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: 1.6228
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 3.0,
-    "eval_loss": 1.6256719827651978,
-    "eval_runtime": 85.7309,
-    "eval_samples_per_second": 291.773,
-    "eval_steps_per_second": 2.286,
-    "train_loss": 0.3325984596243642,
-    "train_runtime": 10519.3103,
-    "train_samples_per_second": 168.762,
+    "eval_loss": 1.6228044033050537,
+    "eval_runtime": 85.8562,
+    "eval_samples_per_second": 291.348,
+    "eval_steps_per_second": 2.283,
+    "train_loss": 0.33127041616516817,
+    "train_runtime": 10514.6739,
+    "train_samples_per_second": 168.836,
     "train_steps_per_second": 1.319
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.0,
-    "eval_loss": 1.6256719827651978,
-    "eval_runtime": 85.7309,
-    "eval_samples_per_second": 291.773,
-    "eval_steps_per_second": 2.286
+    "eval_loss": 1.6228044033050537,
+    "eval_runtime": 85.8562,
+    "eval_samples_per_second": 291.348,
+    "eval_steps_per_second": 2.283
 }

runs/Feb08_06-07-11_d0d182b85cf0/events.out.tfevents.1707383098.d0d182b85cf0.139131.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:addc25ba36cfefa110b38a28ea852155e88d0bef03d65ef6e1926fbe2bab0f9f
+size 359

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.3325984596243642,
-    "train_runtime": 10519.3103,
-    "train_samples_per_second": 168.762,
+    "train_loss": 0.33127041616516817,
+    "train_runtime": 10514.6739,
+    "train_samples_per_second": 168.836,
     "train_steps_per_second": 1.319
 }

trainer_state.json

@@ -11,85 +11,85 @@
     {
       "epoch": 0.11,
       "learning_rate": 4.819780853517878e-05,
-      "loss": 1.0435,
+      "loss": 1.0175,
       "step": 500
     },
     {
       "epoch": 0.22,
       "learning_rate": 4.639561707035756e-05,
-      "loss": 0.6931,
+      "loss": 0.6937,
       "step": 1000
     },
     {
       "epoch": 0.32,
       "learning_rate": 4.459342560553633e-05,
-      "loss": 0.6098,
+      "loss": 0.6068,
       "step": 1500
     },
     {
       "epoch": 0.43,
       "learning_rate": 4.2791234140715114e-05,
-      "loss": 0.5507,
+      "loss": 0.55,
       "step": 2000
     },
     {
       "epoch": 0.54,
       "learning_rate": 4.098904267589389e-05,
-      "loss": 0.5055,
+      "loss": 0.5041,
       "step": 2500
     },
     {
       "epoch": 0.65,
       "learning_rate": 3.9186851211072664e-05,
-      "loss": 0.4718,
+      "loss": 0.4698,
       "step": 3000
     },
     {
       "epoch": 0.76,
       "learning_rate": 3.7384659746251445e-05,
-      "loss": 0.4472,
+      "loss": 0.4439,
       "step": 3500
     },
     {
       "epoch": 0.87,
       "learning_rate": 3.558246828143022e-05,
-      "loss": 0.408,
+      "loss": 0.4068,
       "step": 4000
     },
     {
       "epoch": 0.97,
       "learning_rate": 3.3780276816608994e-05,
-      "loss": 0.386,
+      "loss": 0.3862,
       "step": 4500
     },
     {
       "epoch": 1.08,
       "learning_rate": 3.1978085351787776e-05,
-      "loss": 0.3326,
+      "loss": 0.332,
       "step": 5000
     },
     {
       "epoch": 1.19,
       "learning_rate": 3.0175893886966554e-05,
-      "loss": 0.3051,
+      "loss": 0.3044,
       "step": 5500
     },
     {
       "epoch": 1.3,
       "learning_rate": 2.8373702422145332e-05,
-      "loss": 0.2956,
+      "loss": 0.2945,
       "step": 6000
     },
     {
       "epoch": 1.41,
       "learning_rate": 2.657151095732411e-05,
-      "loss": 0.2854,
+      "loss": 0.2832,
       "step": 6500
     },
     {
       "epoch": 1.51,
       "learning_rate": 2.4769319492502884e-05,
-      "loss": 0.2789,
+      "loss": 0.2809,
       "step": 7000
     },
     {
@@ -101,7 +101,7 @@
     {
       "epoch": 1.73,
       "learning_rate": 2.116493656286044e-05,
-      "loss": 0.2532,
+      "loss": 0.2534,
       "step": 8000
     },
     {
@@ -113,13 +113,13 @@
     {
       "epoch": 1.95,
       "learning_rate": 1.7560553633217993e-05,
-      "loss": 0.2387,
+      "loss": 0.2397,
       "step": 9000
     },
     {
       "epoch": 2.05,
       "learning_rate": 1.575836216839677e-05,
-      "loss": 0.206,
+      "loss": 0.2052,
       "step": 9500
     },
     {
@@ -131,7 +131,7 @@
     {
       "epoch": 2.27,
       "learning_rate": 1.2153979238754325e-05,
-      "loss": 0.1758,
+      "loss": 0.1752,
       "step": 10500
     },
     {
@@ -143,40 +143,40 @@
     {
       "epoch": 2.49,
       "learning_rate": 8.54959630911188e-06,
-      "loss": 0.1632,
+      "loss": 0.1639,
       "step": 11500
     },
     {
       "epoch": 2.6,
       "learning_rate": 6.747404844290659e-06,
-      "loss": 0.1592,
+      "loss": 0.1594,
       "step": 12000
     },
     {
       "epoch": 2.7,
       "learning_rate": 4.945213379469435e-06,
-      "loss": 0.1542,
+      "loss": 0.1538,
       "step": 12500
     },
     {
       "epoch": 2.81,
       "learning_rate": 3.143021914648212e-06,
-      "loss": 0.1542,
+      "loss": 0.154,
       "step": 13000
     },
     {
       "epoch": 2.92,
       "learning_rate": 1.3408304498269898e-06,
-      "loss": 0.1481,
+      "loss": 0.1492,
       "step": 13500
     },
     {
       "epoch": 3.0,
       "step": 13872,
       "total_flos": 2.370754172808069e+17,
-      "train_loss": 0.3325984596243642,
-      "train_runtime": 10519.3103,
-      "train_samples_per_second": 168.762,
+      "train_loss": 0.33127041616516817,
+      "train_runtime": 10514.6739,
+      "train_samples_per_second": 168.836,
       "train_steps_per_second": 1.319
     }
   ],