End of training

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 1.0
+      value: 0.999
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Marqo/nsfw-image-detection-384](https://huggingface.co/Marqo/nsfw-image-detection-384) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0003
-- Accuracy: 1.0
+- Loss: 0.0041
+- Accuracy: 0.999
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.996,
-    "eval_loss": 0.0253806933760643,
-    "eval_runtime": 28.4966,
-    "eval_samples_per_second": 35.092,
-    "eval_steps_per_second": 4.386,
-    "total_flos": 7.036273897136456e+17,
-    "train_loss": 0.009760016079127729,
-    "train_runtime": 1762.5795,
-    "train_samples_per_second": 26.869,
-    "train_steps_per_second": 1.68
+    "epoch": 1.5005065856129685,
+    "eval_accuracy": 0.999,
+    "eval_loss": 0.004140933509916067,
+    "eval_runtime": 28.4475,
+    "eval_samples_per_second": 35.152,
+    "eval_steps_per_second": 2.215,
+    "total_flos": 3.5197712872833024e+17,
+    "train_loss": 0.002488548292009352,
+    "train_runtime": 874.5228,
+    "train_samples_per_second": 27.076,
+    "train_steps_per_second": 1.693
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.996,
-    "eval_loss": 0.0253806933760643,
-    "eval_runtime": 28.4966,
-    "eval_samples_per_second": 35.092,
-    "eval_steps_per_second": 4.386
+    "epoch": 1.5005065856129685,
+    "eval_accuracy": 0.999,
+    "eval_loss": 0.004140933509916067,
+    "eval_runtime": 28.4475,
+    "eval_samples_per_second": 35.152,
+    "eval_steps_per_second": 2.215
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 7.036273897136456e+17,
-    "train_loss": 0.009760016079127729,
-    "train_runtime": 1762.5795,
-    "train_samples_per_second": 26.869,
-    "train_steps_per_second": 1.68
+    "epoch": 1.5005065856129685,
+    "total_flos": 3.5197712872833024e+17,
+    "train_loss": 0.002488548292009352,
+    "train_runtime": 874.5228,
+    "train_samples_per_second": 27.076,
+    "train_steps_per_second": 1.693
 }

trainer_state.json

@@ -1,108 +1,60 @@
 {
-  "best_metric": 1.5119968338694889e-05,
-  "best_model_checkpoint": "./nsfw-fine-tuned-timm/checkpoint-2000",
-  "epoch": 3.0,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.5005065856129685,
   "eval_steps": 500,
-  "global_step": 2961,
+  "global_step": 1481,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.5065856129685917,
-      "grad_norm": 0.025366274639964104,
-      "learning_rate": 4.1556906450523474e-05,
-      "loss": 0.0375,
+      "grad_norm": 0.000192928608157672,
+      "learning_rate": 1.676459191498087e-05,
+      "loss": 0.0041,
       "step": 500
     },
     {
       "epoch": 0.5065856129685917,
-      "eval_accuracy": 0.998,
-      "eval_loss": 0.0066848029382526875,
-      "eval_runtime": 27.7412,
-      "eval_samples_per_second": 36.047,
-      "eval_steps_per_second": 4.506,
+      "eval_accuracy": 0.996,
+      "eval_loss": 0.014655331149697304,
+      "eval_runtime": 27.4552,
+      "eval_samples_per_second": 36.423,
+      "eval_steps_per_second": 2.295,
       "step": 500
     },
     {
       "epoch": 1.0131712259371835,
-      "grad_norm": 0.008881847374141216,
-      "learning_rate": 3.3113812901046945e-05,
-      "loss": 0.0138,
+      "grad_norm": 0.00017229605873581022,
+      "learning_rate": 5.773817382593008e-06,
+      "loss": 0.0023,
       "step": 1000
     },
     {
       "epoch": 1.0131712259371835,
       "eval_accuracy": 1.0,
-      "eval_loss": 0.00014481571270152926,
-      "eval_runtime": 27.3526,
-      "eval_samples_per_second": 36.56,
-      "eval_steps_per_second": 4.57,
+      "eval_loss": 0.00030512735247612,
+      "eval_runtime": 27.4106,
+      "eval_samples_per_second": 36.482,
+      "eval_steps_per_second": 2.298,
       "step": 1000
     },
     {
-      "epoch": 1.5197568389057752,
-      "grad_norm": 0.004049438983201981,
-      "learning_rate": 2.4670719351570416e-05,
-      "loss": 0.0028,
-      "step": 1500
-    },
-    {
-      "epoch": 1.5197568389057752,
-      "eval_accuracy": 1.0,
-      "eval_loss": 9.533474803902209e-05,
-      "eval_runtime": 27.4005,
-      "eval_samples_per_second": 36.496,
-      "eval_steps_per_second": 4.562,
-      "step": 1500
-    },
-    {
-      "epoch": 2.026342451874367,
-      "grad_norm": 0.0020459946244955063,
-      "learning_rate": 1.6227625802093887e-05,
-      "loss": 0.002,
-      "step": 2000
-    },
-    {
-      "epoch": 2.026342451874367,
-      "eval_accuracy": 1.0,
-      "eval_loss": 1.5119968338694889e-05,
-      "eval_runtime": 27.7767,
-      "eval_samples_per_second": 36.001,
-      "eval_steps_per_second": 4.5,
-      "step": 2000
-    },
-    {
-      "epoch": 2.5329280648429586,
-      "grad_norm": 0.025027019903063774,
-      "learning_rate": 7.784532252617359e-06,
-      "loss": 0.0013,
-      "step": 2500
-    },
-    {
-      "epoch": 2.5329280648429586,
-      "eval_accuracy": 1.0,
-      "eval_loss": 2.4935185138019733e-05,
-      "eval_runtime": 27.3766,
-      "eval_samples_per_second": 36.528,
-      "eval_steps_per_second": 4.566,
-      "step": 2500
-    },
-    {
-      "epoch": 3.0,
-      "step": 2961,
-      "total_flos": 7.036273897136456e+17,
-      "train_loss": 0.009760016079127729,
-      "train_runtime": 1762.5795,
-      "train_samples_per_second": 26.869,
-      "train_steps_per_second": 1.68
+      "epoch": 1.5005065856129685,
+      "step": 1481,
+      "total_flos": 3.5197712872833024e+17,
+      "train_loss": 0.002488548292009352,
+      "train_runtime": 874.5228,
+      "train_samples_per_second": 27.076,
+      "train_steps_per_second": 1.693
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2961,
+  "max_steps": 1481,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
-  "save_steps": 1000,
+  "num_train_epochs": 2,
+  "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -115,7 +67,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 7.036273897136456e+17,
+  "total_flos": 3.5197712872833024e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null