feat(logger.py): support swanlab

docs/design-docs/logger.md

@@ -1,6 +1,6 @@
 # Logger
 
-The logger is designed to track key training metrics (including distributed metrics with reductions and timing), as well as providing integration with logging backends like WandB, Tensorboard, and MLflow.
+The logger is designed to track key training metrics (including distributed metrics with reductions and timing), as well as providing integration with logging backends like WandB, Tensorboard, MLflow and Swanlab.
 
 ## Requirements
 
@@ -10,12 +10,13 @@ The logger is designed to track key training metrics (including distributed metr
    * WandB
    * Tensorboard
    * MLflow
+   * Swanlab
 
 ## Overall Design
 
 Since there is a single controller, the single process running the main training loop will gather the metrics and do the logging.
 
-To handle multiple logger backends, we will have a {py:class}`LoggerInterface <nemo_rl.utils.logger.LoggerInterface>` interface that the {py:class}`TensorboardLogger <nemo_rl.utils.logger.TensorboardLogger>`, {py:class}`WandbLogger <nemo_rl.utils.logger.WandbLogger>`, and {py:class}`MLflowLogger <nemo_rl.utils.logger.MLflowLogger>` will implement:
+To handle multiple logger backends, we will have a {py:class}`LoggerInterface <nemo_rl.utils.logger.LoggerInterface>` interface that the {py:class}`TensorboardLogger <nemo_rl.utils.logger.TensorboardLogger>`, {py:class}`WandbLogger <nemo_rl.utils.logger.WandbLogger>`, {py:class}`MLflowLogger <nemo_rl.utils.logger.MLflowLogger>` and {py:class}`SwanlabLogger <nemo_rl.utils.logger.SwanlabLogger>` will implement:
 
 ```python
 class LoggerInterface(ABC):
@@ -35,7 +36,7 @@ class LoggerInterface(ABC):
 A {py:class}`Logger <nemo_rl.utils.logger.Logger>` wrapper class will also implement {py:class}`LoggerInterface <nemo_rl.utils.logger.LoggerInterface>` and maintain a list of loggers to which it delegates writing logs. This will be the main class the user uses in the training loop. Usage example:
 
 ```python
-# Initialize logger with wandb, tensorboard, and mlflow enabled
+# Initialize logger with wandb, tensorboard, mlflow and swanlab enabled
 logging_config = {
     "wandb_enabled": True,
     "tensorboard_enabled": False,
@@ -45,6 +46,10 @@ logging_config = {
         "project": "grpo-dev",
         "name": "grpo-dev-logging",
     },
+    "swanlab": {
+        "project": "nemo-rl",
+        "name": "grpo-dev-logging",
+    },
     "tensorboard": {
         "log_dir": "logs",
     },
@@ -74,6 +79,13 @@ The logger supports three main logging backends:
 - Includes built-in hyperparameter logging
 - Offers rich visualization and collaboration features
 
+### Swanlab
+- Training visualization (Android, iOS, Wechat public account and Web)
+- Automatic logging
+- Hyperparameter recording
+- Experiment comparison
+- Multi-user collaboration
+
 ### Tensorboard
 - Local file-based logging
 - Standard TensorBoard visualization
@@ -121,6 +133,7 @@ The logger supports pretty-formatted logging of validation samples to help visua
 ```python
 logger:
   wandb_enabled: false
+  swanlab_enabled: false
   tensorboard_enabled: false
   mlflow_enabled: false
   num_val_samples_to_print: 10
@@ -140,7 +153,7 @@ When enabled, the pretty logging will generate formatted text similar to:
 
 ## GPU Metric Logging
 
-NeMo RL monitors GPU memory and utilization through [system metrics](https://docs.ray.io/en/latest/ray-observability/reference/system-metrics.html#system-metrics) exposed by Ray nodes. While Ray makes these metrics available for tools like Prometheus, NeMo RL directly polls GPU memory and utilization data and logs them to TensorBoard, WandB, and/or MLflow.
+NeMo RL monitors GPU memory and utilization through [system metrics](https://docs.ray.io/en/latest/ray-observability/reference/system-metrics.html#system-metrics) exposed by Ray nodes. While Ray makes these metrics available for tools like Prometheus, NeMo RL directly polls GPU memory and utilization data and logs them to TensorBoard, WandB, MLflow and/or SwanLab.
 
 This approach allows us to offer the same GPU metric tracking on all loggers and simplifies the implementation greatly.
 
@@ -149,6 +162,7 @@ This feature is enabled with the `monitor_gpus` configuration parameter. The fre
 ```python
 logger:
   wandb_enabled: false
+  swanlab_enabled: false
   tensorboard_enabled: false
   mlflow_enabled: false
   monitor_gpus: true
@@ -162,8 +176,8 @@ While it is feasible to monitor using remote workers, the implementation require
 * Logs sent back to the driver do not introduce significant overhead.
 * Metrics remain clear and interpretable, avoiding issues like double counting caused by colocated workers.
 * Workers can gracefully flush their logs in case of failure.
-* Logging behaves consistently across TensorBoard, WandB, and MLflow.
+* Logging behaves consistently across TensorBoard, WandB, MLflow and Swanlab.
 * Workers that spawn other workers accurately report the total resource usage of any grandchild workers.
 
 Due to these complexities, we opted for a simpler approach: collecting metrics exposed by the Ray metrics server from the driver.
-:::
+:::

examples/configs/dpo.yaml

@@ -155,6 +155,7 @@ logger:
   wandb_enabled: false # Make sure you do a ``wandb login [Your API key]'' before running
   tensorboard_enabled: false
   mlflow_enabled: false  # Disable MLflow logging
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/grpo-deepscaler-1.5b-8K.yaml

@@ -129,6 +129,7 @@ logger:
   wandb_enabled: false
   tensorboard_enabled: false
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: false  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   wandb:
     project: "grpo-dev"

examples/configs/grpo_math_1B.yaml

@@ -138,6 +138,7 @@ logger:
   wandb_enabled: false
   tensorboard_enabled: false
   mlflow_enabled: false  # Disable MLflow logging
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   wandb:
     project: "grpo-dev"

examples/configs/grpo_math_1B_megatron.yaml

@@ -157,6 +157,7 @@ logger:
   wandb_enabled: false
   tensorboard_enabled: false
   mlflow_enabled: false  # Disable MLflow logging
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: false  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   wandb:
     project: "grpo-dev"

examples/configs/grpo_sliding_puzzle.yaml

@@ -53,6 +53,7 @@ logger:
   wandb_enabled: false
   tensorboard_enabled: false
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   wandb:
     project: "grpo-dev"

examples/configs/recipes/llm/dpo-llama3.1-8b-instruct-4n8g-fsdp2tp1.v2.yaml

@@ -79,6 +79,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/dpo-llama3.1-8b-instruct-4n8g-fsdp2tp2-quick.v2.yaml

@@ -79,6 +79,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/dpo-llama3.1-8b-instruct-4n8g-megatron.yaml

@@ -112,6 +112,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/dpo-llama3.1-8b-instruct-4n8g-megatrontp2pp2-quick.yaml

@@ -112,6 +112,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/dpo-llama3.2-1b-instruct-1n8g-fsdp2tp1.v2.yaml

@@ -78,6 +78,7 @@ logger:
   log_dir: "logs"
   wandb_enabled: true
   tensorboard_enabled: true
+  swanlab_enabled: false # Disable SwanLab logging
   mlflow_enabled: false
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal

examples/configs/recipes/llm/grpo-gemma3-1b-it-1n8g-fsdp2tp1.yaml

@@ -108,6 +108,7 @@ logger:
   num_val_samples_to_print: 0
   wandb_enabled: true
   tensorboard_enabled: true
+  swanlab_enabled: false # Disable SwanLab logging
   mlflow_enabled: false
   monitor_gpus: true
   wandb:

examples/configs/recipes/llm/grpo-gemma3-27b-it-16n8g-fsdp2tp8sp-actckpt-long.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/grpo-llama3.1-8b-instruct-4n8g-fsdp2tp1-long.v3.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/grpo-llama3.2-1b-instruct-1n8g-fsdp2tp1.v3.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/grpo-qwen2.5-32b-32n8g-fsdp2tp8sp-actckpt-long.v3.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/grpo-qwen2.5-32b-32n8g-fsdp2tp8sp-actckpt.v3.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/grpo-qwen2.5-7b-instruct-4n8g-fsdp2tp4sp.v3.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/grpo-qwen2.5-math-1.5b-instruct-1n8g-fsdp2tp1.v3.yaml

@@ -110,6 +110,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   wandb:
     project: nemo-rl

examples/configs/recipes/llm/sft-llama3.1-8b-instruct-1n8g-fsdp2tp1-long.v2.yaml

@@ -59,6 +59,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/sft-llama3.1-8b-instruct-1n8g-fsdp2tp2sp.v2.yaml

@@ -59,6 +59,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/sft-llama3.1-8b-instruct-1n8g-megatron.yaml

@@ -103,6 +103,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/sft-llama3.2-1b-1n8g-fsdp2tp1.v2.yaml

@@ -59,6 +59,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/recipes/llm/sft-qwen2.5-32b-4n8g-fsdp2tp8sp-actckpt.v2.yaml

@@ -59,6 +59,7 @@ logger:
   wandb_enabled: true
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/sft.yaml

@@ -138,6 +138,7 @@ logger:
   wandb_enabled: true # Make sure you do a ``wandb login [Your API key]'' before running
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: true  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb:

examples/configs/sft_openmathinstruct2.yaml

@@ -73,6 +73,7 @@ logger:
   wandb_enabled: true # Make sure you do a ``wandb login [Your API key]'' before running
   tensorboard_enabled: true
   mlflow_enabled: false
+  swanlab_enabled: false # Disable SwanLab logging
   monitor_gpus: false  # If true, will monitor GPU usage and log to wandb and/or tensorboard
   num_val_samples_to_print: 0  # Number of validation samples to pretty print on terminal
   wandb: