Avoid nan loss when there are labels with no samples in the training …

…data. (#12)
fcakyon · Dec 16, 2024 · 20f3779 · 20f3779
1 parent 1c71f91
commit 20f3779
Show file tree

Hide file tree

Showing 7 changed files with 73 additions and 46 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -11,39 +11,39 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        operating-system: [ubuntu-latest, windows-latest, macos-latest]
-        python-version: [3.7, 3.8, 3.9]
-        torch-version: [1.10.2, 1.11.0, 1.12.0]
+        os: [ubuntu-latest, windows-latest, macos-latest]
+        python-version: [3.9, "3.10"]
+        torch-version: [1.13.1, 2.5.1]
       fail-fast: false
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
 
       - name: Set up Python
-        uses: actions/setup-python@v2
+        uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
 
       - name: Restore Ubuntu cache
-        uses: actions/cache@v1
-        if: matrix.operating-system == 'ubuntu-latest'
+        uses: actions/cache@v4
+        if: matrix.os == 'ubuntu-latest'
         with:
           path: ~/.cache/pip
           key: ${{ matrix.os }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
           restore-keys: ${{ matrix.os }}-${{ matrix.python-version }}-
 
       - name: Restore MacOS cache
-        uses: actions/cache@v1
-        if: matrix.operating-system == 'macos-latest'
+        uses: actions/cache@v4
+        if: matrix.os == 'macos-latest'
         with:
           path: ~/Library/Caches/pip
           key: ${{ matrix.os }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
           restore-keys: ${{ matrix.os }}-${{ matrix.python-version }}-
 
       - name: Restore Windows cache
-        uses: actions/cache@v1
-        if: matrix.operating-system == 'windows-latest'
+        uses: actions/cache@v4
+        if: matrix.os == 'windows-latest'
         with:
           path: ~\AppData\Local\pip\Cache
           key: ${{ matrix.os }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
@@ -52,9 +52,14 @@ jobs:
       - name: Update pip
         run: python -m pip install --upgrade pip
 
+      - name: Install package in development mode
+        run: pip install -e .[dev]
+
+      - name: Show installed packages
+        run: pip list
+
       - name: Lint with flake8, black and isort
         run: |
-          pip install -e .[dev]
           # stop the build if there are Python syntax errors or undefined names
           flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
           black . --check --config pyproject.toml
@@ -66,17 +71,23 @@ jobs:
         run: >
           pip install numpy
 
-      - name: Install PyTorch on Linux and Windows
+      - name: Install PyTorch==1.13.1 on Linux and Windows
         if: >
-          matrix.operating-system == 'ubuntu-latest' ||
-          matrix.operating-system == 'windows-latest'
+          (matrix.os == 'ubuntu-latest' ||
+          matrix.os == 'windows-latest') &&
+          matrix.torch-version == '1.13.1'
         run: >
           pip install torch==${{ matrix.torch-version }}+cpu
           -f https://download.pytorch.org/whl/torch_stable.html
 
-      - name: Install PyTorch on MacOS
-        if: matrix.operating-system == 'macos-latest'
-        run: pip install torch==${{ matrix.torch-version }}
+      - name: Install PyTorch==2.5.1 on Linux and Windows
+        if: >
+          (matrix.os == 'ubuntu-latest' ||
+          matrix.os == 'windows-latest') &&
+          matrix.torch-version == '2.5.1'
+        run: >
+          pip install torch==${{ matrix.torch-version }}
+          -f https://download.pytorch.org/whl/torch_stable.html
 
       - name: Install balanced-loss package from local setup.py
         run: >

diff --git a/.github/workflows/package_testing.yml b/.github/workflows/package_testing.yml
@@ -10,39 +10,39 @@ jobs:
 
     strategy:
       matrix:
-        operating-system: [ubuntu-latest, windows-latest, macos-latest]
-        python-version: [3.7, 3.8, 3.9]
-        torch-version: [1.10.2, 1.11.0, 1.12.0]
+        os: [ubuntu-latest, windows-latest, macos-latest]
+        python-version: [3.9, "3.10"]
+        torch-version: [1.13.1, 2.5.1]
       fail-fast: false
 
     steps:
       - name: Checkout
-        uses: actions/checkout@v2
+        uses: actions/checkout@v4
 
       - name: Set up Python
-        uses: actions/setup-python@v2
+        uses: actions/setup-python@v5
         with:
           python-version: ${{ matrix.python-version }}
 
       - name: Restore Ubuntu cache
-        uses: actions/cache@v1
-        if: matrix.operating-system == 'ubuntu-latest'
+        uses: actions/cache@v4
+        if: matrix.os == 'ubuntu-latest'
         with:
           path: ~/.cache/pip
           key: ${{ matrix.os }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
           restore-keys: ${{ matrix.os }}-${{ matrix.python-version }}-
 
       - name: Restore MacOS cache
-        uses: actions/cache@v1
-        if: matrix.operating-system == 'macos-latest'
+        uses: actions/cache@v4
+        if: matrix.os == 'macos-latest'
         with:
           path: ~/Library/Caches/pip
           key: ${{ matrix.os }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
           restore-keys: ${{ matrix.os }}-${{ matrix.python-version }}-
 
       - name: Restore Windows cache
-        uses: actions/cache@v1
-        if: matrix.operating-system == 'windows-latest'
+        uses: actions/cache@v4
+        if: matrix.os == 'windows-latest'
         with:
           path: ~\AppData\Local\pip\Cache
           key: ${{ matrix.os }}-${{ matrix.python-version }}-${{ hashFiles('**/setup.py')}}
@@ -55,16 +55,26 @@ jobs:
         run: >
           pip install numpy
 
-      - name: Install PyTorch on Linux and Windows
+      - name: Install PyTorch==1.13.1 on Linux and Windows
         if: >
-          matrix.operating-system == 'ubuntu-latest' ||
-          matrix.operating-system == 'windows-latest'
+          (matrix.os == 'ubuntu-latest' ||
+          matrix.os == 'windows-latest') &&
+          matrix.torch-version == '1.13.1'
         run: >
           pip install torch==${{ matrix.torch-version }}+cpu
           -f https://download.pytorch.org/whl/torch_stable.html
 
+      - name: Install PyTorch==2.5.1 on Linux and Windows
+        if: >
+          (matrix.os == 'ubuntu-latest' ||
+          matrix.os == 'windows-latest') &&
+          matrix.torch-version == '2.5.1'
+        run: >
+          pip install torch==${{ matrix.torch-version }}
+          -f https://download.pytorch.org/whl/torch_stable.html
+
       - name: Install PyTorch on MacOS
-        if: matrix.operating-system == 'macos-latest'
+        if: matrix.os == 'macos-latest' 
         run: pip install torch==${{ matrix.torch-version }}
 
       - name: Install latest balanced-loss package
@@ -74,4 +84,3 @@ jobs:
       - name: Unittest balanced-loss
         run: |
           python -m unittest
-
diff --git a/.github/workflows/publish_pypi.yml b/.github/workflows/publish_pypi.yml
@@ -9,9 +9,9 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-    - uses: actions/checkout@v2
+    - uses: actions/checkout@v4
     - name: Set up Python
-      uses: actions/setup-python@v2
+      uses: actions/setup-python@v5
       with:
         python-version: '3.x'
     - name: Install dependencies

diff --git a/README.md b/README.md
@@ -24,7 +24,6 @@ When training dataset labels are imbalanced, one thing to do is to balance the l
 
 ![alt-text](https://user-images.githubusercontent.com/34196005/180266198-e27d8cba-f5e1-49ca-9f82-d8656333e3c4.png)
 
-
 ## Installation
 
 ```bash
@@ -134,6 +133,7 @@ What is the difference between this repo and vandit15's?
 - This repo implements loss functions as `torch.nn.Module`
 - In addition to class balanced losses, this repo also supports the standard versions of the cross entropy/focal loss etc. over the same API
 - All typos and errors in vandit15's source are fixed
+- Continiously tested on PyTorch 1.13.1 and 2.5.1
 
 ## References
 

diff --git a/balanced_loss/__init__.py b/balanced_loss/__init__.py
@@ -1,3 +1,3 @@
 from .losses import Loss
 
-__version__ = "0.1.0"
+__version__ = "0.1.1"
diff --git a/balanced_loss/losses.py b/balanced_loss/losses.py
@@ -44,6 +44,7 @@ def __init__(
         fl_gamma=2,
         samples_per_class=None,
         class_balanced=False,
+        safe: bool = False,
     ):
         """
         Compute the Class Balanced Loss between `logits` and the ground truth `labels`.
@@ -60,6 +61,7 @@ def __init__(
             samples_per_class: A python list of size [num_classes].
                 Required if class_balance is True.
             class_balanced: bool. Whether to use class balanced loss.
+            safe: bool. Whether to allow labels with no samples.
         Returns:
             Loss instance
         """
@@ -73,12 +75,9 @@ def __init__(
         self.fl_gamma = fl_gamma
         self.samples_per_class = samples_per_class
         self.class_balanced = class_balanced
+        self.safe = safe
 
-    def forward(
-        self,
-        logits: torch.tensor,
-        labels: torch.tensor,
-    ):
+    def forward(self, logits: torch.tensor, labels: torch.tensor):
         """
         Compute the Class Balanced Loss between `logits` and the ground truth `labels`.
         Class Balanced Loss: ((1-beta)/(1-beta^n))*Loss(labels, logits)
@@ -97,8 +96,16 @@ def forward(
 
         if self.class_balanced:
             effective_num = 1.0 - np.power(self.beta, self.samples_per_class)
+            # Avoid division by 0 error for test cases without all labels present.
+            if self.safe:
+                effective_num_classes = np.sum(effective_num != 0)
+                effective_num[effective_num == 0] = np.inf
+
+            else:
+                effective_num_classes = num_classes
+
             weights = (1.0 - self.beta) / np.array(effective_num)
-            weights = weights / np.sum(weights) * num_classes
+            weights = weights / np.sum(weights) * effective_num_classes
             weights = torch.tensor(weights, device=logits.device).float()
 
             if self.loss_type != "cross_entropy":

diff --git a/setup.py b/setup.py
@@ -37,7 +37,7 @@ def get_version():
 setuptools.setup(
     name="balanced-loss",
     version=get_version(),
-    author="",
+    author="fcakyon",
     license="MIT",
     description="Easy to use class-balanced cross-entropy and focal loss implementation for Pytorch.",
     long_description=get_long_description(),
@@ -54,9 +54,9 @@ def get_version():
         "Intended Audience :: Developers",
         "Intended Audience :: Science/Research",
         "Programming Language :: Python :: 3",
-        "Programming Language :: Python :: 3.7",
         "Programming Language :: Python :: 3.8",
         "Programming Language :: Python :: 3.9",
+        "Programming Language :: Python :: 3.10",
         "Topic :: Software Development :: Libraries",
         "Topic :: Software Development :: Libraries :: Python Modules",
         "Topic :: Education",