Training in progress epoch 0

README.md

@@ -1,4 +1,5 @@
 ---
+license: apache-2.0
 tags:
 - generated_from_keras_callback
 model-index:
@@ -11,12 +12,12 @@ probably proofread and complete it, then remove this comment. -->
 
 # kaisar-barlybay-sse/model
 
-This model was trained from scratch on an unknown dataset.
+This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 1.3977
+- Train Loss: 1.3923
 - Validation Loss: 1.3863
-- Train Accuracy: 0.2655
-- Epoch: 8
+- Train Accuracy: 0.2315
+- Epoch: 0
 
 ## Model description
 
@@ -42,15 +43,7 @@ The following hyperparameters were used during training:
 
 | Train Loss | Validation Loss | Train Accuracy | Epoch |
 |:----------:|:---------------:|:--------------:|:-----:|
-| 1.3897     | 1.3862          | 0.3693         | 0     |
-| 1.3996     | 1.3852          | 0.3872         | 1     |
-| 1.3926     | 1.3863          | 0.3513         | 2     |
-| 1.3906     | 1.3863          | 0.2475         | 3     |
-| 1.3933     | 1.3863          | 0.2535         | 4     |
-| 1.3895     | 1.3863          | 0.2375         | 5     |
-| 1.3947     | 1.3863          | 0.2335         | 6     |
-| 1.3928     | 1.3863          | 0.2595         | 7     |
-| 1.3977     | 1.3863          | 0.2655         | 8     |
+| 1.3923     | 1.3863          | 0.2315         | 0     |
 
 
 ### Framework versions

config.json

@@ -1,10 +1,11 @@
 {
-  "_name_or_path": "D:\\src\\pipelines\\domain_bert\\models\\domain_bert_kaz_legal\\model\\checkpoint-16000",
+  "_name_or_path": "bert-base-multilingual-cased",
   "architectures": [
     "BertForMultipleChoice"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
+  "directionality": "bidi",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -16,10 +17,14 @@
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
   "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
   "position_embedding_type": "absolute",
-  "torch_dtype": "float32",
   "transformers_version": "4.28.1",
   "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 30522
+  "vocab_size": 119547
 }

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa5050107d1b4b85944b617a85877eeff1ab98933818d70e5149dd29f617b4e6
-size 438203668
+oid sha256:ea78ba657adec69cff42953adbd4efa18977297a52d315f2b22d1bfcef98a78d
+size 711688468

tokenizer_config.json

@@ -1,11 +1,9 @@
 {
   "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
-  "do_basic_tokenize": true,
-  "do_lower_case": true,
+  "do_lower_case": false,
   "mask_token": "[MASK]",
-  "model_max_length": 1000000000000000019884624838656,
-  "never_split": null,
+  "model_max_length": 512,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
   "strip_accents": null,