350d67a55f0839f29ccdc4aa6512435637b5049feecea36a34232
 size 347498816

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 2.898785558237184e+16,
-    "train_loss": 0.6919973691304525,
-    "train_runtime": 76.6563,
-    "train_samples_per_second": 7.24,
-    "train_steps_per_second": 0.039
+    "epoch": 5.0,
+    "total_flos": 7.24696389559296e+16,
+    "train_loss": 0.6189225832621257,
+    "train_runtime": 239.3324,
+    "train_samples_per_second": 3.865,
+    "train_steps_per_second": 0.063
 }

trainer_state.json

@@ -1,47 +1,80 @@
 {
-  "best_metric": 0.5316455696202531,
-  "best_model_checkpoint": "swin-finetuned-food101\\checkpoint-1",
-  "epoch": 2.0,
+  "best_metric": 0.6075949367088608,
+  "best_model_checkpoint": "swin-finetuned-food101\\checkpoint-12",
+  "epoch": 5.0,
   "eval_steps": 500,
-  "global_step": 3,
+  "global_step": 15,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.67,
-      "eval_accuracy": 0.5316455696202531,
-      "eval_loss": 0.6931473612785339,
-      "eval_runtime": 5.5884,
-      "eval_samples_per_second": 14.136,
-      "eval_steps_per_second": 0.537,
-      "step": 1
-    },
-    {
-      "epoch": 2.0,
+      "epoch": 1.0,
       "eval_accuracy": 0.46835443037974683,
-      "eval_loss": 0.6949594020843506,
-      "eval_runtime": 3.9224,
-      "eval_samples_per_second": 20.141,
-      "eval_steps_per_second": 0.765,
+      "eval_loss": 0.709945797920227,
+      "eval_runtime": 4.7061,
+      "eval_samples_per_second": 16.787,
+      "eval_steps_per_second": 0.637,
       "step": 3
     },
     {
       "epoch": 2.0,
-      "step": 3,
-      "total_flos": 2.898785558237184e+16,
-      "train_loss": 0.6919973691304525,
-      "train_runtime": 76.6563,
-      "train_samples_per_second": 7.24,
-      "train_steps_per_second": 0.039
+      "eval_accuracy": 0.569620253164557,
+      "eval_loss": 0.7113380432128906,
+      "eval_runtime": 5.1386,
+      "eval_samples_per_second": 15.374,
+      "eval_steps_per_second": 0.584,
+      "step": 6
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.4810126582278481,
+      "eval_loss": 0.8271049857139587,
+      "eval_runtime": 4.3712,
+      "eval_samples_per_second": 18.073,
+      "eval_steps_per_second": 0.686,
+      "step": 9
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 7.142857142857143e-05,
+      "loss": 0.6561,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6075949367088608,
+      "eval_loss": 0.6998201608657837,
+      "eval_runtime": 4.5439,
+      "eval_samples_per_second": 17.386,
+      "eval_steps_per_second": 0.66,
+      "step": 12
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5822784810126582,
+      "eval_loss": 0.8232901096343994,
+      "eval_runtime": 6.1137,
+      "eval_samples_per_second": 12.922,
+      "eval_steps_per_second": 0.491,
+      "step": 15
+    },
+    {
+      "epoch": 5.0,
+      "step": 15,
+      "total_flos": 7.24696389559296e+16,
+      "train_loss": 0.6189225832621257,
+      "train_runtime": 239.3324,
+      "train_samples_per_second": 3.865,
+      "train_steps_per_second": 0.063
     }
   ],
   "logging_steps": 10,
-  "max_steps": 3,
+  "max_steps": 15,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 2.898785558237184e+16,
+  "total_flos": 7.24696389559296e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33a0d34bffcd583aa49a3ad059ebf6d6eeeb3d75e3117868411d5c74be66b67f
+oid sha256:35bc3aedffcd97a81146ac1708f5ec50be895ef7811e6fdf6a464670897316c1
 size 4728