rooa commited on
Commit
eda02e2
1 Parent(s): cd6dc07

Initial commit

Browse files
README.md CHANGED
@@ -1,3 +1,56 @@
1
  ---
2
  license: bsd-3-clause
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: bsd-3-clause
3
  ---
4
+ # CodeGen (CodeGen-NL 6B)
5
+
6
+ ## Model description
7
+
8
+ CodeGen is a family of autoregressive language models for **program synthesis** from the paper: [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) by Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong. The models are originally released in [this repository](https://github.com/salesforce/CodeGen), under 3 pre-training data variants (`NL`, `Multi`, `Mono`) and 4 model size variants (`350M`, `2B`, `6B`, `16B`).
9
+
10
+ The checkpoint included in this repository is denoted as **CodeGen-NL 6B** in the paper, where "NL" means it is pre-trained on the Pile and "6B" refers to the number of trainable parameters.
11
+
12
+ ## Training data
13
+
14
+ This checkpoint (CodeGen-NL 6B) was pre-trained on [the Pile](https://github.com/EleutherAI/the-pile), a large-scale curated dataset created by [EleutherAI](https://www.eleuther.ai/). Parts of the dataset include code data.
15
+
16
+ ## Training procedure
17
+
18
+ CodeGen was trained using cross-entropy loss to maximize the likelihood of sequential inputs.
19
+ The family of models are trained using 4 TPU-v4 chips by Google, leveraging data and model parallelism.
20
+ See Section 2.3 of the [paper](https://arxiv.org/abs/2203.13474) for more details.
21
+
22
+ ## Evaluation results
23
+
24
+ We evaluate our models on two code generation benchmark: HumanEval and MTPB. Please refer to the [paper](https://arxiv.org/abs/2203.13474) for more details.
25
+
26
+
27
+ ## Intended Use and Limitations
28
+
29
+ As an autoregressive language model, CodeGen is capable of extracting features from given natural language and programming language texts, and calculating the likelihood of them.
30
+ However, the model is intended for and best at **program synthesis**, that is, generating executable code given English prompts, where the prompts should be in the form of a comment string. The model can complete partially-generated code as well.
31
+
32
+ ## How to use
33
+
34
+ This model can be easily loaded using the `AutoModelForCausalLM` functionality:
35
+
36
+ ```python
37
+ from transformers import AutoTokenizer, AutoModelForCausalLM
38
+ tokenizer = AutoTokenizer.from_pretrained('Salesforce/codegen-6B-nl')
39
+ model = AutoModelForCausalLM.from_pretrained('Salesforce/codegen-6B-nl')
40
+
41
+ text = "def hello_world():"
42
+ input_ids = tokenizer(text, return_tensors="pt").input_ids
43
+ generated_ids = model.generate(input_ids, max_length=128)
44
+ print(tokenizer.decode(generated_ids[0], skip_special_tokens=True))
45
+ ```
46
+
47
+ ## BibTeX entry and citation info
48
+
49
+ ```bibtex
50
+ @article{Nijkamp2022ACP,
51
+ title={A Conversational Paradigm for Program Synthesis},
52
+ author={Nijkamp, Erik and Pang, Bo and Hayashi, Hiroaki and Tu, Lifu and Wang, Huan and Zhou, Yingbo and Savarese, Silvio and Xiong, Caiming},
53
+ journal={arXiv preprint},
54
+ year={2022}
55
+ }
56
+ ```
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "codegen-6B-nl",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "CodeGenForCausalLM"
6
+ ],
7
+ "attn_pdrop": 0.0,
8
+ "bos_token_id": 1,
9
+ "embd_pdrop": 0.0,
10
+ "eos_token_id": 2,
11
+ "gradient_checkpointing": false,
12
+ "initializer_range": 0.02,
13
+ "layer_norm_epsilon": 1e-05,
14
+ "model_type": "codegen",
15
+ "n_ctx": 2048,
16
+ "n_embd": 4096,
17
+ "n_head": 16,
18
+ "n_inner": null,
19
+ "n_layer": 33,
20
+ "n_positions": 2048,
21
+ "resid_pdrop": 0.0,
22
+ "rotary_dim": 64,
23
+ "scale_attn_weights": true,
24
+ "summary_activation": null,
25
+ "summary_first_dropout": 0.1,
26
+ "summary_proj_to_labels": true,
27
+ "summary_type": "cls_index",
28
+ "summary_use_proj": true,
29
+ "task_specific_params": {
30
+ "text-generation": {
31
+ "do_sample": true,
32
+ "max_length": 50,
33
+ "temperature": 1.0
34
+ }
35
+ },
36
+ "tokenizer_class": "GPT2Tokenizer",
37
+ "torch_dtype": "float16",
38
+ "transformers_version": "4.16.2",
39
+ "use_cache": true,
40
+ "vocab_size": 51200
41
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe50e2bcbd63fd4a06f2af033761435ceb5c34ba7ef2fe95149c070f9ee3705b
3
+ size 14132732279
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
vocab.json ADDED
The diff for this file is too large to render. See raw diff