End of training

README.md

@@ -2,7 +2,7 @@
 tags:
 - generated_from_trainer
 datasets:
-- coco_dataset_script
+- ydshieh/coco_dataset_script
 model-index:
 - name: clip-roberta-finetuned
   results: []
@@ -13,7 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # clip-roberta-finetuned
 
-This model was trained from scratch on the coco_dataset_script dataset.
+This model was trained from scratch on the ydshieh/coco_dataset_script 2017 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.6496
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 3.0,
-    "eval_loss": 1.5614525079727173,
-    "eval_runtime": 22.7741,
-    "eval_samples_per_second": 1098.355,
-    "eval_steps_per_second": 2.152,
-    "train_loss": 0.3106431312077048,
-    "train_runtime": 2677.4197,
-    "train_samples_per_second": 663.048,
-    "train_steps_per_second": 1.295
+    "epoch": 10.0,
+    "eval_loss": 1.649561882019043,
+    "eval_runtime": 22.7131,
+    "eval_samples_per_second": 1101.305,
+    "eval_steps_per_second": 2.157,
+    "train_loss": 0.17783270721617042,
+    "train_runtime": 9016.469,
+    "train_samples_per_second": 656.302,
+    "train_steps_per_second": 1.282
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 3.0,
-    "eval_loss": 1.5614525079727173,
-    "eval_runtime": 22.7741,
-    "eval_samples_per_second": 1098.355,
-    "eval_steps_per_second": 2.152
+    "epoch": 10.0,
+    "eval_loss": 1.649561882019043,
+    "eval_runtime": 22.7131,
+    "eval_samples_per_second": 1101.305,
+    "eval_steps_per_second": 2.157
 }

runs/Feb13_02-52-08_28fc6ffbaa77/events.out.tfevents.1707801827.28fc6ffbaa77.111637.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08938ef2066d05f6ade5f4f1e3a949104f2dff7e493736834b8ff9f29b2aa837
+size 359

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 3.0,
-    "train_loss": 0.3106431312077048,
-    "train_runtime": 2677.4197,
-    "train_samples_per_second": 663.048,
-    "train_steps_per_second": 1.295
+    "epoch": 10.0,
+    "train_loss": 0.17783270721617042,
+    "train_runtime": 9016.469,
+    "train_samples_per_second": 656.302,
+    "train_steps_per_second": 1.282
 }

trainer_state.json

@@ -1,65 +1,167 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
+  "epoch": 10.0,
   "eval_steps": 500,
-  "global_step": 3468,
+  "global_step": 11560,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.43,
-      "learning_rate": 4.2791234140715114e-05,
-      "loss": 0.7219,
+      "learning_rate": 4.7837370242214535e-05,
+      "loss": 0.7277,
       "step": 500
     },
     {
       "epoch": 0.87,
-      "learning_rate": 3.558246828143022e-05,
-      "loss": 0.3833,
+      "learning_rate": 4.567474048442907e-05,
+      "loss": 0.3994,
       "step": 1000
     },
     {
       "epoch": 1.3,
-      "learning_rate": 2.8373702422145332e-05,
-      "loss": 0.2855,
+      "learning_rate": 4.35121107266436e-05,
+      "loss": 0.3097,
       "step": 1500
     },
     {
       "epoch": 1.73,
-      "learning_rate": 2.116493656286044e-05,
-      "loss": 0.243,
+      "learning_rate": 4.134948096885813e-05,
+      "loss": 0.2747,
       "step": 2000
     },
     {
       "epoch": 2.16,
-      "learning_rate": 1.395617070357555e-05,
-      "loss": 0.2044,
+      "learning_rate": 3.9186851211072664e-05,
+      "loss": 0.2418,
       "step": 2500
     },
     {
       "epoch": 2.6,
-      "learning_rate": 6.747404844290659e-06,
-      "loss": 0.1665,
+      "learning_rate": 3.70242214532872e-05,
+      "loss": 0.2109,
       "step": 3000
     },
     {
-      "epoch": 3.0,
-      "step": 3468,
-      "total_flos": 2.370754172808069e+17,
-      "train_loss": 0.3106431312077048,
-      "train_runtime": 2677.4197,
-      "train_samples_per_second": 663.048,
-      "train_steps_per_second": 1.295
+      "epoch": 3.03,
+      "learning_rate": 3.4861591695501735e-05,
+      "loss": 0.2044,
+      "step": 3500
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 3.269896193771627e-05,
+      "loss": 0.1703,
+      "step": 4000
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 3.05363321799308e-05,
+      "loss": 0.1709,
+      "step": 4500
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 2.8373702422145332e-05,
+      "loss": 0.1466,
+      "step": 5000
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 2.6211072664359864e-05,
+      "loss": 0.1433,
+      "step": 5500
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 2.4048442906574396e-05,
+      "loss": 0.1302,
+      "step": 6000
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 2.188581314878893e-05,
+      "loss": 0.1202,
+      "step": 6500
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 1.972318339100346e-05,
+      "loss": 0.1169,
+      "step": 7000
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 1.7560553633217993e-05,
+      "loss": 0.1016,
+      "step": 7500
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 1.5397923875432525e-05,
+      "loss": 0.1027,
+      "step": 8000
+    },
+    {
+      "epoch": 7.35,
+      "learning_rate": 1.323529411764706e-05,
+      "loss": 0.0892,
+      "step": 8500
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 1.1072664359861593e-05,
+      "loss": 0.087,
+      "step": 9000
+    },
+    {
+      "epoch": 8.22,
+      "learning_rate": 8.910034602076126e-06,
+      "loss": 0.0804,
+      "step": 9500
+    },
+    {
+      "epoch": 8.65,
+      "learning_rate": 6.747404844290659e-06,
+      "loss": 0.0749,
+      "step": 10000
+    },
+    {
+      "epoch": 9.08,
+      "learning_rate": 4.584775086505191e-06,
+      "loss": 0.0718,
+      "step": 10500
+    },
+    {
+      "epoch": 9.52,
+      "learning_rate": 2.4221453287197232e-06,
+      "loss": 0.0654,
+      "step": 11000
+    },
+    {
+      "epoch": 9.95,
+      "learning_rate": 2.5951557093425607e-07,
+      "loss": 0.0644,
+      "step": 11500
+    },
+    {
+      "epoch": 10.0,
+      "step": 11560,
+      "total_flos": 7.90251390936023e+17,
+      "train_loss": 0.17783270721617042,
+      "train_runtime": 9016.469,
+      "train_samples_per_second": 656.302,
+      "train_steps_per_second": 1.282
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3468,
+  "max_steps": 11560,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 2.370754172808069e+17,
+  "total_flos": 7.90251390936023e+17,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null