Merge branch 'master' into docs/csv-logger

Author: bhimrazy

.azure/gpu-tests-fabric.yml

@@ -130,7 +130,7 @@ jobs:
       - bash: |
           set -e
           extra=$(python -c "print({'lightning': 'fabric-'}.get('$(PACKAGE_NAME)', ''))")
-          pip install -e ".[${extra}dev]" -U --extra-index-url="${TORCH_URL}"
+          pip install -e ".[${extra}dev]" -U --upgrade-strategy=eager --extra-index-url="${TORCH_URL}"
         displayName: "Install package & dependencies"
 
       - bash: |

.azure/gpu-tests-pytorch.yml

@@ -134,7 +134,7 @@ jobs:
       - bash: |
           set -e
           extra=$(python -c "print({'lightning': 'pytorch-'}.get('$(PACKAGE_NAME)', ''))")
-          pip install -e ".[${extra}dev]" -U --extra-index-url="${TORCH_URL}"
+          pip install -e ".[${extra}dev]" -U --upgrade-strategy=eager --extra-index-url="${TORCH_URL}"
         displayName: "Install package & dependencies"
 
       - bash: pip uninstall -y lightning

.github/workflows/ci-tests-fabric.yml

@@ -138,7 +138,8 @@ jobs:
       - name: Install package & dependencies
         timeout-minutes: 20
         run: |
-          pip install -e ".[${EXTRA_PREFIX}test,${EXTRA_PREFIX}strategies]" -U --prefer-binary \
+          pip install -e ".[${EXTRA_PREFIX}test,${EXTRA_PREFIX}strategies]" \
+            -U --upgrade-strategy=eager --prefer-binary \
             --extra-index-url="${TORCH_URL}" --find-links="${PYPI_CACHE_DIR}"
           pip list
       - name: Dump handy wheels

.github/workflows/ci-tests-pytorch.yml

@@ -136,7 +136,8 @@ jobs:
       - name: Install package & dependencies
         timeout-minutes: 20
         run: |
-          pip install ".[${EXTRA_PREFIX}extra,${EXTRA_PREFIX}test,${EXTRA_PREFIX}strategies]" -U --prefer-binary \
+          pip install ".[${EXTRA_PREFIX}extra,${EXTRA_PREFIX}test,${EXTRA_PREFIX}strategies]" \
+            -U --upgrade-strategy=eager --prefer-binary \
             -r requirements/_integrations/accelerators.txt \
             --extra-index-url="${TORCH_URL}" --find-links="${PYPI_CACHE_DIR}"
           pip list

.github/workflows/probot-check-group.yml

@@ -14,7 +14,7 @@ jobs:
     if: github.event.pull_request.draft == false
     timeout-minutes: 61 # in case something is wrong with the internal timeout
     steps:
-      - uses: Lightning-AI/probot@v5.4
+      - uses: Lightning-AI/probot@v5.5
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         with:

docs/source-pytorch/advanced/transfer_learning.rst

@@ -32,7 +32,7 @@ Let's use the `AutoEncoder` as a feature extractor in a separate model.
     class CIFAR10Classifier(LightningModule):
         def __init__(self):
             # init the pretrained LightningModule
-            self.feature_extractor = AutoEncoder.load_from_checkpoint(PATH)
+            self.feature_extractor = AutoEncoder.load_from_checkpoint(PATH).encoder
             self.feature_extractor.freeze()
 
             # the autoencoder outputs a 100-dim representation and CIFAR-10 has 10 classes

docs/source-pytorch/common/precision_intermediate.rst

@@ -165,7 +165,7 @@ Under the hood, we use `transformer_engine.pytorch.fp8_autocast <https://docs.nv
 Quantization via Bitsandbytes
 *****************************
 
-`bitsandbytes <https://github.com/TimDettmers/bitsandbytes>`__ (BNB) is a library that supports quantizing :class:`torch.nn.Linear` weights.
+`bitsandbytes <https://github.com/bitsandbytes-foundation/bitsandbytes>`__ (BNB) is a library that supports quantizing :class:`torch.nn.Linear` weights.
 
 Both 4-bit (`paper reference <https://arxiv.org/abs/2305.14314v1>`__) and 8-bit (`paper reference <https://arxiv.org/abs/2110.02861>`__) quantization is supported.
 Specifically, we support the following modes:

docs/source-pytorch/glossary/index.rst

@@ -209,7 +209,7 @@ Glossary
 
 .. displayitem::
    :header: LightningModule
-   :description: A base class organizug your neural network module
+   :description: A base class organizing your neural network module
    :col_css: col-md-12
    :button_link: ../common/lightning_module.html
    :height: 100

requirements/fabric/strategies.txt

@@ -6,5 +6,4 @@
 # note: is a bug around 0.10 with `MPS_Accelerator must implement all abstract methods`
 #  shall be resolved by https://github.com/microsoft/DeepSpeed/issues/4372
 deepspeed >=0.9.3, <=0.9.3; platform_system != "Windows" and platform_system != "Darwin"  # strict
-# skip bitsandbytes==0.46, due to ValueError: too many values to unpack (expected 2)
-bitsandbytes >=0.45.2,!=0.46,<0.47.0; platform_system != "Darwin"
+bitsandbytes >=0.45.2,<0.47.0; platform_system != "Darwin"

requirements/pytorch/extra.txt

@@ -8,5 +8,4 @@ hydra-core >=1.2.0, <1.4.0
 jsonargparse[signatures,jsonnet] >=4.39.0, <4.41.0
 rich >=12.3.0, <14.1.0
 tensorboardX >=2.2, <2.7.0  # min version is set by torch.onnx missing attribute
-# skip bitsandbytes==0.46, due to ValueError: too many values to unpack (expected 2)
-bitsandbytes >=0.45.2,!=0.46,<0.47.0; platform_system != "Darwin"
+bitsandbytes >=0.45.2,<0.47.0; platform_system != "Darwin"