Text Generation
Transformers
Safetensors
English
codegen
code
chincyk commited on
Commit
a29dd1a
·
verified ·
1 Parent(s): f04c66a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +33 -27
README.md CHANGED
@@ -1,12 +1,14 @@
1
- ---
2
- library_name: transformers
3
- tags:
4
- - code
5
- license: mit
6
- datasets:
7
- - iamtarun/python_code_instructions_18k_alpaca
8
- pipeline_tag: text-generation
9
- ---
 
 
10
 
11
  # PyCodeGen 350M
12
 
@@ -32,6 +34,10 @@ Finally model has been adapted to the Python language by training on the BigPyth
32
  The dataset contains problem descriptions and code in python language.
33
  This dataset is taken from sahil2801/code_instructions_120k, which adds a prompt column in alpaca style.
34
 
 
 
 
 
35
 
36
  ## Example of usage
37
 
@@ -68,25 +74,25 @@ print(tokenizer.decode(output[0], skip_special_tokens=True))
68
  ## Training parameters
69
 
70
  BitsAndBytes:
71
- - load_in_4bit=True,
72
- - bnb_4bit_quant_type="nf4",
73
- - bnb_4bit_use_double_quant=True,
74
- - bnb_4bit_compute_dtype=torch.bfloat16
75
 
76
  LoraConfig:
77
- - r=32,
78
- - lora_alpha=16,
79
- - target_modules='all-linear',
80
- - lora_dropout=0.1,
81
- - bias='none',
82
- - task_type='CASUAL_LM'
83
 
84
  Finetuning:
85
- - num_epochs = 15
86
- - train_batch_size = 4
87
- - eval_batch_size = 8
88
- - gradient_accumulation_steps = 8
89
- - learning_rate = 3e-4
90
- - weight_decay = 0.01
91
- - lr_scheduler_name = "cosine"
92
- - num_warmup_steps = 190
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - code
5
+ license: mit
6
+ datasets:
7
+ - iamtarun/python_code_instructions_18k_alpaca
8
+ pipeline_tag: text-generation
9
+ language:
10
+ - en
11
+ ---
12
 
13
  # PyCodeGen 350M
14
 
 
34
  The dataset contains problem descriptions and code in python language.
35
  This dataset is taken from sahil2801/code_instructions_120k, which adds a prompt column in alpaca style.
36
 
37
+ ## Intended uses
38
+
39
+ The model can be used to generate python code that solves task with optionally given input data.
40
+
41
 
42
  ## Example of usage
43
 
 
74
  ## Training parameters
75
 
76
  BitsAndBytes:
77
+ - load_in_4bit: True,
78
+ - bnb_4bit_quant_type: nf4,
79
+ - bnb_4bit_use_double_quant: True,
80
+ - bnb_4bit_compute_dtype: torch.bfloat16
81
 
82
  LoraConfig:
83
+ - r: 32,
84
+ - lora_alpha: 16,
85
+ - target_modules: all-linear,
86
+ - lora_dropout: 0.1,
87
+ - bias: none,
88
+ - task_type: CASUAL_LM
89
 
90
  Finetuning:
91
+ - num_epochs: 15
92
+ - train_batch_size: 4
93
+ - eval_batch_size: 8
94
+ - gradient_accumulation_steps: 8
95
+ - learning_rate: 3e-4
96
+ - weight_decay: 0.01
97
+ - lr_scheduler_name: cosine
98
+ - num_warmup_steps: 190