diff --git a/.flake8 b/.flake8 deleted file mode 100644 index 229856aa4366..000000000000 --- a/.flake8 +++ /dev/null @@ -1,22 +0,0 @@ -[flake8] -ignore = - ;W503 line break before binary operator - W503, - ;E203 whitespace before ':' - E203, - -; exclude file -exclude = - .tox, - .git, - __pycache__, - build, - dist, - *.pyc, - *.egg-info, - .cache, - .eggs - -max-line-length = 120 - -per-file-ignores = __init__.py:F401 diff --git a/.github/ISSUE_TEMPLATE/config.yml b/.github/ISSUE_TEMPLATE/config.yml index b310fcfefc15..436bdf887c69 100644 --- a/.github/ISSUE_TEMPLATE/config.yml +++ b/.github/ISSUE_TEMPLATE/config.yml @@ -1,7 +1,7 @@ blank_issues_enabled: true contact_links: - name: ❓ Simple question - Slack Chat - url: https://join.slack.com/t/colossalaiworkspace/shared_invite/zt-z7b26eeb-CBp7jouvu~r0~lcFzX832w + url: https://github.com/hpcaitech/public_assets/tree/main/colossalai/contact/slack about: This issue tracker is not for technical support. Please use our Slack chat, and ask the community for help. - name: ❓ Simple question - WeChat url: https://github.com/hpcaitech/ColossalAI/blob/main/docs/images/WeChat.png diff --git a/.github/workflows/build_on_pr.yml b/.github/workflows/build_on_pr.yml index d112d61dd91d..e2114d43bcd0 100644 --- a/.github/workflows/build_on_pr.yml +++ b/.github/workflows/build_on_pr.yml @@ -61,7 +61,7 @@ jobs: run: shell: bash concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-repare-cache cancel-in-progress: true steps: - name: Copy testmon cache @@ -87,7 +87,7 @@ jobs: anyLibraryFileChanged: ${{ steps.find-lib-change.outputs.any_changed }} runs-on: ubuntu-latest concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-detect-change cancel-in-progress: true steps: - uses: actions/checkout@v2 @@ -141,13 +141,13 @@ jobs: runs-on: [self-hosted, gpu] container: image: hpcaitech/pytorch-cuda:1.12.0-11.3.0 - options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 + options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 -v /data/scratch/llama-tiny:/data/scratch/llama-tiny timeout-minutes: 60 defaults: run: shell: bash concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-run-test cancel-in-progress: true steps: - name: Checkout TensorNVMe @@ -208,12 +208,13 @@ jobs: - name: Execute Unit Testing run: | - CURL_CA_BUNDLE="" PYTHONPATH=$PWD pytest --testmon --testmon-cov=. --durations=10 tests/ + CURL_CA_BUNDLE="" PYTHONPATH=$PWD pytest -m "not largedist" --testmon --testmon-forceselect --testmon-cov=. --durations=10 tests/ env: DATA: /data/scratch/cifar-10 NCCL_SHM_DISABLE: 1 LD_LIBRARY_PATH: /github/home/.tensornvme/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 TESTMON_CORE_PKGS: /__w/ColossalAI/ColossalAI/requirements/requirements.txt,/__w/ColossalAI/ColossalAI/requirements/requirements-test.txt + LLAMA_PATH: /data/scratch/llama-tiny - name: Store Testmon Cache run: | diff --git a/.github/workflows/build_on_schedule.yml b/.github/workflows/build_on_schedule.yml index 03b47e6cb5b6..6c77377be34f 100644 --- a/.github/workflows/build_on_schedule.yml +++ b/.github/workflows/build_on_schedule.yml @@ -13,7 +13,7 @@ jobs: runs-on: [self-hosted, 8-gpu] container: image: hpcaitech/pytorch-cuda:1.12.0-11.3.0 - options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 + options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 -v /data/scratch/llama-tiny:/data/scratch/llama-tiny timeout-minutes: 40 steps: - name: Check GPU Availability # ensure all GPUs have enough memory @@ -64,6 +64,7 @@ jobs: env: DATA: /data/scratch/cifar-10 LD_LIBRARY_PATH: /github/home/.tensornvme/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 + LLAMA_PATH: /data/scratch/llama-tiny - name: Notify Lark id: message-preparation diff --git a/.github/workflows/compatiblity_test_on_dispatch.yml b/.github/workflows/compatiblity_test_on_dispatch.yml index 1778d64ee287..5083212993cc 100644 --- a/.github/workflows/compatiblity_test_on_dispatch.yml +++ b/.github/workflows/compatiblity_test_on_dispatch.yml @@ -44,13 +44,13 @@ jobs: name: Test for PyTorch Compatibility needs: matrix_preparation if: github.repository == 'hpcaitech/ColossalAI' - runs-on: [self-hosted, gpu] + runs-on: [self-hosted, 8-gpu] strategy: fail-fast: false matrix: ${{fromJson(needs.matrix_preparation.outputs.matrix)}} container: image: ${{ matrix.container }} - options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 + options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 -v /data/scratch/llama-tiny:/data/scratch/llama-tiny timeout-minutes: 120 steps: - name: Install dependencies @@ -64,7 +64,7 @@ jobs: - name: Install tensornvme run: | cd TensorNVMe - conda install cmake + apt update && apt install -y cmake pip install -r requirements.txt pip install -v . - uses: actions/checkout@v2 @@ -83,8 +83,7 @@ jobs: fi - name: Install Colossal-AI run: | - pip install -r requirements/requirements.txt - pip install -v --no-cache-dir . + CUDA_EXT=1 pip install -v . pip install -r requirements/requirements-test.txt - name: Unit Testing run: | @@ -93,3 +92,4 @@ jobs: DATA: /data/scratch/cifar-10 NCCL_SHM_DISABLE: 1 LD_LIBRARY_PATH: /github/home/.tensornvme/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 + LLAMA_PATH: /data/scratch/llama-tiny diff --git a/.github/workflows/compatiblity_test_on_pr.yml b/.github/workflows/compatiblity_test_on_pr.yml index 0aa9dffeb632..cc17c66f9c3a 100644 --- a/.github/workflows/compatiblity_test_on_pr.yml +++ b/.github/workflows/compatiblity_test_on_pr.yml @@ -13,7 +13,7 @@ jobs: outputs: matrix: ${{ steps.set-matrix.outputs.matrix }} concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-prepare-matrix cancel-in-progress: true steps: - uses: actions/checkout@v3 @@ -35,16 +35,16 @@ jobs: name: Test for PyTorch Compatibility needs: matrix_preparation if: github.repository == 'hpcaitech/ColossalAI' - runs-on: [self-hosted, gpu] + runs-on: [self-hosted, 8-gpu] strategy: fail-fast: false matrix: ${{fromJson(needs.matrix_preparation.outputs.matrix)}} container: image: ${{ matrix.container }} - options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 + options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 -v /data/scratch/llama-tiny:/data/scratch/llama-tiny timeout-minutes: 120 concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-run-test-${{ matrix.container }} cancel-in-progress: true steps: - name: Install dependencies @@ -58,7 +58,7 @@ jobs: - name: Install tensornvme run: | cd TensorNVMe - conda install cmake + apt update && apt install -y cmake pip install -r requirements.txt pip install -v . - uses: actions/checkout@v2 @@ -78,7 +78,7 @@ jobs: - name: Install Colossal-AI run: | - pip install -v --no-cache-dir . + CUDA_EXT=1 pip install -v . pip install -r requirements/requirements-test.txt - name: Unit Testing run: | @@ -87,3 +87,4 @@ jobs: DATA: /data/scratch/cifar-10 NCCL_SHM_DISABLE: 1 LD_LIBRARY_PATH: /github/home/.tensornvme/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 + LLAMA_PATH: /data/scratch/llama-tiny diff --git a/.github/workflows/compatiblity_test_on_schedule.yml b/.github/workflows/compatiblity_test_on_schedule.yml index 15ac4f1a92bb..158fe751bf2e 100644 --- a/.github/workflows/compatiblity_test_on_schedule.yml +++ b/.github/workflows/compatiblity_test_on_schedule.yml @@ -32,13 +32,13 @@ jobs: name: Test for PyTorch Compatibility needs: matrix_preparation if: github.repository == 'hpcaitech/ColossalAI' - runs-on: [self-hosted, gpu] + runs-on: [self-hosted, 8-gpu] strategy: fail-fast: false matrix: ${{fromJson(needs.matrix_preparation.outputs.matrix)}} container: image: ${{ matrix.container }} - options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 + options: --gpus all --rm -v /data/scratch/cifar-10:/data/scratch/cifar-10 -v /data/scratch/llama-tiny:/data/scratch/llama-tiny timeout-minutes: 120 steps: - name: Install dependencies @@ -54,7 +54,7 @@ jobs: - name: Install tensornvme run: | cd TensorNVMe - conda install cmake + apt update && apt install -y cmake pip install -r requirements.txt pip install -v . - uses: actions/checkout@v2 @@ -75,7 +75,7 @@ jobs: - name: Install Colossal-AI run: | - pip install -v --no-cache-dir . + CUDA_EXT=1 pip install -v . pip install -r requirements/requirements-test.txt - name: Unit Testing @@ -85,6 +85,7 @@ jobs: DATA: /data/scratch/cifar-10 NCCL_SHM_DISABLE: 1 LD_LIBRARY_PATH: /github/home/.tensornvme/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64 + LLAMA_PATH: /data/scratch/llama-tiny - name: Notify Lark id: message-preparation diff --git a/.github/workflows/doc_check_on_pr.yml b/.github/workflows/doc_check_on_pr.yml index ae9e311649f7..ee8a82128dd7 100644 --- a/.github/workflows/doc_check_on_pr.yml +++ b/.github/workflows/doc_check_on_pr.yml @@ -17,7 +17,7 @@ jobs: github.event.pull_request.base.repo.full_name == 'hpcaitech/ColossalAI' runs-on: ubuntu-latest concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-check-i18n cancel-in-progress: true steps: - uses: actions/checkout@v2 @@ -35,7 +35,7 @@ jobs: github.event.pull_request.base.repo.full_name == 'hpcaitech/ColossalAI' runs-on: ubuntu-latest concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-check-doc cancel-in-progress: true steps: - uses: actions/checkout@v2 diff --git a/.github/workflows/doc_test_on_pr.yml b/.github/workflows/doc_test_on_pr.yml index bf9ed64c8a7e..f1e7a2d0cab0 100644 --- a/.github/workflows/doc_test_on_pr.yml +++ b/.github/workflows/doc_test_on_pr.yml @@ -20,7 +20,7 @@ jobs: any_changed: ${{ steps.changed-files.outputs.any_changed }} changed_files: ${{ steps.changed-files.outputs.all_changed_files }} concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-detect-change cancel-in-progress: true name: Detect changed example files steps: @@ -63,7 +63,7 @@ jobs: run: shell: bash concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-run-doctest cancel-in-progress: true steps: - name: Checkout ColossalAI-Documentation @@ -89,7 +89,7 @@ jobs: - name: Install ColossalAI run: | source activate pytorch - pip install -v . + CUDA_EXT=1 pip install -v . - name: Test the Doc run: | diff --git a/.github/workflows/doc_test_on_schedule.yml b/.github/workflows/doc_test_on_schedule.yml index 6b4f5d1f908c..027fbfd0aaeb 100644 --- a/.github/workflows/doc_test_on_schedule.yml +++ b/.github/workflows/doc_test_on_schedule.yml @@ -32,7 +32,7 @@ jobs: - name: Install ColossalAI run: | - pip install -v . + CUDA_EXT=1 pip install -v . - name: Install Doc Test Requirements run: | diff --git a/.github/workflows/example_check_on_dispatch.yml b/.github/workflows/example_check_on_dispatch.yml index 620d4771af55..9d3bd9a48235 100644 --- a/.github/workflows/example_check_on_dispatch.yml +++ b/.github/workflows/example_check_on_dispatch.yml @@ -53,7 +53,7 @@ jobs: uses: actions/checkout@v3 - name: Install Colossal-AI run: | - pip install -v . + CUDA_EXT=1 pip install -v . - name: Test the example run: | dir=${{ matrix.directory }} diff --git a/.github/workflows/example_check_on_pr.yml b/.github/workflows/example_check_on_pr.yml index d990a76ca6db..5934704f4102 100644 --- a/.github/workflows/example_check_on_pr.yml +++ b/.github/workflows/example_check_on_pr.yml @@ -21,7 +21,7 @@ jobs: anyChanged: ${{ steps.setup-matrix.outputs.anyChanged }} name: Detect changed example files concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-detect-change cancel-in-progress: true steps: - uses: actions/checkout@v3 @@ -81,14 +81,14 @@ jobs: options: --gpus all --rm -v /data/scratch/examples-data:/data/ timeout-minutes: 10 concurrency: - group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} + group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}-run-example-${{ matrix.directory }} cancel-in-progress: true steps: - uses: actions/checkout@v3 - name: Install Colossal-AI run: | - pip install -v . + CUDA_EXT=1 pip install -v . - name: Test the example run: | diff --git a/.github/workflows/example_check_on_schedule.yml b/.github/workflows/example_check_on_schedule.yml index bd52ca4321a2..5ed128c3ebc5 100644 --- a/.github/workflows/example_check_on_schedule.yml +++ b/.github/workflows/example_check_on_schedule.yml @@ -42,7 +42,7 @@ jobs: - name: Install Colossal-AI run: | - pip install -v . + CUDA_EXT=1 pip install -v . - name: Traverse all files run: | diff --git a/.github/workflows/run_chatgpt_examples.yml b/.github/workflows/run_chatgpt_examples.yml index a336526897e2..f9e9f400962e 100644 --- a/.github/workflows/run_chatgpt_examples.yml +++ b/.github/workflows/run_chatgpt_examples.yml @@ -49,5 +49,5 @@ jobs: NCCL_SHM_DISABLE: 1 MAX_JOBS: 8 SFT_DATASET: /data/scratch/github_actions/chat/data.json - PROMPT_PATH: /data/scratch/github_actions/chat/prompts_en.jsonl + PROMPT_DATASET: /data/scratch/github_actions/chat/prompts_en.jsonl PRETRAIN_DATASET: /data/scratch/github_actions/chat/alpaca_data.json diff --git a/.github/workflows/scripts/check_doc_i18n.py b/.github/workflows/scripts/check_doc_i18n.py index 1aa7283e9e52..1e7f0c33a785 100644 --- a/.github/workflows/scripts/check_doc_i18n.py +++ b/.github/workflows/scripts/check_doc_i18n.py @@ -22,13 +22,13 @@ def compare_dirs(dir1, dir2): # If the corresponding item doesn't exist in the second directory, the directories are different if not os.path.exists(item_path2): - print(f'Found mismatch: {item_path1}, {item_path2}') + print(f"Found mismatch: {item_path1}, {item_path2}") return False # If the corresponding item is a directory, we compare the two directories recursively if os.path.isdir(item_path1) and os.path.isdir(item_path2): if not compare_dirs(item_path1, item_path2): - print(f'Found mismatch: {item_path1}, {item_path2}') + print(f"Found mismatch: {item_path1}, {item_path2}") return False # both are files @@ -37,16 +37,16 @@ def compare_dirs(dir1, dir2): # If the corresponding item is not a file or a directory, the directories are different else: - print(f'Found mismatch: {item_path1}, {item_path2}') + print(f"Found mismatch: {item_path1}, {item_path2}") return False # If all items are the same, the directories are the same return True -if __name__ == '__main__': +if __name__ == "__main__": parser = argparse.ArgumentParser() - parser.add_argument('-d', '--directory', help="The directory where the multi-language source files are kept.") + parser.add_argument("-d", "--directory", help="The directory where the multi-language source files are kept.") args = parser.parse_args() i18n_folders = os.listdir(args.directory) @@ -56,7 +56,7 @@ def compare_dirs(dir1, dir2): for i in range(1, len(i18n_folders)): dir1 = i18n_folders[0] dir2 = i18n_folders[i] - print(f'comparing {dir1} vs {dir2}') + print(f"comparing {dir1} vs {dir2}") match = compare_dirs(i18n_folders[0], i18n_folders[i]) if not match: diff --git a/.github/workflows/scripts/example_checks/check_dispatch_inputs.py b/.github/workflows/scripts/example_checks/check_dispatch_inputs.py index 5bec96187e0c..91778f692cc6 100644 --- a/.github/workflows/scripts/example_checks/check_dispatch_inputs.py +++ b/.github/workflows/scripts/example_checks/check_dispatch_inputs.py @@ -4,7 +4,7 @@ def check_inputs(input_list): for path in input_list: - real_path = os.path.join('examples', path) + real_path = os.path.join("examples", path) if not os.path.exists(real_path): return False return True @@ -12,16 +12,16 @@ def check_inputs(input_list): def main(): parser = argparse.ArgumentParser() - parser.add_argument('-f', '--fileNameList', type=str, help="List of file names") + parser.add_argument("-f", "--fileNameList", type=str, help="List of file names") args = parser.parse_args() name_list = args.fileNameList.split(",") is_correct = check_inputs(name_list) if is_correct: - print('success') + print("success") else: - print('failure') + print("failure") -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/.github/workflows/scripts/example_checks/check_example_weekly.py b/.github/workflows/scripts/example_checks/check_example_weekly.py index 83eff644e315..95a3d24c9a78 100644 --- a/.github/workflows/scripts/example_checks/check_example_weekly.py +++ b/.github/workflows/scripts/example_checks/check_example_weekly.py @@ -17,21 +17,21 @@ def show_files(path, all_files): def join(input_list, sep=None): - return (sep or ' ').join(input_list) + return (sep or " ").join(input_list) def main(): - contents = show_files('examples/', []) + contents = show_files("examples/", []) all_loc = [] for file_loc in contents: - split_loc = file_loc.split('/') + split_loc = file_loc.split("/") # must have two sub-folder levels after examples folder, such as examples/images/vit is acceptable, examples/images/README.md is not, examples/requirements.txt is not. if len(split_loc) >= 4: - re_loc = '/'.join(split_loc[1:3]) + re_loc = "/".join(split_loc[1:3]) if re_loc not in all_loc: all_loc.append(re_loc) print(all_loc) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/.github/workflows/scripts/example_checks/detect_changed_example.py b/.github/workflows/scripts/example_checks/detect_changed_example.py index c69d95a552e9..95f671dfb32b 100644 --- a/.github/workflows/scripts/example_checks/detect_changed_example.py +++ b/.github/workflows/scripts/example_checks/detect_changed_example.py @@ -3,7 +3,7 @@ def main(): parser = argparse.ArgumentParser() - parser.add_argument('-f', '--fileNameList', type=str, help="The list of changed files") + parser.add_argument("-f", "--fileNameList", type=str, help="The list of changed files") args = parser.parse_args() name_list = args.fileNameList.split(":") folder_need_check = set() @@ -15,10 +15,10 @@ def main(): # - application # - file if loc.split("/")[0] == "examples" and len(loc.split("/")) >= 4: - folder_need_check.add('/'.join(loc.split("/")[1:3])) + folder_need_check.add("/".join(loc.split("/")[1:3])) # Output the result using print. Then the shell can get the values. print(list(folder_need_check)) -if __name__ == '__main__': +if __name__ == "__main__": main() diff --git a/.github/workflows/scripts/generate_leaderboard_and_send_to_lark.py b/.github/workflows/scripts/generate_leaderboard_and_send_to_lark.py index 2884e38dd3dd..412b14c7b283 100644 --- a/.github/workflows/scripts/generate_leaderboard_and_send_to_lark.py +++ b/.github/workflows/scripts/generate_leaderboard_and_send_to_lark.py @@ -74,16 +74,16 @@ def get_organization_repositories(github_token, organization_name) -> List[str]: # prepare header headers = { - 'Authorization': f'Bearer {github_token}', - 'Accept': 'application/vnd.github+json', - 'X-GitHub-Api-Version': '2022-11-28' + "Authorization": f"Bearer {github_token}", + "Accept": "application/vnd.github+json", + "X-GitHub-Api-Version": "2022-11-28", } res = requests.get(url, headers=headers).json() repo_list = [] for item in res: - repo_list.append(item['name']) + repo_list.append(item["name"]) return repo_list @@ -97,9 +97,9 @@ def get_issue_pull_request_comments(github_token: str, org_name: str, repo_name: """ # prepare header headers = { - 'Authorization': f'Bearer {github_token}', - 'Accept': 'application/vnd.github+json', - 'X-GitHub-Api-Version': '2022-11-28' + "Authorization": f"Bearer {github_token}", + "Accept": "application/vnd.github+json", + "X-GitHub-Api-Version": "2022-11-28", } user_engagement_count = {} @@ -107,28 +107,28 @@ def get_issue_pull_request_comments(github_token: str, org_name: str, repo_name: # do pagination to the API page = 1 while True: - comment_api = f'https://api.github.com/repos/{org_name}/{repo_name}/issues/comments?since={since}&page={page}' + comment_api = f"https://api.github.com/repos/{org_name}/{repo_name}/issues/comments?since={since}&page={page}" comment_response = requests.get(comment_api, headers=headers).json() if len(comment_response) == 0: break else: for item in comment_response: - comment_author_relationship = item['author_association'] - if comment_author_relationship != 'MEMBER': + comment_author_relationship = item["author_association"] + if comment_author_relationship != "MEMBER": # if the comment is not made by our member # we don't count this comment towards user engagement continue - issue_id = item['issue_url'].split('/')[-1] - issue_api = f'https://api.github.com/repos/{org_name}/{repo_name}/issues/{issue_id}' + issue_id = item["issue_url"].split("/")[-1] + issue_api = f"https://api.github.com/repos/{org_name}/{repo_name}/issues/{issue_id}" issue_response = requests.get(issue_api, headers=headers).json() - issue_author_relationship = issue_response['author_association'] + issue_author_relationship = issue_response["author_association"] - if issue_author_relationship != 'MEMBER': + if issue_author_relationship != "MEMBER": # this means that the issue/PR is not created by our own people # any comments in this issue/PR by our member will be counted towards the leaderboard - member_name = item['user']['login'] + member_name = item["user"]["login"] if member_name in user_engagement_count: user_engagement_count[member_name] += 1 @@ -153,7 +153,7 @@ def _generate_discussion_query(num, cursor: str = None): if cursor is None: offset_str = "" else: - offset_str = f", after: \"{cursor}\"" + offset_str = f', after: "{cursor}"' query = f""" {{ repository(owner: "{org_name}", name: "{repo_name}"){{ @@ -182,7 +182,7 @@ def _generate_comment_reply_count_for_discussion(discussion_number, num, cursor: if cursor is None: offset_str = "" else: - offset_str = f", before: \"{cursor}\"" + offset_str = f', before: "{cursor}"' query = f""" {{ repository(owner: "{org_name}", name: "{repo_name}"){{ @@ -220,8 +220,8 @@ def _generate_comment_reply_count_for_discussion(discussion_number, num, cursor: # a utility function to make call to Github GraphQL API def _call_graphql_api(query): headers = {"Authorization": f"Bearer {github_token}"} - json_data = {'query': query} - response = requests.post('https://api.github.com/graphql', json=json_data, headers=headers) + json_data = {"query": query} + response = requests.post("https://api.github.com/graphql", json=json_data, headers=headers) data = response.json() return data @@ -234,21 +234,21 @@ def _call_graphql_api(query): data = _call_graphql_api(query) found_discussion_out_of_time_range = False - edges = data['data']['repository']['discussions']['edges'] + edges = data["data"]["repository"]["discussions"]["edges"] if len(edges) == 0: break else: # keep the discussion whose author is not a member for edge in edges: # print the discussion title - discussion = edge['node'] - discussion_updated_at = str2datetime(discussion['updatedAt']) + discussion = edge["node"] + discussion_updated_at = str2datetime(discussion["updatedAt"]) # check if the updatedAt is within the last 7 days # if yes, add it to discussion_numbers if discussion_updated_at > since: - if discussion['authorAssociation'] != 'MEMBER': - discussion_numbers.append(discussion['number']) + if discussion["authorAssociation"] != "MEMBER": + discussion_numbers.append(discussion["number"]) else: found_discussion_out_of_time_range = True @@ -256,7 +256,7 @@ def _call_graphql_api(query): break else: # update cursor - cursor = edges[-1]['cursor'] + cursor = edges[-1]["cursor"] # get the discussion comments and replies made by our member user_engagement_count = {} @@ -269,42 +269,42 @@ def _call_graphql_api(query): data = _call_graphql_api(query) # get the comments - edges = data['data']['repository']['discussion']['comments']['edges'] + edges = data["data"]["repository"]["discussion"]["comments"]["edges"] # update the cursor if len(edges) == 0: break else: # update cursor for pagination - cursor = edges[-1]['cursor'] + cursor = edges[-1]["cursor"] for edge in edges: - comment = edge['node'] - if comment['authorAssociation'] == 'MEMBER': + comment = edge["node"] + if comment["authorAssociation"] == "MEMBER": # check if the updatedAt is within the last 7 days # if yes, add it to user_engagement_count - comment_updated_at = datetime.strptime(comment['updatedAt'], "%Y-%m-%dT%H:%M:%SZ") + comment_updated_at = datetime.strptime(comment["updatedAt"], "%Y-%m-%dT%H:%M:%SZ") if comment_updated_at > since: - member_name = comment['author']['login'] + member_name = comment["author"]["login"] if member_name in user_engagement_count: user_engagement_count[member_name] += 1 else: user_engagement_count[member_name] = 1 # get the replies - reply_edges = comment['replies']['edges'] + reply_edges = comment["replies"]["edges"] if len(reply_edges) == 0: continue else: for reply_edge in reply_edges: - reply = reply_edge['node'] - if reply['authorAssociation'] == 'MEMBER': + reply = reply_edge["node"] + if reply["authorAssociation"] == "MEMBER": # check if the updatedAt is within the last 7 days # if yes, add it to discussion_numbers - reply_updated_at = datetime.strptime(reply['updatedAt'], "%Y-%m-%dT%H:%M:%SZ") + reply_updated_at = datetime.strptime(reply["updatedAt"], "%Y-%m-%dT%H:%M:%SZ") if reply_updated_at > since: - member_name = reply['author']['login'] + member_name = reply["author"]["login"] if member_name in user_engagement_count: user_engagement_count[member_name] += 1 else: @@ -312,7 +312,9 @@ def _call_graphql_api(query): return user_engagement_count -def generate_user_engagement_leaderboard_image(github_token: str, org_name: str, repo_list: List[str], output_path: str) -> bool: +def generate_user_engagement_leaderboard_image( + github_token: str, org_name: str, repo_list: List[str], output_path: str +) -> bool: """ Generate the user engagement leaderboard image for stats within the last 7 days @@ -335,16 +337,19 @@ def _update_count(counter): else: total_engagement_count[name] = count - for repo_name in repo_list: print(f"Fetching user engagement count for {repo_name}/{repo_name}") - issue_pr_engagement_count = get_issue_pull_request_comments(github_token=github_token, org_name=org_name, repo_name=repo_name, since=start_datetime_str) - discussion_engagement_count = get_discussion_comments(github_token=github_token, org_name=org_name, repo_name=repo_name, since=start_datetime) + issue_pr_engagement_count = get_issue_pull_request_comments( + github_token=github_token, org_name=org_name, repo_name=repo_name, since=start_datetime_str + ) + discussion_engagement_count = get_discussion_comments( + github_token=github_token, org_name=org_name, repo_name=repo_name, since=start_datetime + ) # update the total engagement count _update_count(issue_pr_engagement_count) _update_count(discussion_engagement_count) - + # prepare the data for plotting x = [] y = [] @@ -363,7 +368,7 @@ def _update_count(counter): # plot the leaderboard xlabel = f"Number of Comments made (since {start_datetime_str})" ylabel = "Member" - title = 'Active User Engagement Leaderboard' + title = "Active User Engagement Leaderboard" plot_bar_chart(x, y, xlabel=xlabel, ylabel=ylabel, title=title, output_path=output_path) return True else: @@ -380,16 +385,16 @@ def generate_contributor_leaderboard_image(github_token, org_name, repo_list, ou """ # request to the Github API to get the users who have contributed in the last 7 days headers = { - 'Authorization': f'Bearer {github_token}', - 'Accept': 'application/vnd.github+json', - 'X-GitHub-Api-Version': '2022-11-28' + "Authorization": f"Bearer {github_token}", + "Accept": "application/vnd.github+json", + "X-GitHub-Api-Version": "2022-11-28", } counter = Counter() start_datetime = get_utc_time_one_week_ago() def _get_url(org_name, repo_name, page): - return f'https://api.github.com/repos/{org_name}/{repo_name}/pulls?per_page=50&page={page}&state=closed' + return f"https://api.github.com/repos/{org_name}/{repo_name}/pulls?per_page=50&page={page}&state=closed" def _iterate_by_page(org_name, repo_name): page = 1 @@ -415,8 +420,8 @@ def _iterate_by_page(org_name, repo_name): # count the pull request and author from response for pr_data in response: - merged_at = pr_data['merged_at'] - author = pr_data['user']['login'] + merged_at = pr_data["merged_at"] + author = pr_data["user"]["login"] if merged_at is None: continue @@ -439,7 +444,7 @@ def _iterate_by_page(org_name, repo_name): _iterate_by_page(org_name, repo_name) # convert unix timestamp to Beijing datetime - bj_start_datetime = datetime.fromtimestamp(start_datetime.timestamp(), tz=pytz.timezone('Asia/Shanghai')) + bj_start_datetime = datetime.fromtimestamp(start_datetime.timestamp(), tz=pytz.timezone("Asia/Shanghai")) bj_start_datetime_str = datetime2str(bj_start_datetime) contribution_list = counter.to_sorted_list() @@ -452,7 +457,7 @@ def _iterate_by_page(org_name, repo_name): if len(author_list) > 0: xlabel = f"Number of Pull Requests (since {bj_start_datetime_str})" ylabel = "Contributor" - title = 'Active Contributor Leaderboard' + title = "Active Contributor Leaderboard" plot_bar_chart(num_commit_list, author_list, xlabel=xlabel, ylabel=ylabel, title=title, output_path=output_path) return True else: @@ -468,14 +473,14 @@ def upload_image_to_lark(lark_tenant_token: str, image_path: str) -> str: image_path (str): the path to the image to be uploaded """ url = "https://open.feishu.cn/open-apis/im/v1/images" - form = {'image_type': 'message', 'image': (open(image_path, 'rb'))} # 需要替换具体的path + form = {"image_type": "message", "image": (open(image_path, "rb"))} # 需要替换具体的path multi_form = MultipartEncoder(form) headers = { - 'Authorization': f'Bearer {lark_tenant_token}', ## 获取tenant_access_token, 需要替换为实际的token + "Authorization": f"Bearer {lark_tenant_token}", ## 获取tenant_access_token, 需要替换为实际的token } - headers['Content-Type'] = multi_form.content_type + headers["Content-Type"] = multi_form.content_type response = requests.request("POST", url, headers=headers, data=multi_form).json() - return response['data']['image_key'] + return response["data"]["image_key"] def generate_lark_tenant_access_token(app_id: str, app_secret: str) -> str: @@ -486,10 +491,10 @@ def generate_lark_tenant_access_token(app_id: str, app_secret: str) -> str: app_id (str): Lark app id app_secret (str): Lark app secret """ - url = 'https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal' - data = {'app_id': app_id, 'app_secret': app_secret} + url = "https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal" + data = {"app_id": app_id, "app_secret": app_secret} response = requests.post(url, json=data).json() - return response['tenant_access_token'] + return response["tenant_access_token"] def send_image_to_lark(image_key: str, webhook_url: str) -> None: @@ -516,10 +521,10 @@ def send_message_to_lark(message: str, webhook_url: str): requests.post(webhook_url, json=data) -if __name__ == '__main__': - GITHUB_TOKEN = os.environ['GITHUB_TOKEN'] - CONTRIBUTOR_IMAGE_PATH = 'contributor_leaderboard.png' - USER_ENGAGEMENT_IMAGE_PATH = 'engagement_leaderboard.png' +if __name__ == "__main__": + GITHUB_TOKEN = os.environ["GITHUB_TOKEN"] + CONTRIBUTOR_IMAGE_PATH = "contributor_leaderboard.png" + USER_ENGAGEMENT_IMAGE_PATH = "engagement_leaderboard.png" ORG_NAME = "hpcaitech" # get all open source repositories @@ -527,17 +532,19 @@ def send_message_to_lark(message: str, webhook_url: str): # generate images contrib_success = generate_contributor_leaderboard_image(GITHUB_TOKEN, ORG_NAME, REPO_LIST, CONTRIBUTOR_IMAGE_PATH) - engagement_success = generate_user_engagement_leaderboard_image(GITHUB_TOKEN, ORG_NAME, REPO_LIST, USER_ENGAGEMENT_IMAGE_PATH) + engagement_success = generate_user_engagement_leaderboard_image( + GITHUB_TOKEN, ORG_NAME, REPO_LIST, USER_ENGAGEMENT_IMAGE_PATH + ) # upload images - APP_ID = os.environ['LARK_APP_ID'] - APP_SECRET = os.environ['LARK_APP_SECRET'] + APP_ID = os.environ["LARK_APP_ID"] + APP_SECRET = os.environ["LARK_APP_SECRET"] LARK_TENANT_TOKEN = generate_lark_tenant_access_token(app_id=APP_ID, app_secret=APP_SECRET) contributor_image_key = upload_image_to_lark(LARK_TENANT_TOKEN, CONTRIBUTOR_IMAGE_PATH) user_engagement_image_key = upload_image_to_lark(LARK_TENANT_TOKEN, USER_ENGAGEMENT_IMAGE_PATH) # send message to lark - LARK_WEBHOOK_URL = os.environ['LARK_WEBHOOK_URL'] + LARK_WEBHOOK_URL = os.environ["LARK_WEBHOOK_URL"] message = """本周的社区榜单出炉啦！ 1. 开发贡献者榜单 2. 用户互动榜单 diff --git a/.github/workflows/scripts/generate_release_draft.py b/.github/workflows/scripts/generate_release_draft.py index dc592e4c977b..7374481005ef 100644 --- a/.github/workflows/scripts/generate_release_draft.py +++ b/.github/workflows/scripts/generate_release_draft.py @@ -7,27 +7,27 @@ import requests -COMMIT_API = 'https://api.github.com/repos/hpcaitech/ColossalAI/commits' -TAGS_API = 'https://api.github.com/repos/hpcaitech/ColossalAI/tags' +COMMIT_API = "https://api.github.com/repos/hpcaitech/ColossalAI/commits" +TAGS_API = "https://api.github.com/repos/hpcaitech/ColossalAI/tags" def parse_args(): parser = argparse.ArgumentParser() - parser.add_argument('--out', type=str, help='output path for the release draft', required=True) - parser.add_argument('--version', type=str, help='current version to release', required=True) + parser.add_argument("--out", type=str, help="output path for the release draft", required=True) + parser.add_argument("--version", type=str, help="current version to release", required=True) return parser.parse_args() def get_latest_tag_commit(headers=None): res = requests.get(url=TAGS_API, headers=headers) data = res.json() - commit_hash = data[0]['commit']['sha'] - version = data[0]['name'] + commit_hash = data[0]["commit"]["sha"] + version = data[0]["name"] return commit_hash, version def get_commit_info(commit_hash, headers=None): - api = f'{COMMIT_API}/{commit_hash}' + api = f"{COMMIT_API}/{commit_hash}" res = requests.get(url=api, headers=headers) return res.json() @@ -37,7 +37,7 @@ def get_all_commit_info(since, headers=None): results = [] while True: - api = f'{COMMIT_API}?since={since}&per_page=100&page={page}' + api = f"{COMMIT_API}?since={since}&per_page=100&page={page}" resp = requests.get(url=api, headers=headers) data = resp.json() @@ -53,21 +53,21 @@ def get_all_commit_info(since, headers=None): def collate_release_info(commit_info_list): results = dict() - pattern = pattern = r'\[.*\]' + pattern = pattern = r"\[.*\]" for commit_info in commit_info_list: - author = commit_info['commit']['author']['name'] + author = commit_info["commit"]["author"]["name"] try: - author_url = commit_info['author']['url'] + author_url = commit_info["author"]["url"] except: # author can be None author_url = None - msg = commit_info['commit']['message'] + msg = commit_info["commit"]["message"] match = re.search(pattern, msg) if match: - tag = match.group().lstrip('[').rstrip(']').capitalize() + tag = match.group().lstrip("[").rstrip("]").capitalize() if tag not in results: results[tag] = [] results[tag].append((msg, author, author_url)) @@ -89,42 +89,43 @@ def generate_release_post_markdown(current_version, last_version, release_info): for msg, author, author_url in v: # only keep the first line - msg = msg.split('\n')[0] + msg = msg.split("\n")[0] if author_url: - item = f'{msg} by [{author}]({author_url})\n' + item = f"{msg} by [{author}]({author_url})\n" else: - item = f'{msg} by {author}\n' - text.append(f'- {item}') + item = f"{msg} by {author}\n" + text.append(f"- {item}") - text.append('\n') + text.append("\n") # add full change log text.append( - f'**Full Changelog**: https://github.com/hpcaitech/ColossalAI/compare/{current_version}...{last_version}') + f"**Full Changelog**: https://github.com/hpcaitech/ColossalAI/compare/{current_version}...{last_version}" + ) return text -if __name__ == '__main__': +if __name__ == "__main__": args = parse_args() - token = os.environ['GITHUB_API_TOKEN'] - headers = {'Authorization': token} + token = os.environ["GITHUB_API_TOKEN"] + headers = {"Authorization": token} # get previous release tag last_release_commit, last_version = get_latest_tag_commit(headers) last_release_commit_info = get_commit_info(last_release_commit, headers=headers) - last_release_date = last_release_commit_info['commit']['author']['date'] + last_release_date = last_release_commit_info["commit"]["author"]["date"] # get the commits since last release commit_info = get_all_commit_info(since=last_release_date, headers=headers) - commit_info = commit_info[:-1] # remove the release commit + commit_info = commit_info[:-1] # remove the release commit # collate into markdown release_info = collate_release_info(commit_info) markdown_text = generate_release_post_markdown(args.version, last_version, release_info) # write into a file - with open(args.out, 'w') as f: + with open(args.out, "w") as f: for line in markdown_text: f.write(line) diff --git a/.github/workflows/scripts/send_message_to_lark.py b/.github/workflows/scripts/send_message_to_lark.py index a113327a786e..bc005d93c3f5 100644 --- a/.github/workflows/scripts/send_message_to_lark.py +++ b/.github/workflows/scripts/send_message_to_lark.py @@ -5,8 +5,8 @@ def parse_args(): parser = argparse.ArgumentParser() - parser.add_argument('-m', '--message', type=str) - parser.add_argument('-u', '--url', type=str) + parser.add_argument("-m", "--message", type=str) + parser.add_argument("-u", "--url", type=str) return parser.parse_args() @@ -15,6 +15,6 @@ def send_message_to_lark(message, webhook_url): requests.post(webhook_url, json=data) -if __name__ == '__main__': +if __name__ == "__main__": args = parse_args() send_message_to_lark(args.message, args.url) diff --git a/.isort.cfg b/.isort.cfg index 090aa28e39f3..ccbf575fdbfa 100644 --- a/.isort.cfg +++ b/.isort.cfg @@ -3,3 +3,5 @@ line_length = 120 multi_line_output=3 include_trailing_comma = true ignore_comments = true +profile = black +honor_noqa = true diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml index 725d266375ef..9871e1184462 100644 --- a/.pre-commit-config.yaml +++ b/.pre-commit-config.yaml @@ -1,23 +1,31 @@ repos: + - repo: https://github.com/PyCQA/autoflake + rev: v2.2.1 + hooks: + - id: autoflake + name: autoflake (python) + args: ['--in-place', '--remove-unused-variables', '--remove-all-unused-imports', '--ignore-init-module-imports'] + - repo: https://github.com/pycqa/isort rev: 5.12.0 hooks: - id: isort name: sort all imports (python) - - repo: https://github.com/pre-commit/mirrors-yapf - rev: v0.32.0 + - repo: https://github.com/psf/black-pre-commit-mirror + rev: 23.9.1 hooks: - - id: yapf - name: yapf formatter - args: ['--style=.style.yapf', '--parallel', '--in-place'] + - id: black + name: black formatter + args: ['--line-length=120', '--target-version=py37', '--target-version=py38', '--target-version=py39','--target-version=py310'] - repo: https://github.com/pre-commit/mirrors-clang-format rev: v13.0.1 hooks: - id: clang-format name: clang formatter + types_or: [c++, c] - repo: https://github.com/pre-commit/pre-commit-hooks rev: v4.3.0 diff --git a/.style.yapf b/.style.yapf deleted file mode 100644 index 05be0dc6a3a5..000000000000 --- a/.style.yapf +++ /dev/null @@ -1,5 +0,0 @@ -[style] -based_on_style = google -spaces_before_comment = 4 -split_before_logical_operator = true -column_limit = 120 diff --git a/LICENSE b/LICENSE index c7a5bb16880e..59d456c5b8a1 100644 --- a/LICENSE +++ b/LICENSE @@ -396,3 +396,84 @@ Copyright 2021- HPC-AI Technology Inc. All rights reserved. CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + + ---------------- LICENSE FOR VLLM TEAM ---------------- + + from VLLM TEAM: + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + https://github.com/vllm-project/vllm/blob/main/LICENSE + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. + + ---------------- LICENSE FOR LIGHTLLM TEAM ---------------- + + from LIGHTLLM TEAM: + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + https://github.com/ModelTC/lightllm/blob/main/LICENSE + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. + ---------------- LICENSE FOR AutoGPTQ ---------------- + + From AutoGPTQ: + + MIT License + + Copyright (c) 2023 潘其威(William) + + Permission is hereby granted, free of charge, to any person obtaining a copy + of this software and associated documentation files (the "Software"), to deal + in the Software without restriction, including without limitation the rights + to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + copies of the Software, and to permit persons to whom the Software is + furnished to do so, subject to the following conditions: + + The above copyright notice and this permission notice shall be included in all + copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. + + ---------------- LICENSE FOR exllama ---------------- + + From exllama: + + MIT License + + Permission is hereby granted, free of charge, to any person obtaining a copy + of this software and associated documentation files (the "Software"), to deal + in the Software without restriction, including without limitation the rights + to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + copies of the Software, and to permit persons to whom the Software is + furnished to do so, subject to the following conditions: + + The above copyright notice and this permission notice shall be included in all + copies or substantial portions of the Software. + + THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE + SOFTWARE. diff --git a/README.md b/README.md index 0ddcdab741a4..68f8b5dcabf7 100644 --- a/README.md +++ b/README.md @@ -4,7 +4,7 @@ [![logo](https://raw.githubusercontent.com/hpcaitech/public_assets/main/colossalai/img/colossal-ai_logo_vertical.png)](https://www.colossalai.org/) Colossal-AI: Making large AI models cheaper, faster, and more accessible - +test

Paper | Documentation | Examples | @@ -16,7 +16,7 @@ [![Documentation](https://readthedocs.org/projects/colossalai/badge/?version=latest)](https://colossalai.readthedocs.io/en/latest/?badge=latest) [![CodeFactor](https://www.codefactor.io/repository/github/hpcaitech/colossalai/badge)](https://www.codefactor.io/repository/github/hpcaitech/colossalai) [![HuggingFace badge](https://img.shields.io/badge/%F0%9F%A4%97HuggingFace-Join-yellow)](https://huggingface.co/hpcai-tech) - [![slack badge](https://img.shields.io/badge/Slack-join-blueviolet?logo=slack&)](https://join.slack.com/t/colossalaiworkspace/shared_invite/zt-z7b26eeb-CBp7jouvu~r0~lcFzX832w) + [![slack badge](https://img.shields.io/badge/Slack-join-blueviolet?logo=slack&)](https://github.com/hpcaitech/public_assets/tree/main/colossalai/contact/slack) [![WeChat badge](https://img.shields.io/badge/微信-加入-green?logo=wechat&)](https://raw.githubusercontent.com/hpcaitech/public_assets/main/colossalai/img/WeChat.png) @@ -25,6 +25,7 @@ ## Latest News +* [2023/09] [One Half-Day of Training Using a Few Hundred Dollars Yields Similar Results to Mainstream Large Models, Open-Source and Commercial-Free Domain-Specific Llm Solution](https://www.hpc-ai.tech/blog/one-half-day-of-training-using-a-few-hundred-dollars-yields-similar-results-to-mainstream-large-models-open-source-and-commercial-free-domain-specific-llm-solution) * [2023/09] [70 Billion Parameter LLaMA2 Model Training Accelerated by 195%](https://www.hpc-ai.tech/blog/70b-llama2-training) * [2023/07] [HPC-AI Tech Raises 22 Million USD in Series A Funding](https://www.hpc-ai.tech/blog/hpc-ai-tech-raises-22-million-usd-in-series-a-funding-to-fuel-team-expansion-and-business-growth) * [2023/07] [65B Model Pretraining Accelerated by 38%, Best Practices for Building LLaMA-Like Base Models Open-Source](https://www.hpc-ai.tech/blog/large-model-pretraining) @@ -33,8 +34,6 @@ * [2023/03] [AWS and Google Fund Colossal-AI with Startup Cloud Programs](https://www.hpc-ai.tech/blog/aws-and-google-fund-colossal-ai-with-startup-cloud-programs) * [2023/02] [Open Source Solution Replicates ChatGPT Training Process! Ready to go with only 1.6GB GPU Memory](https://www.hpc-ai.tech/blog/colossal-ai-chatgpt) * [2023/01] [Hardware Savings Up to 46 Times for AIGC and Automatic Parallelism](https://medium.com/pytorch/latest-colossal-ai-boasts-novel-automatic-parallelism-and-offers-savings-up-to-46x-for-stable-1453b48f3f02) -* [2022/11] [Diffusion Pretraining and Hardware Fine-Tuning Can Be Almost 7X Cheaper](https://www.hpc-ai.tech/blog/diffusion-pretraining-and-hardware-fine-tuning-can-be-almost-7x-cheaper) -* [2022/10] [Use a Laptop to Analyze 90% of Proteins, With a Single-GPU Inference Sequence Exceeding 10,000](https://www.hpc-ai.tech/blog/use-a-laptop-to-analyze-90-of-proteins-with-a-single-gpu-inference-sequence-exceeding) ## Table of Contents