Model save

README.md

@@ -1,6 +1,6 @@
 ---
 license: cc-by-nc-4.0
-base_model: nguyenvulebinh/vi-mrc-large
+base_model: SalmonAI123/Bestclean-ViMrcbase-TFIDFPNS-ver1
 tags:
 - generated_from_trainer
 model-index:
@@ -13,9 +13,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # BestClean-ViMrclarge-TFIDFPNS-bigbadata-ver1
 
-This model is a fine-tuned version of [nguyenvulebinh/vi-mrc-large](https://huggingface.co/nguyenvulebinh/vi-mrc-large) on the None dataset.
+This model is a fine-tuned version of [SalmonAI123/Bestclean-ViMrcbase-TFIDFPNS-ver1](https://huggingface.co/SalmonAI123/Bestclean-ViMrcbase-TFIDFPNS-ver1) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8174
+- Loss: 0.3409
 
 ## Model description
 
@@ -46,8 +46,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log        | 1.0   | 47   | 0.8854          |
-| No log        | 2.0   | 94   | 0.8174          |
+| No log        | 1.0   | 10   | 0.3527          |
+| No log        | 2.0   | 20   | 0.3409          |
 
 
 ### Framework versions

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "nguyenvulebinh/vi-mrc-large",
+  "_name_or_path": "SalmonAI123/Bestclean-ViMrcbase-TFIDFPNS-ver1",
   "architectures": [
     "RobertaForQuestionAnswering"
   ],
@@ -10,14 +10,14 @@
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 1024,
+  "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
+  "intermediate_size": 3072,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,
   "model_type": "roberta",
-  "num_attention_heads": 16,
-  "num_hidden_layers": 24,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
   "output_past": true,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c4b281c75652d5c756c0e857ae8b4043019364f60bee9fd6009fdc9b5a752a52
-size 2235506473
+oid sha256:1ceba87f18a2dff2ec9382570139a7c1bb2cec1686585c2b60d4eac894e15dde
+size 1109886697

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4698f248a6d17dc6c6365af091d903340cfb427ecbb3b8630b057bf9feacaa68
+oid sha256:63ae0c0b0987cea6916d77f95e0bc6a012a04285090fe7cdb57a5a68b69f8d55
 size 4091