├── LICENSE ├── README.md ├── assets ├── ds-coder-base.png ├── magicoder-s-ds.png ├── overview.svg └── result.png ├── data └── prompt.txt ├── demo ├── README.md └── magicoder_demo.py ├── experiments ├── README.md ├── __init__.py ├── __pycache__ │ ├── __init__.cpython-310.pyc │ ├── text2code.cpython-310.pyc │ └── utils.cpython-310.pyc ├── avg_multipl_e.py ├── collect_seed_documents.py ├── data_embedding │ ├── cosine_similarity.py │ ├── instructor_embedding.py │ └── length.py ├── ds_1000.py ├── extract_function_comment_pairs.py ├── lang_ablation.py ├── text2code.py ├── train_comment_code_pairs.py └── utils.py ├── pdm.lock ├── pyproject.toml ├── src └── magicoder │ ├── __init__.py │ ├── clean_data.py │ ├── convert_dense_to_moe.py │ ├── convert_moe_to_ffn.py │ ├── convert_moe_to_weighted.py │ ├── convert_weighted_to_dense.py │ ├── decontamination │ ├── __init__.py │ ├── benchmark_data.py │ ├── find_substrings.py │ └── utils.py │ ├── generate_data.py │ ├── llm_wrapper.py │ ├── minhash_deduplication.py │ ├── model │ ├── __pycache__ │ │ ├── configuration_llama_moe_upscaling.cpython-310.pyc │ │ ├── modeling_llama_moe_upscaling_hf.cpython-310.pyc │ │ └── modeling_llama_weighted_hf.cpython-310.pyc │ ├── configuration_llama_moe_upscaling.py │ ├── modeling_llama_moe_upscaling_hf.py │ └── modeling_llama_weighted_hf.py │ ├── preprocess_data.py │ ├── prompt_template.py │ ├── train.py │ ├── train_dense.sh │ ├── train_moe.sh │ ├── train_weighted.sh │ └── utils.py ├── test.sh ├── test_moe.sh ├── tests └── __init__.py └── xft_env.yml /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/README.md -------------------------------------------------------------------------------- /assets/ds-coder-base.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/assets/ds-coder-base.png -------------------------------------------------------------------------------- /assets/magicoder-s-ds.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/assets/magicoder-s-ds.png -------------------------------------------------------------------------------- /assets/overview.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/assets/overview.svg -------------------------------------------------------------------------------- /assets/result.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/assets/result.png -------------------------------------------------------------------------------- /data/prompt.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/data/prompt.txt -------------------------------------------------------------------------------- /demo/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/demo/README.md -------------------------------------------------------------------------------- /demo/magicoder_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/demo/magicoder_demo.py -------------------------------------------------------------------------------- /experiments/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/README.md -------------------------------------------------------------------------------- /experiments/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /experiments/__pycache__/__init__.cpython-310.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/__pycache__/__init__.cpython-310.pyc -------------------------------------------------------------------------------- /experiments/__pycache__/text2code.cpython-310.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/__pycache__/text2code.cpython-310.pyc -------------------------------------------------------------------------------- /experiments/__pycache__/utils.cpython-310.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/__pycache__/utils.cpython-310.pyc -------------------------------------------------------------------------------- /experiments/avg_multipl_e.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/avg_multipl_e.py -------------------------------------------------------------------------------- /experiments/collect_seed_documents.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/collect_seed_documents.py -------------------------------------------------------------------------------- /experiments/data_embedding/cosine_similarity.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/data_embedding/cosine_similarity.py -------------------------------------------------------------------------------- /experiments/data_embedding/instructor_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/data_embedding/instructor_embedding.py -------------------------------------------------------------------------------- /experiments/data_embedding/length.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/data_embedding/length.py -------------------------------------------------------------------------------- /experiments/ds_1000.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/ds_1000.py -------------------------------------------------------------------------------- /experiments/extract_function_comment_pairs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/extract_function_comment_pairs.py -------------------------------------------------------------------------------- /experiments/lang_ablation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/lang_ablation.py -------------------------------------------------------------------------------- /experiments/text2code.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/text2code.py -------------------------------------------------------------------------------- /experiments/train_comment_code_pairs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/train_comment_code_pairs.py -------------------------------------------------------------------------------- /experiments/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/experiments/utils.py -------------------------------------------------------------------------------- /pdm.lock: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/pdm.lock -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/pyproject.toml -------------------------------------------------------------------------------- /src/magicoder/__init__.py: -------------------------------------------------------------------------------- 1 | from . import utils 2 | -------------------------------------------------------------------------------- /src/magicoder/clean_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/clean_data.py -------------------------------------------------------------------------------- /src/magicoder/convert_dense_to_moe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/convert_dense_to_moe.py -------------------------------------------------------------------------------- /src/magicoder/convert_moe_to_ffn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/convert_moe_to_ffn.py -------------------------------------------------------------------------------- /src/magicoder/convert_moe_to_weighted.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/convert_moe_to_weighted.py -------------------------------------------------------------------------------- /src/magicoder/convert_weighted_to_dense.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/convert_weighted_to_dense.py -------------------------------------------------------------------------------- /src/magicoder/decontamination/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/decontamination/__init__.py -------------------------------------------------------------------------------- /src/magicoder/decontamination/benchmark_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/decontamination/benchmark_data.py -------------------------------------------------------------------------------- /src/magicoder/decontamination/find_substrings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/decontamination/find_substrings.py -------------------------------------------------------------------------------- /src/magicoder/decontamination/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/decontamination/utils.py -------------------------------------------------------------------------------- /src/magicoder/generate_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/generate_data.py -------------------------------------------------------------------------------- /src/magicoder/llm_wrapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/llm_wrapper.py -------------------------------------------------------------------------------- /src/magicoder/minhash_deduplication.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/minhash_deduplication.py -------------------------------------------------------------------------------- /src/magicoder/model/__pycache__/configuration_llama_moe_upscaling.cpython-310.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/model/__pycache__/configuration_llama_moe_upscaling.cpython-310.pyc -------------------------------------------------------------------------------- /src/magicoder/model/__pycache__/modeling_llama_moe_upscaling_hf.cpython-310.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/model/__pycache__/modeling_llama_moe_upscaling_hf.cpython-310.pyc -------------------------------------------------------------------------------- /src/magicoder/model/__pycache__/modeling_llama_weighted_hf.cpython-310.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/model/__pycache__/modeling_llama_weighted_hf.cpython-310.pyc -------------------------------------------------------------------------------- /src/magicoder/model/configuration_llama_moe_upscaling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/model/configuration_llama_moe_upscaling.py -------------------------------------------------------------------------------- /src/magicoder/model/modeling_llama_moe_upscaling_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/model/modeling_llama_moe_upscaling_hf.py -------------------------------------------------------------------------------- /src/magicoder/model/modeling_llama_weighted_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/model/modeling_llama_weighted_hf.py -------------------------------------------------------------------------------- /src/magicoder/preprocess_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/preprocess_data.py -------------------------------------------------------------------------------- /src/magicoder/prompt_template.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/prompt_template.py -------------------------------------------------------------------------------- /src/magicoder/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/train.py -------------------------------------------------------------------------------- /src/magicoder/train_dense.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/train_dense.sh -------------------------------------------------------------------------------- /src/magicoder/train_moe.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/train_moe.sh -------------------------------------------------------------------------------- /src/magicoder/train_weighted.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/train_weighted.sh -------------------------------------------------------------------------------- /src/magicoder/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/src/magicoder/utils.py -------------------------------------------------------------------------------- /test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/test.sh -------------------------------------------------------------------------------- /test_moe.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/test_moe.sh -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /xft_env.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/ise-uiuc/xft/HEAD/xft_env.yml --------------------------------------------------------------------------------