Hugging Face's logo

CVR123
/
Muril-base-finetune-Tamil-qc

Text Classification
Transformers
TensorBoard
Safetensors
bert
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
CVR123 commited on
Commit
f7ced65
•
1 Parent(s): 73c1be3

MuRIL_for_TamilQC

Browse files
Files changed (6) hide show
  1. README.md +21 -21
  2. config.json +1 -1
  3. model.safetensors +1 -1
  4. runs/Apr23_10-46-36_f0560cf996e3/events.out.tfevents.1713869409.f0560cf996e3.1073.0 +3 -0
  5. runs/Apr23_11-01-18_f0560cf996e3/events.out.tfevents.1713870091.f0560cf996e3.1073.1 +3 -0
  6. training_args.bin +2 -2
README.md CHANGED
@@ -18,11 +18,11 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [google/muril-large-cased](https://huggingface.co/google/muril-large-cased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.8454
22
- - Precision: 0.8906
23
- - Recall: 0.8887
24
- - Accuracy: 0.8887
25
- - F1-score: 0.8891
26
 
27
  ## Model description
28
 
@@ -42,8 +42,8 @@ More information needed
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 5e-05
45
- - train_batch_size: 16
46
- - eval_batch_size: 16
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
@@ -53,21 +53,21 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Accuracy | F1-score |
55
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:--------:|:--------:|
56
- | 0.8303 | 1.0 | 309 | 1.7865 | 0.0487 | 0.2206 | 0.2206 | 0.0798 |
57
- | 0.4838 | 2.0 | 618 | 0.4928 | 0.8792 | 0.8725 | 0.8725 | 0.8729 |
58
- | 0.2588 | 3.0 | 927 | 0.5815 | 0.8742 | 0.8725 | 0.8725 | 0.8722 |
59
- | 0.2287 | 4.0 | 1236 | 0.5803 | 0.8767 | 0.8765 | 0.8765 | 0.8760 |
60
- | 0.1217 | 5.0 | 1545 | 0.7215 | 0.8773 | 0.8725 | 0.8725 | 0.8739 |
61
- | 0.0752 | 6.0 | 1854 | 0.6988 | 0.8938 | 0.8907 | 0.8907 | 0.8914 |
62
- | 0.0416 | 7.0 | 2163 | 0.7899 | 0.8966 | 0.8927 | 0.8927 | 0.8937 |
63
- | 0.0247 | 8.0 | 2472 | 0.8132 | 0.8927 | 0.8907 | 0.8907 | 0.8912 |
64
- | 0.0242 | 9.0 | 2781 | 0.8443 | 0.8944 | 0.8927 | 0.8927 | 0.8930 |
65
- | 0.0129 | 10.0 | 3090 | 0.8454 | 0.8906 | 0.8887 | 0.8887 | 0.8891 |
66
 
67
 
68
  ### Framework versions
69
 
70
- - Transformers 4.39.3
71
- - Pytorch 2.1.2
72
- - Datasets 2.18.0
73
- - Tokenizers 0.15.2
 
18
 
19
  This model is a fine-tuned version of [google/muril-large-cased](https://huggingface.co/google/muril-large-cased) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 0.6739
22
+ - Precision: 0.9145
23
+ - Recall: 0.9130
24
+ - Accuracy: 0.9130
25
+ - F1-score: 0.9135
26
 
27
  ## Model description
28
 
 
42
 
43
  The following hyperparameters were used during training:
44
  - learning_rate: 5e-05
45
+ - train_batch_size: 32
46
+ - eval_batch_size: 32
47
  - seed: 42
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Accuracy | F1-score |
55
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:--------:|:--------:|
56
+ | 0.6155 | 1.0 | 155 | 0.3583 | 0.8905 | 0.8887 | 0.8887 | 0.8889 |
57
+ | 0.2625 | 2.0 | 310 | 0.4334 | 0.8741 | 0.8623 | 0.8623 | 0.8618 |
58
+ | 0.1603 | 3.0 | 465 | 0.4511 | 0.8890 | 0.8866 | 0.8866 | 0.8862 |
59
+ | 0.0885 | 4.0 | 620 | 0.4819 | 0.9087 | 0.9028 | 0.9028 | 0.9034 |
60
+ | 0.0541 | 5.0 | 775 | 0.6202 | 0.8934 | 0.8927 | 0.8927 | 0.8929 |
61
+ | 0.0207 | 6.0 | 930 | 0.6179 | 0.9117 | 0.9089 | 0.9089 | 0.9093 |
62
+ | 0.0152 | 7.0 | 1085 | 0.6444 | 0.9060 | 0.9049 | 0.9049 | 0.9053 |
63
+ | 0.0057 | 8.0 | 1240 | 0.6534 | 0.9120 | 0.9109 | 0.9109 | 0.9114 |
64
+ | 0.0039 | 9.0 | 1395 | 0.6734 | 0.9146 | 0.9130 | 0.9130 | 0.9136 |
65
+ | 0.0039 | 10.0 | 1550 | 0.6739 | 0.9145 | 0.9130 | 0.9130 | 0.9135 |
66
 
67
 
68
  ### Framework versions
69
 
70
+ - Transformers 4.40.0
71
+ - Pytorch 2.2.1+cu121
72
+ - Datasets 2.19.0
73
+ - Tokenizers 0.19.1
config.json CHANGED
@@ -36,7 +36,7 @@
36
  "position_embedding_type": "absolute",
37
  "problem_type": "single_label_classification",
38
  "torch_dtype": "float32",
39
- "transformers_version": "4.39.3",
40
  "type_vocab_size": 2,
41
  "use_cache": true,
42
  "vocab_size": 197285
 
36
  "position_embedding_type": "absolute",
37
  "problem_type": "single_label_classification",
38
  "torch_dtype": "float32",
39
+ "transformers_version": "4.40.0",
40
  "type_vocab_size": 2,
41
  "use_cache": true,
42
  "vocab_size": 197285
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:10b4f5a70c4508f5517a17b5ebec3d4aa90c3a3128c8d58ca3fd45982e7382db
3
  size 2023700848
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59f90ce6358023d1894bb6b7c826df3bfc5faf79b62c2373202361ca7c5ab73e
3
  size 2023700848
runs/Apr23_10-46-36_f0560cf996e3/events.out.tfevents.1713869409.f0560cf996e3.1073.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec2a4f230cee066d02daac1f4de7e3a02c5ec0281daa825f0d623b38d255212f
3
+ size 12219
runs/Apr23_11-01-18_f0560cf996e3/events.out.tfevents.1713870091.f0560cf996e3.1073.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b4899fe1dc27dc40334ab71897732e97b2e64164290bbf482c0f60604d1fd2d1
3
+ size 12219
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b0854a1dd956f36148e18672401cbc9ff46ab3dc1c5d07a8ff2006bc231e9af5
3
- size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96eb1ad978430a32168c7ca305381619968f45ba84a149347ec9980d10888b41
3
+ size 4984