File: pre-tokenizer-hashes.yml

package info (click to toggle)
llama.cpp 7965%2Bdfsg-1
  • links: PTS, VCS
  • area: main
  • in suites: sid
  • size: 75,824 kB
  • sloc: cpp: 348,634; ansic: 49,792; python: 33,481; lisp: 10,836; sh: 6,289; objc: 1,392; javascript: 924; xml: 384; makefile: 233
file content (45 lines) | stat: -rw-r--r-- 1,606 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
name: Check Pre-Tokenizer Hashes

on:
    push:
        paths:
            - 'convert_hf_to_gguf.py'
            - 'convert_hf_to_gguf_update.py'
    pull_request:
        paths:
            - 'convert_hf_to_gguf.py'
            - 'convert_hf_to_gguf_update.py'

jobs:
    pre-tokenizer-hashes:
        runs-on: ubuntu-slim

        steps:
        - name: Checkout repository
          uses: actions/checkout@v6

        - name: Set up Python
          uses: actions/setup-python@v6
          with:
              python-version: '3.11'

        - name: Install Python dependencies
          run: |
              python3 -m venv .venv
              .venv/bin/pip install -r requirements/requirements-convert_hf_to_gguf_update.txt

        - name: Update pre-tokenizer hashes
          run: |
              cp convert_hf_to_gguf.py /tmp
              .venv/bin/python convert_hf_to_gguf_update.py --check-missing

        - name: Check if committed pre-tokenizer hashes matches generated version
          run: |
              if ! diff -q convert_hf_to_gguf.py /tmp/convert_hf_to_gguf.py; then
                  echo "Model pre-tokenizer hashes (in convert_hf_to_gguf.py) do not match generated hashes (from convert_hf_to_gguf_update.py)."
                  echo "To fix: run ./convert_hf_to_gguf_update.py and commit the updated convert_hf_to_gguf.py along with your changes"
                  echo "Differences found:"
                  diff convert_hf_to_gguf.py /tmp/convert_hf_to_gguf.py || true
                  exit 1
              fi
              echo "Model pre-tokenizer hashes are up to date."