Hugging Face's logo

neuropark
/
sahajBERT-NCC

Text Classification
Transformers
PyTorch
Bengali
albert
collaborative
bengali
SequenceClassification
Inference Endpoints
Model card Files Community
1
Upload commited on
Commit
3f4e734
•
1 Parent(s): e2f4304

Step 19519

Browse files
Files changed (3) hide show
  1. README.md +12 -12
  2. config.json +2 -2
  3. pytorch_model.bin +1 -1
README.md CHANGED
@@ -59,7 +59,7 @@ WIP
59
 
60
  ## Training data
61
 
62
- The model was initialized with pre-trained weights of [sahajBERT](https://huggingface.co/neuropark/sahajBERT) at step 19169 and trained on the `sna.bn` split of [IndicGlue](https://huggingface.co/datasets/indic_glue).
63
 
64
  ## Training procedure
65
 
@@ -72,27 +72,27 @@ Coming soon!
72
 
73
  ## Eval results
74
 
75
- accuracy: 0.9192062367115521
76
 
77
- loss: 0.2882556915283203
78
 
79
- macro_f1: 0.8987860792763027
80
 
81
- macro_precision: 0.9047413783524961
82
 
83
- macro_recall: 0.8938654072561182
84
 
85
- micro_f1: 0.919206236711552
86
 
87
- micro_precision: 0.9192062367115521
88
 
89
- micro_recall: 0.9192062367115521
90
 
91
- weighted_f1: 0.9190050640322454
92
 
93
- weighted_precision: 0.9191484504954501
94
 
95
- weighted_recall: 0.9192062367115521
96
 
97
 
98
 
 
59
 
60
  ## Training data
61
 
62
+ The model was initialized with pre-trained weights of [sahajBERT](https://huggingface.co/neuropark/sahajBERT) at step 19519 and trained on the `sna.bn` split of [IndicGlue](https://huggingface.co/datasets/indic_glue).
63
 
64
  ## Training procedure
65
 
 
72
 
73
  ## Eval results
74
 
75
+ accuracy: 0.9163713678242381
76
 
77
+ loss: 0.29771897196769714
78
 
79
+ macro_f1: 0.8951960933373831
80
 
81
+ macro_precision: 0.8958313840463195
82
 
83
+ macro_recall: 0.8962088356299692
84
 
85
+ micro_f1: 0.9163713678242381
86
 
87
+ micro_precision: 0.9163713678242381
88
 
89
+ micro_recall: 0.9163713678242381
90
 
91
+ weighted_f1: 0.916670480049282
92
 
93
+ weighted_precision: 0.9180146709071523
94
 
95
+ weighted_recall: 0.9163713678242381
96
 
97
 
98
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "albertvillanova/autonlp-indic_glue-multi_class_classification-0d675eb-1291121",
3
  "_num_labels": 6,
4
  "architectures": [
5
  "AlbertForSequenceClassification"
@@ -34,7 +34,7 @@
34
  "5": 5
35
  },
36
  "layer_norm_eps": 1e-12,
37
- "max_length": 192,
38
  "max_position_embeddings": 512,
39
  "model_type": "albert",
40
  "net_structure_type": 0,
 
1
  {
2
+ "_name_or_path": "albertvillanova/autonlp-indic_glue-multi_class_classification-1e67664-1311135",
3
  "_num_labels": 6,
4
  "architectures": [
5
  "AlbertForSequenceClassification"
 
34
  "5": 5
35
  },
36
  "layer_norm_eps": 1e-12,
37
+ "max_length": 128,
38
  "max_position_embeddings": 512,
39
  "model_type": "albert",
40
  "net_structure_type": 0,
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:77849cec43637ffb280fd0daf6214328bfc66b5d92d061adf47685dce59e89b4
3
  size 71800235
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eedcd154db07528d67ee486cb6fa1809eba8539719a3ca4e312e025193f4f4d6
3
  size 71800235