Merge pull request #26 from NVIDIA-NeMo/pablo-garay/rufF_linting

pablo-garay · web-flow · commit 3edd04d17c28 · 2025-11-05T11:34:28.000-08:00
feat: ruff for linting
diff --git a/.github/workflows/changelog-build.yml b/.github/workflows/changelog-build.yml
@@ -16,7 +16,7 @@ on:
         type: string
         required: false
         default: ''
-    
+
 jobs:
   changelog:
     runs-on: ubuntu-latest
@@ -41,7 +41,7 @@ jobs:
           toTag: ${{ inputs.release-branch }}
 
       - name: Update changelog file
-        env: 
+        env:
           RELEASE_BRANCH: ${{ inputs.release-branch }}
           CHANGELOG: ${{ steps.github_tag.outputs.changelog }}
           MAIN_CONTENT: ${{ inputs.changelog-main-content }}
diff --git a/.github/workflows/cicd-approve-test-queue.yml b/.github/workflows/cicd-approve-test-queue.yml
@@ -150,4 +150,3 @@ jobs:
             -H 'Content-type: application/json' \
             --data "{\"text\":\":robot_joy: <https://github.com/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}|Test-queue-approval-bot workflow> failed. Please review manually.\n\ncc ${SLACK_WEBHOOK_ADMIN}\"}" \
             $SLACK_WEBHOOK
-
diff --git a/.github/workflows/ruff.yml b/.github/workflows/ruff.yml
@@ -0,0 +1,25 @@
+# Copyright (c) 2025, NVIDIA CORPORATION.  All rights reserved.
+#
+# 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
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# 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.
+
+name: Ruff Linter
+on: pull_request
+jobs:
+  ruff:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: astral-sh/ruff-action@v3
+        with:
+          version: 0.11.4
+          args: "check --output-format=github"
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -12,40 +12,53 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+default_language_version:
+  python: python3
+
+ci:
+  autofix_prs: false
+  autoupdate_commit_msg: '[pre-commit.ci] pre-commit suggestions'
+  autoupdate_schedule: quarterly
+
+default_install_hook_types: [pre-commit, commit-msg]
+
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v5.0.0
     hooks:
-    - id: end-of-file-fixer
-      # only include python files
-      files: \.py$
-    - id: trailing-whitespace
-      # only include python files
-      files: \.py$
+      - id: check-added-large-files
+        args: ['--maxkb=1000']
+      - id: check-case-conflict
+      - id: check-yaml
+      - id: detect-private-key
+      - id: end-of-file-fixer
+        exclude: docs/
+      - id: requirements-txt-fixer
+      - id: trailing-whitespace
 
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: "v0.9.9" # Use the appropriate version
+    rev: v0.11.4
+    hooks:
+      # Run the linter
+      - id: ruff
+        args: [--fix]  # Enables auto-fixing lint errors
+      # Run the formatter
+      - id: ruff-format
+
+  - repo: local
     hooks:
-    - id: ruff
-      args: ["--fix"]
-    - id: ruff
-      args: ["check", "--select", "I", "--fix"]
-    - id: ruff-format
+      - id: check-signoff
+        name: Check Signed-off-by
+        entry: bash -c 'if ! grep -q "Signed-off-by:" "$1"; then echo "❌ Commit message must be signed off. Use git commit -s to add it automatically."; exit 1; fi' --
+        language: system
+        always_run: true
+        stages: [commit-msg]
+        types: [text]
 
-  # - repo: local
-  #   hooks:
-  #     - id: no-underscore-md
-  #       name: "Disallow '_' in Markdown filenames"
-  #       language: system
-  #       entry: |
-  #         bash -c '
-  #           # Report the offending files
-  #           echo "[pre-commit] ERROR: Found Markdown files with underscores:" >&2
-  #           for file in "$@"; do
-  #             echo "  - $file (use hyphens instead)" >&2
-  #           done
-  #           exit 1
-  #         '
-  #       files: '.*\/[^\/]*_[^\/]*\.md$'
-  #       exclude: '^\.github/'
-  #       types: [file]
+      - id: pre-commit-reminder
+        name: Pre-commit Installation Reminder
+        entry: echo "⚠️  Remember to install pre-commit and hooks (pip install pre-commit && pre-commit install --install-hooks)"
+        language: system
+        always_run: true
+        pass_filenames: false
+        stages: [pre-commit]
diff --git a/README.md b/README.md
@@ -27,4 +27,4 @@ If you find our code useful, please consider citing the following papers:
   journal={arXiv preprint arXiv:2501.03575},
   year={2025}
 }
-```
+```
diff --git a/dfm/examples/Automodel/finetune/wan2_1_t2v_flow.yaml b/dfm/examples/Automodel/finetune/wan2_1_t2v_flow.yaml
@@ -56,4 +56,4 @@ checkpoint:
   checkpoint_dir: /opt/DFM/wan_t2v_flow_outputs_base_recipe_fsdp_run_1/
   model_save_format: torch_save
   save_consolidated: false
-  restore_from: null
+  restore_from: null
diff --git a/dfm/examples/Automodel/finetune/wan2_1_t2v_flow_multinode.yaml b/dfm/examples/Automodel/finetune/wan2_1_t2v_flow_multinode.yaml
@@ -56,4 +56,4 @@ checkpoint:
   checkpoint_dir: /opt/DFM/wan_t2v_flow_outputs_base_recipe_multi_node_fsdp_run_3/
   model_save_format: torch_save
   save_consolidated: false
-  restore_from: null
+  restore_from: null
diff --git a/dfm/src/Automodel/README.md b/dfm/src/Automodel/README.md
@@ -1,4 +1,3 @@
 # Automodel
 
 Models implemented using Dtensors (NeMo Automodel)
-
diff --git a/dfm/src/megatron/README.md b/dfm/src/megatron/README.md
@@ -1,4 +1,3 @@
 # MBridge
 
 Megatron models including DiT and dLLM implementations.
-
diff --git a/dfm/src/megatron/base/README.md b/dfm/src/megatron/base/README.md
@@ -1,4 +1,3 @@
 # Base
 
 Base classes and interfaces for Megatron models.
-
diff --git a/dfm/src/megatron/model/README.md b/dfm/src/megatron/model/README.md
@@ -1,4 +1,3 @@
 # Model
 
 Model implementations for DiT and dLLM architectures.
-
diff --git a/examples/README.md b/examples/README.md
@@ -1,4 +1,3 @@
 # Examples
 
 Training and inference examples for dtensor and megatron models
-
diff --git a/examples/dtensor/README.md b/examples/dtensor/README.md
@@ -1,4 +1,3 @@
 # DTensor Models (NeMo Automodel)
 
 Examples using NeMo Automodel with distributed tensor parallelism.
-
diff --git a/examples/dtensor/configs/README.md b/examples/dtensor/configs/README.md
@@ -1,4 +1,3 @@
 # Configs
 
 Configuration files for various Wan model versions.
-
diff --git a/examples/dtensor/scripts/README.md b/examples/dtensor/scripts/README.md
@@ -1,4 +1,3 @@
 # Scripts
 
 Training scripts for pretraining and finetuning.
-
diff --git a/examples/megatron/README.md b/examples/megatron/README.md
@@ -1,4 +1,3 @@
 # Megatron Examples
 
 Recipes and configuration overrides for megatron training.
-
diff --git a/examples/megatron/override_configs/README.md b/examples/megatron/override_configs/README.md
@@ -1,4 +1,3 @@
 # Override Configs
 
 Parallelism configuration overrides for different CP/TP/SP sizes.
-
diff --git a/examples/megatron/recipe/README.md b/examples/megatron/recipe/README.md
@@ -1,4 +1,3 @@
 # Recipe
 
 Training recipes for Wan2.1 pretraining, finetuning, and weight verification.
-
diff --git a/nemo_vfm/diffusion/data/readme.rst b/nemo_vfm/diffusion/data/readme.rst
@@ -1,4 +1,4 @@
-Preparing Image / Video Megatron Energon WebDataset with Cosmos Tokenizer
+﻿Preparing Image / Video Megatron Energon WebDataset with Cosmos Tokenizer
 ===========================
 
 This script is an example on preparing a WebDataset for an image / video + text dataset using distributed processing with the Cosmos Tokenizer. It processes each sample by generating a **continuous** image / video latent using the Cosmos video tokenizer and a T5 embedding from the text caption. Then, the processed data is stored in a WebDataset-compatible format.
@@ -22,5 +22,5 @@ Usage
 
 3. **Running the Script**:
    To run the script on 8 GPUs, use the following command:
-   
+
    ``bash torchrun --nproc_per_node=8 nemo/collections/diffusion/data/prepare_energon_dataset.py``
diff --git a/nemo_vfm/diffusion/readme.rst b/nemo_vfm/diffusion/readme.rst
@@ -1,10 +1,10 @@
-Diffusion Training Framework
+﻿Diffusion Training Framework
 =============
 
 Overview
 --------
 
-The NeMo Diffusion Training Framework provides a scalable training platform for diffusion models with transformer backbones.  Our new features streamline the training process, allowing developers to efficiently train state-of-the-art models with ease. 
+The NeMo Diffusion Training Framework provides a scalable training platform for diffusion models with transformer backbones. Our new features streamline the training process, allowing developers to efficiently train state-of-the-art models with ease.
 
 
 Some of the features we currently support include:
@@ -58,7 +58,7 @@ Energon Dataloader for Webscale Dataloading
 Webscale Dataloading
 ^^^^^^^^^^^^^^^^^^^^
 
-Megatron-Energon is an optimized multi-modal dataloader for large-scale deep learning with Megatron. Energon allows for distributed loading of large training training data for multi-modal model training. Energon allows for blending many datasets together and distributing the dataloading workflow across multiple cluster nodes/processes while ensuring reproducibility and resumability. 
+Megatron-Energon is an optimized multi-modal dataloader for large-scale deep learning with Megatron. Energon allows for distributed loading of large training training data for multi-modal model training. Energon allows for blending many datasets together and distributing the dataloading workflow across multiple cluster nodes/processes while ensuring reproducibility and resumability.
 
 Dataloader Checkpointing
 ^^^^^^^^^^^^^^^^^^^^^^^^
@@ -84,7 +84,7 @@ Our dataloader provides support for mixed image-video training by using the NeMo
 
 Model and Data Parallelism
 --------------------------
-NeMo provides support for training models using tensor parallelism, sequence parallelism, pipeline parallelism, and context parallelism. To support pipeline parallelism with conditional diffusion training, we duplicate the conditional embeddings across the pipeline stages, and perform an all-reduce during the backward pass. This approach uses more compute, but it has a lower communication cost than sending the conditional embeddings through different pipeline stages. 
+NeMo provides support for training models using tensor parallelism, sequence parallelism, pipeline parallelism, and context parallelism. To support pipeline parallelism with conditional diffusion training, we duplicate the conditional embeddings across the pipeline stages, and perform an all-reduce during the backward pass. This approach uses more compute, but it has a lower communication cost than sending the conditional embeddings through different pipeline stages.
 
 .. image:: assets/pipeline_conditioning.png
    :alt: Conditioning mechanism for pipeline parallelism
@@ -123,7 +123,7 @@ this will generate a folder a tar files. .pth contains image/video latent repres
    ├── samples/sample_0001.pickle
    ├── samples/sample_0001.json
    └── ...
-   shard_001.tar   
+   shard_001.tar
 
 The following is a sample command to prepare prepare webdataset into energon dataset:
 
@@ -187,4 +187,4 @@ To launch training on multiple nodes using Slurm
 Citations
 ---------
 
-.. [1] William Peebles and Saining Xie, "Scalable Diffusion Models with Transformers," *arXiv preprint arXiv:2212.09748*, 2022.
+.. [1] William Peebles and Saining Xie, "Scalable Diffusion Models with Transformers," *arXiv preprint arXiv:2212.09748*, 2022.
diff --git a/nemo_vfm/diffusion/vae/readme.rst b/nemo_vfm/diffusion/vae/readme.rst
@@ -1,4 +1,4 @@
-============================
+﻿============================
 Pretraining Variational AutoEncoder
 ============================
 
@@ -64,7 +64,7 @@ Data Preparation
 3. then select dataset type `ImageWebdataset` and specify the type `jpg`. Below is an example of the interactive setup:
 
     .. code-block:: bash
-        
+
         Found 2925 tar files in total. The first and last ones are:
         - 000000.tar
         - 002924.tar
@@ -124,8 +124,3 @@ We provide a sample training script for launching multi-node training. Simply co
 
    bash nemo/collections/diffusion/vae/train_vae.sh \
    data.path=xxx
-
-
-
-
-
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/autoregressive/nemo/post_training/tokenizer/README.md b/nemo_vfm/physicalai/Cosmos/cosmos1/models/autoregressive/nemo/post_training/tokenizer/README.md
@@ -1,4 +1,4 @@
-# Cosmos Autoregressive-based World Foundation Models: Changing the Video Tokenizer
+﻿# Cosmos Autoregressive-based World Foundation Models: Changing the Video Tokenizer
 
 Learn how to post-train Cosmos Autoregressive-based World Foundation Models (WFMs) using the NVIDIA NeMo Framework when swapping out the original tokenizer for a new one. This recipe provides a conceptual overview of how to adapt your existing model to handle a different discrete video (DV) tokenizer, ensuring high-quality video generation under a new compression setting.
 
@@ -49,7 +49,7 @@ The NeMo Framework supports the following Cosmos Autoregressive (AR) models for
 
 ## Why Change Tokenizers?
 
-Cosmos Autoregressive models are typically trained on a specific tokenizer configuration (e.g., 8×16×16). If you’d like to reduce patch size or change compression (e.g., to 4×8×8), you can post-train the existing weights so that the model effectively aligns its internal representations with the new token embeddings—without re-training the tokenizer.
+Cosmos Autoregressive models are typically trained on a specific tokenizer configuration (e.g., 8×16×16). If you'd like to reduce patch size or change compression (e.g., to 4×8×8), you can post-train the existing weights so that the model effectively aligns its internal representations with the new token embeddings—without re-training the tokenizer.
 
 ## Tutorial: Finetuning Cosmos-4B on 10k Videos with a New Tokenizer
 
@@ -62,7 +62,7 @@ In this tutorial, we will:
 
 - **Original Sequence Length:** 12,800 tokens (33 frames * 640 px width * 1024 px height, when tokenized with an 8×16×16 tokenizer becomes ⌈33/8⌉ * (640/16) * (1024/16) = 5 * 64 * 80 = 12,800 tokens)
 - **New Sequence Length:** 12,800 tokens (17 frames * 320 px width * 512 px height, when tokenized with a 4×8×8 tokenizer becomes ⌈17/8⌉ * (320/4) * (512/8) = 5 * 80 * 64 = 12,800 tokens)
-- For other resolutions or frame counts, recalculate your maximum tokens to ensure you do not exceed the model’s capacity.
+- For other resolutions or frame counts, recalculate your maximum tokens to ensure you do not exceed the model's capacity.
 
 ```bash
 export WIDTH=512
@@ -145,7 +145,7 @@ torchrun --nproc-per-node=$NUM_GPUS cosmos1/models/autoregressive/nemo/post_trai
 After training, monitor checkpoints and generate sample outputs. Please keep in mind that you will have to
 disable the diffusion decoder, as this is not yet supported for non-standard tokenizers.
 
-1. **Checkpoints:**  
+1. **Checkpoints:**
    Checkpoints are saved under `./experiments/example_log_dir/default/checkpoints/`. For example:
    ```bash
    epoch=1-step=49
@@ -154,7 +154,7 @@ disable the diffusion decoder, as this is not yet supported for non-standard tok
    ```
    Choose a checkpoint (e.g., `epoch=1-step=99`) for inference.
 
-2. **Run Inference:**  
+2. **Run Inference:**
    Generate sample video outputs:
    ```bash
    # Set the checkpoint directory
@@ -177,7 +177,7 @@ disable the diffusion decoder, as this is not yet supported for non-standard tok
        --tokenizer_compression_factor $TOKENIZER_COMPRESSION_FACTOR
    ```
 
-3. **(Optional) Evaluate with TokenBench:**  
+3. **(Optional) Evaluate with TokenBench:**
    To compute quantitative metrics (e.g., PSNR, SSIM), generate a large number of videos using the script above and place them into the `evals`
    folder of the checkpoint. Please ensure that the input filename in the original folder (e.g. `my-video.mp4` matches the output filename in the
    `evals` folder). This is because PSNR and SSIM compare video quality against a reference, and our script will match the output video with
@@ -245,7 +245,7 @@ python cosmos1/models/autoregressive/nemo/post_training/video2world_prepare_data
 
 ### 3. Training/Finetuning for Video2World
 
-Fine-tune a Video2World–specific checkpoint (or any compatible Cosmos AR model) on your new data. For example, using the 5B Video2World model:
+Fine-tune a Video2World—specific checkpoint (or any compatible Cosmos AR model) on your new data. For example, using the 5B Video2World model:
 
 ```bash
 torchrun --nproc-per-node=$NUM_GPUS cosmos1/models/autoregressive/nemo/post_training/video2world_finetuning.py \
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/README.md b/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/README.md
@@ -439,7 +439,7 @@ srun \
    torchrun --nproc_per_node=8 cosmos1/models/diffusion/nemo/post_training/multicamera.py  --factory cosmos_multicamera_diffusion_7b_text2world_overfit   --yes   trainer.max_steps=1000     optim.config.lr=1e-6
    ```
 
-4. (Optional) Run Largescale Full Post-Training  
+4. (Optional) Run Largescale Full Post-Training
    You will need to prepare you own data similar with the overfit dataset in step 3. Change XDG_CACHE_HOME to point to the custom dataset
    ```bash
    torchrun --nproc_per_node=8 cosmos1/models/diffusion/nemo/post_training/multicamera.py --yes   trainer.max_steps=1000     optim.config.lr=1e-6
@@ -461,9 +461,9 @@ srun \
    # --factory cosmos_multicamera_diffusion_7b_image2world_finetune_w_traj
    ```
 
-For different post-training recipes make sure to change the autoresume directory so that checkpoints generated from recipe A are not used for recipe B. Modify 'recipe.trainer.callbacks.dirpath' and 'recipe.resume.resume_from_directory'.  
+For different post-training recipes make sure to change the autoresume directory so that checkpoints generated from recipe A are not used for recipe B. Modify 'recipe.trainer.callbacks.dirpath' and 'recipe.resume.resume_from_directory'.
 
-'recipe.trainer.val_check_interval' and 'recipe.trainer.limit_val_batches' can be set to 0 to disable validation during post-training.  
+'recipe.trainer.val_check_interval' and 'recipe.trainer.limit_val_batches' can be set to 0 to disable validation during post-training.
 
 Under `HF_HOME`, there should be a `hub` folder, which contains the base checkpoints. Under `multicamera` folder, there should be view embeddings, `mkdir -p $HF_HOME/multicamera && symlink -s cosmos1/models/diffusion/nemo/post_training/multicamera/*.pt $HF_HOME/multicamera`
 ```
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/dataverse/av22_tar_index_full.json b/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/dataverse/av22_tar_index_full.json
@@ -1 +1 @@
-{}
+{}
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/matplotlib/fontlist-v390.json b/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/matplotlib/fontlist-v390.json
@@ -1071,4 +1071,4 @@
     }
   ],
   "__class__": "FontManager"
-}
+}
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/mapping.json b/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/mapping.json
@@ -4,5 +4,5 @@
     "1a1c7950-01f0-4e37-9949-dad68e3d1072.camera_cross_right_120fov": "cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.tar",
     "1a1c7950-01f0-4e37-9949-dad68e3d1072.camera_rear_tele_30fov": "cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.tar",
     "1a1c7950-01f0-4e37-9949-dad68e3d1072.camera_rear_left_70fov": "cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.tar",
-    "1a1c7950-01f0-4e37-9949-dad68e3d1072.camera_rear_right_70fov": "cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.tar"    
-}
+    "1a1c7950-01f0-4e37-9949-dad68e3d1072.camera_rear_right_70fov": "cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.tar"
+}
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.json b/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/multicamera/AV-V2.2/alpamayo_caption_t5/qwen_t5_tars/trainv2-2-chunk-00/part_000000/001365.json
diff --git a/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/torch/megatron/megatron-gpt-345m_vocab b/nemo_vfm/physicalai/Cosmos/cosmos1/models/diffusion/nemo/post_training/torch/megatron/megatron-gpt-345m_vocab
diff --git a/nemo_vfm/physicalai/Dockerfile b/nemo_vfm/physicalai/Dockerfile
diff --git a/nemo_vfm/physicalai/datasets/dataverse/dataverse/utils/alpamayo/generator_config.yaml b/nemo_vfm/physicalai/datasets/dataverse/dataverse/utils/alpamayo/generator_config.yaml
diff --git a/nemo_vfm/physicalai/diffusion/post_training/multicamera/README.md b/nemo_vfm/physicalai/diffusion/post_training/multicamera/README.md
diff --git a/nemo_vfm/physicalai/diffusion/post_training/multicamera/train.sh b/nemo_vfm/physicalai/diffusion/post_training/multicamera/train.sh
diff --git a/nemo_vfm/physicalai/tokenizer/README.md b/nemo_vfm/physicalai/tokenizer/README.md
diff --git a/nemo_vfm/sparse_attention/README.md b/nemo_vfm/sparse_attention/README.md
diff --git a/nemo_vfm/sparse_attention/accel-config.yml b/nemo_vfm/sparse_attention/accel-config.yml
diff --git a/nemo_vfm/sparse_attention/training.md b/nemo_vfm/sparse_attention/training.md
diff --git a/tests/conftest.py b/tests/conftest.py

Original file line number	Diff line number	Diff line change
`@@ -27,4 +27,4 @@ If you find our code useful, please consider citing the following papers:`
`27`	`27`	`journal={arXiv preprint arXiv:2501.03575},`
`28`	`28`	`year={2025}`
`29`	`29`	`}`
`30`		-```
	`30`	+```
Original file line number	Diff line number	Diff line change
`@@ -1,4 +1,3 @@`
`1`	`1`	`# Automodel`
`2`	`2`
`3`	`3`	`Models implemented using Dtensors (NeMo Automodel)`
`4`		`-`
Original file line number	Diff line number	Diff line change
`@@ -1,4 +1,3 @@`
`1`	`1`	`# MBridge`
`2`	`2`
`3`	`3`	`Megatron models including DiT and dLLM implementations.`
`4`		`-`
Original file line number	Diff line number	Diff line change
`@@ -1,4 +1,3 @@`
`1`	`1`	`# Base`
`2`	`2`
`3`	`3`	`Base classes and interfaces for Megatron models.`
`4`		`-`