End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: distilbert-base-uncased
 tags:
+- question-answering
 - generated_from_trainer
+datasets:
+- squad
 model-index:
 - name: out
   results: []
@@ -13,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # out
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 10784,
-    "exact_match": 72.26111636707664,
-    "f1": 81.62230488906296,
-    "train_loss": 0.7383389687951589,
-    "train_runtime": 2418.3642,
+    "exact_match": 46.07379375591296,
+    "f1": 59.615253741406285,
+    "train_loss": 0.9091191572950066,
+    "train_runtime": 4840.8355,
     "train_samples": 88524,
-    "train_samples_per_second": 366.049,
-    "train_steps_per_second": 1.431
+    "train_samples_per_second": 365.739,
+    "train_steps_per_second": 1.43
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e6392f2477459d5693b48db94ee43d9a1d0d90747dc005d9505d5d6aa8358fb
+oid sha256:26b9976813d721358fcb7d0da6a5d5e6dc67432507a8e9890fbc48828c18e7c5
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fcd504c9f2dc1971ef98521e18441e60758e1e762b3444e5ae615b51f44ceb5f
-size 49254973
+oid sha256:3876c71c4d0c1b62abfa24ba0e3b9860114e49702643e1bc29d684bd13fec948
+size 50977373

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 10784,
-    "exact_match": 72.26111636707664,
-    "f1": 81.62230488906296
+    "exact_match": 46.07379375591296,
+    "f1": 59.615253741406285
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 0.7383389687951589,
-    "train_runtime": 2418.3642,
+    "epoch": 20.0,
+    "train_loss": 0.9091191572950066,
+    "train_runtime": 4840.8355,
     "train_samples": 88524,
-    "train_samples_per_second": 366.049,
-    "train_steps_per_second": 1.431
+    "train_samples_per_second": 365.739,
+    "train_steps_per_second": 1.43
 }

trainer_state.json

@@ -1,64 +1,106 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 3460,
+  "global_step": 6920,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.45,
-      "learning_rate": 4.2774566473988445e-05,
-      "loss": 1.6654,
+      "learning_rate": 4.6387283236994224e-05,
+      "loss": 1.6616,
       "step": 500
     },
     {
       "epoch": 2.89,
-      "learning_rate": 3.554913294797688e-05,
-      "loss": 0.9341,
+      "learning_rate": 4.2774566473988445e-05,
+      "loss": 0.9259,
       "step": 1000
     },
     {
       "epoch": 4.34,
-      "learning_rate": 2.832369942196532e-05,
-      "loss": 0.6779,
+      "learning_rate": 3.916184971098266e-05,
+      "loss": 0.6562,
       "step": 1500
     },
     {
       "epoch": 5.78,
-      "learning_rate": 2.1098265895953757e-05,
-      "loss": 0.5258,
+      "learning_rate": 3.554913294797688e-05,
+      "loss": 0.4915,
       "step": 2000
     },
     {
       "epoch": 7.23,
-      "learning_rate": 1.3872832369942197e-05,
-      "loss": 0.4303,
+      "learning_rate": 3.19364161849711e-05,
+      "loss": 0.3773,
       "step": 2500
     },
     {
       "epoch": 8.67,
-      "learning_rate": 6.647398843930635e-06,
-      "loss": 0.4153,
+      "learning_rate": 2.832369942196532e-05,
+      "loss": 0.3321,
       "step": 3000
     },
     {
-      "epoch": 10.0,
-      "step": 3460,
-      "total_flos": 8.674451270424576e+16,
-      "train_loss": 0.7383389687951589,
-      "train_runtime": 2418.3642,
-      "train_samples_per_second": 366.049,
-      "train_steps_per_second": 1.431
+      "epoch": 10.12,
+      "learning_rate": 2.471098265895954e-05,
+      "loss": 0.347,
+      "step": 3500
+    },
+    {
+      "epoch": 11.56,
+      "learning_rate": 2.1098265895953757e-05,
+      "loss": 0.3954,
+      "step": 4000
+    },
+    {
+      "epoch": 13.01,
+      "learning_rate": 1.748554913294798e-05,
+      "loss": 0.5166,
+      "step": 4500
+    },
+    {
+      "epoch": 14.45,
+      "learning_rate": 1.3872832369942197e-05,
+      "loss": 0.6966,
+      "step": 5000
+    },
+    {
+      "epoch": 15.9,
+      "learning_rate": 1.0260115606936416e-05,
+      "loss": 1.072,
+      "step": 5500
+    },
+    {
+      "epoch": 17.34,
+      "learning_rate": 6.647398843930635e-06,
+      "loss": 1.4468,
+      "step": 6000
+    },
+    {
+      "epoch": 18.79,
+      "learning_rate": 3.0346820809248555e-06,
+      "loss": 1.8735,
+      "step": 6500
+    },
+    {
+      "epoch": 20.0,
+      "step": 6920,
+      "total_flos": 1.7348902540849152e+17,
+      "train_loss": 0.9091191572950066,
+      "train_runtime": 4840.8355,
+      "train_samples_per_second": 365.739,
+      "train_steps_per_second": 1.43
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3460,
-  "num_train_epochs": 10,
+  "max_steps": 6920,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 8.674451270424576e+16,
+  "total_flos": 1.7348902540849152e+17,
   "trial_name": null,
   "trial_params": null
 }