End of training

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: google/muril-base-cased
+base_model: bert-base-multilingual-cased
 tags:
 - generated_from_trainer
 metrics:
@@ -18,13 +18,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # hindi-muril-ner
 
-This model is a fine-tuned version of [google/muril-base-cased](https://huggingface.co/google/muril-base-cased) 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:
-- Loss: 0.0006
-- Precision: 0.9963
-- Recall: 0.9984
-- F1: 0.9973
-- Accuracy: 0.9997
+- Loss: 0.0050
+- Precision: 0.9870
+- Recall: 0.9892
+- F1: 0.9881
+- Accuracy: 0.9989
 
 ## Model description
 
@@ -43,7 +43,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
+- learning_rate: 0.0001
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
@@ -53,11 +53,11 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch | Step   | Validation Loss | Precision | Recall | F1     | Accuracy |
-|:-------------:|:-----:|:------:|:---------------:|:---------:|:------:|:------:|:--------:|
-| 0.0065        | 1.0   | 47891  | 0.0044          | 0.9866    | 0.9900 | 0.9883 | 0.9988   |
-| 0.0016        | 2.0   | 95782  | 0.0011          | 0.9952    | 0.9975 | 0.9963 | 0.9996   |
-| 0.0008        | 3.0   | 143673 | 0.0006          | 0.9963    | 0.9984 | 0.9973 | 0.9997   |
+| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| 0.1968        | 1.0   | 882  | 0.0436          | 0.8199    | 0.8751 | 0.8466 | 0.9884   |
+| 0.0212        | 2.0   | 1764 | 0.0106          | 0.9695    | 0.9704 | 0.9700 | 0.9975   |
+| 0.0038        | 3.0   | 2646 | 0.0050          | 0.9870    | 0.9892 | 0.9881 | 0.9989   |
 
 
 ### Framework versions

config.json

@@ -1,11 +1,11 @@
 {
-  "_name_or_path": "google/muril-base-cased",
+  "_name_or_path": "bert-base-multilingual-cased",
   "architectures": [
     "BertForTokenClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
-  "embedding_size": 768,
+  "directionality": "bidi",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -45,10 +45,15 @@
   "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": "4.33.0",
   "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 197285
+  "vocab_size": 119547
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:500151558271b0bb02600903c071d5b7fdfd218ae8f0ba366f1e887372f7efdd
-size 947967145
+oid sha256:b0b8ec08cfe250ae84a7de5575efdb15357a30588b407c1c341e75dbf61268c6
+size 709156009

tokenizer_config.json

@@ -1,15 +1,12 @@
 {
   "clean_up_tokenization_spaces": true,
   "cls_token": "[CLS]",
-  "do_basic_tokenize": true,
   "do_lower_case": false,
-  "lowercase": false,
   "mask_token": "[MASK]",
   "model_max_length": 512,
-  "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",
-  "strip_accents": false,
+  "strip_accents": null,
   "tokenize_chinese_chars": true,
   "tokenizer_class": "BertTokenizer",
   "unk_token": "[UNK]"

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8ff6d35c147c6178ac7bcc5dc6e4f42d026b45bee25a92b34b5a799ad12fde5
+oid sha256:3a02899ca77a62009a9ab6091ed608c6798f2e943d8d5f2f72dc3bddbf244566
 size 4027