🛠️ Check comptability of transformers with molformer + tests

.github/workflows/python-package.yml

@@ -5,38 +5,38 @@ name: Python package
 
 on:
   push:
-    branches: [ "main" ]
+    branches: ["main"]
   pull_request:
-    branches: [ "main" ]
+    branches: ["main"]
 
 jobs:
   build:
-
     runs-on: ubuntu-latest
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9", "3.11", "3.13"]
+        python-version: ["3.10", "3.12", "3.13"]
 
     steps:
-    - uses: actions/checkout@v6
-    - name: Set up Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v6
-      with:
-        python-version: ${{ matrix.python-version }}
-    - name: Install dependencies
-      run: |
-        python -m pip install --upgrade pip
-        python -m pip install flake8 pytest sentencepiece rdkit
-        python -m pip install git+https://github.com/novonordisk-research/pepfunn.git
-        python -m pip install .
-        if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
-    - name: Lint with flake8
-      run: |
-        # stop the build if there are Python syntax errors or undefined names
-        # flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
-        # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
-        flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
-    - name: Test with pytest
-      run: |
-        pytest
+      - uses: actions/checkout@v6
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v6
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install uv
+          uv pip install flake8 pytest sentencepiece rdkit --system
+          uv pip install git+https://github.com/novonordisk-research/pepfunn.git --system
+          uv pip install .  "transformers<5" --system
+          if [ -f requirements.txt ]; then pip install -r requirements.txt; fi
+      - name: Lint with flake8
+        run: |
+          # stop the build if there are Python syntax errors or undefined names
+          # flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
+          # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
+          flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
+      - name: Test with pytest
+        run: |
+          pytest

autopeptideml/reps/lms.py

@@ -4,9 +4,11 @@
 os.environ['TOKENIZERS_PARALLELISM'] = 'False'
 
 import numpy as np
+from packaging.version import parse as V
 import torch
 import transformers
 from transformers import AutoModel, AutoTokenizer, T5Tokenizer, T5EncoderModel
+
 from typing import *
 
 from .engine import RepEngineBase
@@ -200,7 +202,8 @@ def _load_model(self, model: str):
         elif 'ankh' in model.lower():
             self.lab = 'ElnaggarLab'
         elif 'molformer' in model.lower():
-            print("Warning: Molformer does not support transformers>=5.0.0, we recommend using an earlier version, e.g., transformers==4.41.2")
+            if V(transformers.__version__) >= V('5.0.0'):
+                print("Warning: Molformer does not support transformers>=5.0.0, we recommend using an earlier version, e.g., transformers==4.41.2")
             self.lab = 'ibm'
         elif 'chemberta' in model.lower():
             self.lab = 'DeepChem'

tests/test_reps.py

@@ -53,6 +53,13 @@ def test_rostlab_family():
     assert np.array(a).shape == (1, re.dim())
 
 
+def test_molformer_family():
+    re = RepEngineLM('molformer-xl')
+    a = re.compute_reps(['C[C@H](N)C(=O)N[C@@H](CCCNC(=N)N)C(=O)N[C@H]'], batch_size=12)
+    assert re.dim() == 768
+    assert np.array(a).shape == (1, re.dim())
+
+
 if __name__ == '__main__':
     test_esm_family()
     print('ESM OK')