Training Complete

README.md

@@ -6,27 +6,27 @@ tags:
 datasets:
 - ag_news
 model-index:
-- name: roberta-based_uncased-finetuned-ag-news
+- name: bert-base-uncased-ag-news-finetuned
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# roberta-based_uncased-finetuned-ag-news
+# bert-base-uncased-ag-news-finetuned
 
 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the ag_news dataset.
 It achieves the following results on the evaluation set:
-- epoch: 2.0
-- eval_accuracy: 0.9462
-- eval_f1: 0.9462
-- eval_loss: 0.1799
-- eval_precision: 0.9464
-- eval_recall: 0.9462
-- eval_runtime: 39.6297
-- eval_samples_per_second: 302.804
-- eval_steps_per_second: 9.463
-- step: 6750
+- eval_loss: 0.2589
+- eval_accuracy: 0.9424
+- eval_f1(weighted): 0.9425
+- eval_precision(weighted): 0.9427
+- eval_recall(weighted): 0.9424
+- eval_runtime: 73.6126
+- eval_samples_per_second: 326.031
+- eval_steps_per_second: 6.792
+- epoch: 1.0
+- step: 2000
 
 ## Model description
 
@@ -46,12 +46,12 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 32
-- eval_batch_size: 32
+- train_batch_size: 48
+- eval_batch_size: 48
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 2
 
 ### Framework versions
 

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdd19575c8fb8b69ad64bee9727d0a082b150dd364179af48a100f035048d02d
+oid sha256:0e4abd28a302a38e1e1696caf6923f12fa5b5152c3c343d4ef52dd1e8ed237d2
 size 438009649