Training in progress epoch 0

README.md

@@ -1,5 +1,6 @@
 ---
-base_model: Aleksandar/electra-srb-ner
+license: mit
+base_model: sagorsarker/mbert-bengali-ner
 tags:
 - generated_from_keras_callback
 model-index:
@@ -12,11 +13,11 @@ probably proofread and complete it, then remove this comment. -->
 
 # imtiaz114/mbert-ner-baseline-1
 
-This model is a fine-tuned version of [Aleksandar/electra-srb-ner](https://huggingface.co/Aleksandar/electra-srb-ner) on an unknown dataset.
+This model is a fine-tuned version of [sagorsarker/mbert-bengali-ner](https://huggingface.co/sagorsarker/mbert-bengali-ner) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.0633
-- Validation Loss: 0.0885
-- Epoch: 3
+- Train Loss: 0.0902
+- Validation Loss: 0.0541
+- Epoch: 0
 
 ## Model description
 
@@ -35,17 +36,14 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 3184, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
+- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 4776, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}
 - training_precision: mixed_float16
 
 ### Training results
 
 | Train Loss | Validation Loss | Epoch |
 |:----------:|:---------------:|:-----:|
-| 0.1621     | 0.1146          | 0     |
-| 0.1105     | 0.1092          | 1     |
-| 0.0833     | 0.0845          | 2     |
-| 0.0633     | 0.0885          | 3     |
+| 0.0902     | 0.0541          | 0     |
 
 
 ### Framework versions

config.json

@@ -1,11 +1,12 @@
 {
-  "_name_or_path": "Aleksandar/electra-srb-ner",
+  "_name_or_path": "sagorsarker/mbert-bengali-ner",
   "architectures": [
-    "ElectraForTokenClassification"
+    "BertForTokenClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
-  "embedding_size": 768,
+  "directionality": "bidi",
+  "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -31,18 +32,18 @@
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,
-  "model_type": "electra",
+  "model_type": "bert",
   "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",
-  "summary_activation": "gelu",
-  "summary_last_dropout": 0.1,
-  "summary_type": "first",
-  "summary_use_proj": true,
-  "torch_dtype": "float32",
   "transformers_version": "4.31.0.dev0",
   "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 30522
+  "vocab_size": 105879
 }

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee211e6349346e7d6aa4f20b5979ba999d29edecfb4f8be188a74f3013f8a86d
-size 435876320
+oid sha256:8e057ce0484dfbed0f156e64ccdca303b190d3e7f9376bb33f5450e144a0d88f
+size 667358180

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e4a3974847fda36dc728132432b910a0f67598ca1cc7f0686901cf3ef82d4f03
-size 772540
+oid sha256:ce6b3bc23946dfbbdf84cf044e0a29b0378b3783e7d18b1c2339de5da4840056
+size 2563468

tokenizer_config.json

@@ -8,6 +8,6 @@
   "sep_token": "[SEP]",
   "strip_accents": null,
   "tokenize_chinese_chars": true,
-  "tokenizer_class": "ElectraTokenizer",
+  "tokenizer_class": "BertTokenizer",
   "unk_token": "[UNK]"
 }