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": 5000,
-    "exact_match": 55.1,
-    "f1": 65.11208894562436,
-    "train_loss": 1.1906075286865234,
-    "train_runtime": 268.4069,
+    "exact_match": 55.64,
+    "f1": 65.9414733930188,
+    "train_loss": 0.6052761435508728,
+    "train_runtime": 530.4541,
     "train_samples": 5000,
-    "train_samples_per_second": 186.284,
-    "train_steps_per_second": 1.49
+    "train_samples_per_second": 188.518,
+    "train_steps_per_second": 1.508
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c61c4316a22f7e8110c0f0fc63c32f34a4c7da94474271471842aeeb6f910f55
+oid sha256:d84e9ea00131be5e82d3e7b5d5c98975f2a899a75e2edd6dfa9103b865020d55
 size 5588

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:96e86c4768705ddc3813082ed8a74da57e42e563732aa5a296b2171f03082e0c
-size 22922481
+oid sha256:28538ee4151eeb932aecd40a67ef3f169af848d36e27fc3ee4ce46056a90d806
+size 23089559

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 10.0,
+    "epoch": 20.0,
     "eval_samples": 5000,
-    "exact_match": 55.1,
-    "f1": 65.11208894562436
+    "exact_match": 55.64,
+    "f1": 65.9414733930188
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "train_loss": 1.1906075286865234,
-    "train_runtime": 268.4069,
+    "epoch": 20.0,
+    "train_loss": 0.6052761435508728,
+    "train_runtime": 530.4541,
     "train_samples": 5000,
-    "train_samples_per_second": 186.284,
-    "train_steps_per_second": 1.49
+    "train_samples_per_second": 188.518,
+    "train_steps_per_second": 1.508
 }

trainer_state.json

@@ -1,28 +1,34 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 10.0,
+  "epoch": 20.0,
   "eval_steps": 500,
-  "global_step": 400,
+  "global_step": 800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 10.0,
-      "step": 400,
-      "total_flos": 4899491251200000.0,
-      "train_loss": 1.1906075286865234,
-      "train_runtime": 268.4069,
-      "train_samples_per_second": 186.284,
-      "train_steps_per_second": 1.49
+      "epoch": 12.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.9181,
+      "step": 500
+    },
+    {
+      "epoch": 20.0,
+      "step": 800,
+      "total_flos": 9798982502400000.0,
+      "train_loss": 0.6052761435508728,
+      "train_runtime": 530.4541,
+      "train_samples_per_second": 188.518,
+      "train_steps_per_second": 1.508
     }
   ],
   "logging_steps": 500,
-  "max_steps": 400,
-  "num_train_epochs": 10,
+  "max_steps": 800,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 4899491251200000.0,
+  "total_flos": 9798982502400000.0,
   "trial_name": null,
   "trial_params": null
 }