Fine-tuned POS tagger for Sanskrit

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-base-multilingual-cased](https://huggingface.co/bert-base-multilingual-cased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6171
-- Accuracy: 0.7924
+- Loss: 0.6378
+- Accuracy: 0.7854
 
 ## Model description
 
@@ -50,9 +50,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.9435        | 1.0   | 136  | 0.8051          | 0.7273   |
-| 0.6803        | 2.0   | 272  | 0.6515          | 0.7868   |
-| 0.5335        | 3.0   | 408  | 0.6171          | 0.7924   |
+| 0.951         | 1.0   | 136  | 0.8292          | 0.7296   |
+| 0.7118        | 2.0   | 272  | 0.6733          | 0.7765   |
+| 0.5504        | 3.0   | 408  | 0.6378          | 0.7854   |
 
 
 ### Framework versions

config.json

@@ -23,7 +23,11 @@
     "10": "LABEL_10",
     "11": "LABEL_11",
     "12": "LABEL_12",
-    "13": "LABEL_13"
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
@@ -34,6 +38,10 @@
     "LABEL_11": 11,
     "LABEL_12": 12,
     "LABEL_13": 13,
+    "LABEL_14": 14,
+    "LABEL_15": 15,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
     "LABEL_2": 2,
     "LABEL_3": 3,
     "LABEL_4": 4,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11f9e3ee0c6d35854aec97e7b29ebe7c7d2dff9b39925469cb3a0fb10e6159fb
-size 709117816
+oid sha256:00e65e244de6481a1d82b00f3b276a37aa9009ae6efc3d7fb2141b14d5e02310
+size 709130120

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:948dc94c805006c96c565fcdca8441cdba4feb2c2f06a549da263440412dc570
-size 5304
+oid sha256:eb762535dc1c3e44e410c5e389936f24dc2517d27621c686b081e5996d95af4c
+size 5240