End of training

README.md

@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the imdb dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6732
 - Accuracy: 0.9298
+- Loss: 0.6732
 
 ## Model description
 
@@ -68,12 +68,12 @@ The following hyperparameters were used during training:
 | 0.0177        | 2.0   | 3126  | 0.9300   | 0.5305          |
 | 0.0322        | 3.0   | 4689  | 0.9225   | 0.4810          |
 | 0.0122        | 4.0   | 6252  | 0.9244   | 0.5850          |
-| 0.0092        | 5.0   | 7815  | 0.5996   | 0.9282          |
-| 0.0064        | 6.0   | 9378  | 0.5723   | 0.9305          |
-| 0.0127        | 7.0   | 10941 | 0.6555   | 0.9283          |
-| 0.0035        | 8.0   | 12504 | 0.6549   | 0.9300          |
-| 0.0015        | 9.0   | 14067 | 0.6716   | 0.9288          |
-| 0.0004        | 10.0  | 15630 | 0.6732   | 0.9298          |
+| 0.0092        | 5.0   | 7815  | 0.9282   | 0.5996          |
+| 0.0064        | 6.0   | 9378  | 0.9305   | 0.5723          |
+| 0.0127        | 7.0   | 10941 | 0.9283   | 0.6555          |
+| 0.0035        | 8.0   | 12504 | 0.9300   | 0.6549          |
+| 0.0015        | 9.0   | 14067 | 0.9288   | 0.6716          |
+| 0.0004        | 10.0  | 15630 | 0.9298   | 0.6732          |
 
 
 ### Framework versions

config.json

@@ -22,7 +22,6 @@
   "n_heads": 12,
   "n_layers": 6,
   "pad_token_id": 0,
-  "problem_type": "single_label_classification",
   "qa_dropout": 0.1,
   "seq_classif_dropout": 0.2,
   "sinusoidal_pos_embds": false,

runs/Nov20_08-19-47_HAN/events.out.tfevents.1700439595.HAN.8728.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bff3b27babbeaf91d328b797b81393d6186dd09cbd9dd7025ca40fd29bda63f
+size 4609

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc27a1371876d51a967234d59715ddaec2646035b8250e655fd29bd91e61f2a9
+oid sha256:5c2d97f4822368453c27b0788b5af22cd6db8b9c137cc751ea7c49c95563cbca
 size 4091