Hugging Face's logo

kodetr
/
stunting-qa-v4

Text Generation
Transformers
Safetensors
Indonesian
llama
stunting
kesehatan
anak
conversational
text-generation-inference
Model card Files Community
kodetr commited on
Commit
caf799c
·
verified ·
1 Parent(s): 2057e93

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +68 -24
README.md CHANGED
@@ -1,19 +1,19 @@
1
  ---
2
  library_name: transformers
 
 
 
 
3
  license: mit
4
  datasets:
5
- - kodetr/stunting-qa-v1
6
  language:
7
  - id
8
  metrics:
9
- - bleu
10
- - glue
11
  - rouge
12
- base_model:
13
- - meta-llama/Llama-3.2-3B-Instruct
14
  pipeline_tag: text-generation
15
- tags:
16
- - llama-factory
17
  ---
18
 
19
  ### Model Description
@@ -27,25 +27,69 @@ Konsultasi(Q&A) stunting pada anak
27
 
28
  ### Training
29
 
30
- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65d6d2f8b06abf924b24349d/6QshLUs9eAnsKJZOGP2jc.png)
31
 
32
  ### Information Result Training
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
33
 
34
- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65d6d2f8b06abf924b24349d/eHj9f-z3ZkLrtwvBcMzy-.png)
35
-
36
- ### Evaluation (GLUE)
37
-
38
- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/65d6d2f8b06abf924b24349d/obTYsbLRg9gD2_8rzN8G5.png)
39
-
40
- ### Evaluation (ROUGE)
41
-
42
- - **ROUGE Base Model:** {'rouge1': 0.32442426784015754, 'rouge2': 0.21091087320145585, 'rougeL': 0.2997913044723025, 'rougeLsum': 0.3028164031975752}
43
- - **ROUGE Finetuned Model:** {'rouge1': 0.45885281015363844, 'rouge2': 0.33303682535111623, 'rougeL': 0.42486098123248783, 'rougeLsum': 0.43147191004805796}
44
-
45
- ### BERTSCORE
46
-
47
- - **BERTScore Base Model - F1:** 0.7268
48
- - **BERTScore Finetuned Model - F1:** 0.7687
49
 
50
  ### Use with transformers
51
 
@@ -55,7 +99,7 @@ Pastikan untuk memperbarui instalasi transformer Anda melalui pip install --upgr
55
  import torch
56
  from transformers import pipeline
57
 
58
- model_id = "kodetr/stunting-qa-v3"
59
  pipe = pipeline(
60
  "text-generation",
61
  model=model_id,
 
1
  ---
2
  library_name: transformers
3
+ tags:
4
+ - stunting
5
+ - kesehatan
6
+ - anak
7
  license: mit
8
  datasets:
9
+ - kodetr/stunting-qa-2025
10
  language:
11
  - id
12
  metrics:
 
 
13
  - rouge
 
 
14
  pipeline_tag: text-generation
15
+ base_model:
16
+ - meta-llama/Llama-3.1-8B-Instruct
17
  ---
18
 
19
  ### Model Description
 
27
 
28
  ### Training
29
 
30
+ ![image/webp](https://cdn-uploads.huggingface.co/production/uploads/65d6d2f8b06abf924b24349d/YoWso_UhiKhCW6blruH2w.webp)
31
 
32
  ### Information Result Training
33
+ ```
34
+ ***** train metrics *****
35
+ epoch = 2.9987
36
+ num_input_tokens_seen = 1900976
37
+ total_flos = 79944066GF
38
+ train_loss = 0.872
39
+ train_runtime = 1:06:36.18
40
+ train_samples_per_second = 5.737
41
+ train_steps_per_second = 0.358
42
+ ```
43
+
44
+ ### Evaluation
45
+
46
+ ```
47
+ {
48
+ "predict_bleu-4": 46.238530502486256,
49
+ "predict_model_preparation_time": 0.0054,
50
+ "predict_rouge-1": 50.236485540434444,
51
+ "predict_rouge-2": 33.20428471604292,
52
+ "predict_rouge-l": 46.93391739073541,
53
+ "predict_runtime": 10532.8745,
54
+ "predict_samples_per_second": 0.726,
55
+ "predict_steps_per_second": 0.363
56
+ }
57
+ ```
58
+
59
+ ### Parameter
60
+
61
+ ```
62
+ LlamaConfig {
63
+ "architectures": [
64
+ "LlamaForCausalLM"
65
+ ],
66
+ "attention_bias": false,
67
+ "attention_dropout": 0.0,
68
+ "bos_token_id": 128000,
69
+ "eos_token_id": 128009,
70
+ "head_dim": 128,
71
+ "hidden_act": "silu",
72
+ "hidden_size": 4096,
73
+ "initializer_range": 0.02,
74
+ "intermediate_size": 14336,
75
+ "max_position_embeddings": 8192,
76
+ "mlp_bias": false,
77
+ "model_type": "llama",
78
+ "num_attention_heads": 32,
79
+ "num_hidden_layers": 32,
80
+ "num_key_value_heads": 8,
81
+ "pretraining_tp": 1,
82
+ "rms_norm_eps": 1e-05,
83
+ "rope_scaling": null,
84
+ "rope_theta": 500000.0,
85
+ "tie_word_embeddings": false,
86
+ "torch_dtype": "bfloat16",
87
+ "transformers_version": "4.51.3",
88
+ "use_cache": true,
89
+ "vocab_size": 128256
90
+ }
91
+ ```
92
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
93
 
94
  ### Use with transformers
95
 
 
99
  import torch
100
  from transformers import pipeline
101
 
102
+ model_id = "kodetr/stunting-qa-v5"
103
  pipe = pipeline(
104
  "text-generation",
105
  model=model_id,