train and eval working

.gitignore

@@ -20,4 +20,6 @@ app/core/__pycache__/
 src/__pycache__/test_infra.cpython-310.pyc
 app/core/__pycache__/config.cpython-310.pyc
 data/
-!configs/data/
+!configs/data/
+checkpoints/
+logs/

configs/callbacks/default.yaml

@@ -1,7 +1,7 @@
 defaults:
   - model_checkpoint
   - early_stopping
-  - model_summary
+  - rich_model_summary
   - rich_progress_bar
   - _self_
 
@@ -17,7 +17,7 @@ early_stopping:
   patience: 3
   mode: "min"
 
-model_summary:
+rich_model_summary:
   max_depth: -1
 
 rich_progress_bar:

configs/callbacks/{model_summary.yaml → rich_model_summary.yaml}

@@ -1,2 +1,2 @@
-model_summary:
+rich_model_summary:
   max_depth: 1

configs/data/catdog.yaml

@@ -6,5 +6,4 @@ num_workers: 4
 batch_size: 32
 train_val_split: [0.8, 0.2]
 pin_memory: False
-image_size: 160
-dataset_url: "https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip"
+image_size: 160

configs/experiment/catdog_experiment.yaml

@@ -38,7 +38,7 @@ model:
 
 trainer:
   min_epochs: 1
-  max_epochs: 6
+  max_epochs: 10
 
 callbacks:
   model_checkpoint:
@@ -54,7 +54,7 @@ callbacks:
     mode: "max"
     verbose: True
 
-  model_summary:
+  rich_model_summary:
     max_depth: 1
 
   rich_progress_bar:

configs/paths/catdog.yaml

@@ -16,7 +16,7 @@ ckpt_dir: ${paths.root_dir}/checkpoints
 artifact_dir: ${paths.root_dir}/artifacts/
 
 # download url for the dataset
-data_url: "https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip"
+data_url: "https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip"
 
 # path to output directory, created dynamically by hydra
 # path generation pattern is specified in `configs/hydra/default.yaml`

configs/train.yaml

@@ -39,3 +39,6 @@ seed: 42
 
 # name of the experiment
 name: "dogbreed_experiment"
+
+# optimization metric
+optimization_metric: "val_acc"

logs/train/runs/2024-11-08_15-27-10/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: false
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: false
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-27-10/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-27-10
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-27-10/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

logs/train/runs/2024-11-08_15-29-07/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: false
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: false
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-29-07/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-29-07
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-29-07/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

logs/train/runs/2024-11-08_15-29-42/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: false
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: false
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-29-42/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-29-42
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-29-42/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

logs/train/runs/2024-11-08_15-30-22/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: false
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: false
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-30-22/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-30-22
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-30-22/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

logs/train/runs/2024-11-08_15-35-40/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: false
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: false
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-35-40/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test2
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-35-40
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-35-40/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

logs/train/runs/2024-11-08_15-35-57/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: false
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: false
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-35-57/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test2
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-35-57
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-35-57/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

logs/train/runs/2024-11-08_15-37-45/.hydra/config.yaml

@@ -1,95 +0,0 @@
-task_name: train
-tags:
-- dev
-train: true
-test: false
-ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt
-seed: 42
-name: catdog_experiment
-data:
-  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule
-  data_dir: ${paths.data_dir}
-  url: ${paths.data_url}
-  num_workers: 8
-  batch_size: 64
-  train_val_split:
-  - 0.8
-  - 0.2
-  pin_memory: true
-  image_size: 160
-  dataset_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip
-model:
-  _target_: src.models.catdog_model.ViTTinyClassifier
-  img_size: 160
-  patch_size: 16
-  num_classes: 2
-  embed_dim: 64
-  depth: 6
-  num_heads: 2
-  mlp_ratio: 3
-  pre_norm: false
-  lr: 0.001
-  weight_decay: 1.0e-05
-  factor: 0.1
-  patience: 10
-  min_lr: 1.0e-06
-callbacks:
-  model_checkpoint:
-    dirpath: ${paths.ckpt_dir}
-    filename: best-checkpoint
-    monitor: val_acc
-    verbose: true
-    save_last: true
-    save_top_k: 1
-    mode: max
-    auto_insert_metric_name: false
-    save_weights_only: false
-    every_n_train_steps: null
-    train_time_interval: null
-    every_n_epochs: null
-    save_on_train_epoch_end: null
-  early_stopping:
-    monitor: val_acc
-    min_delta: 0.0
-    patience: 10
-    verbose: true
-    mode: max
-    strict: true
-    check_finite: true
-    stopping_threshold: null
-    divergence_threshold: null
-    check_on_train_epoch_end: null
-  model_summary:
-    max_depth: 1
-  rich_progress_bar:
-    refresh_rate: 1
-logger:
-  csv:
-    save_dir: ${paths.output_dir}
-    name: csv/
-    prefix: ''
-  tensorboard:
-    save_dir: ${paths.output_dir}/tensorboard/
-    name: null
-    log_graph: false
-    default_hp_metric: true
-    prefix: ''
-trainer:
-  _target_: lightning.Trainer
-  default_root_dir: ${paths.output_dir}
-  min_epochs: 1
-  max_epochs: 6
-  accelerator: auto
-  devices: auto
-  deterministic: true
-  log_every_n_steps: 10
-  fast_dev_run: false
-paths:
-  root_dir: ${oc.env:PROJECT_ROOT}
-  data_dir: ${paths.root_dir}/data/
-  log_dir: ${paths.root_dir}/logs/
-  ckpt_dir: ${paths.root_dir}/checkpoints
-  artifact_dir: ${paths.root_dir}/artifacts/
-  data_url: https://storage.googleapis.com/mledu-datasets/cats_and_dogs_filtered.zip
-  output_dir: ${hydra:runtime.output_dir}
-  work_dir: ${hydra:runtime.cwd}

logs/train/runs/2024-11-08_15-37-45/.hydra/hydra.yaml

@@ -1,174 +0,0 @@
-hydra:
-  run:
-    dir: ${paths.log_dir}/${task_name}/runs/${now:%Y-%m-%d}_${now:%H-%M-%S}
-  sweep:
-    dir: ${paths.log_dir}/${task_name}/multiruns/${now:%Y-%m-%d}_${now:%H-%M-%S}
-    subdir: ${hydra.job.num}
-  launcher:
-    _target_: hydra._internal.core_plugins.basic_launcher.BasicLauncher
-  sweeper:
-    _target_: hydra._internal.core_plugins.basic_sweeper.BasicSweeper
-    max_batch_size: null
-    params: null
-  help:
-    app_name: ${hydra.job.name}
-    header: '${hydra.help.app_name} is powered by Hydra.
-
-      '
-    footer: 'Powered by Hydra (https://hydra.cc)
-
-      Use --hydra-help to view Hydra specific help
-
-      '
-    template: '${hydra.help.header}
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (group=option)
-
-
-      $APP_CONFIG_GROUPS
-
-
-      == Config ==
-
-      Override anything in the config (foo.bar=value)
-
-
-      $CONFIG
-
-
-      ${hydra.help.footer}
-
-      '
-  hydra_help:
-    template: 'Hydra (${hydra.runtime.version})
-
-      See https://hydra.cc for more info.
-
-
-      == Flags ==
-
-      $FLAGS_HELP
-
-
-      == Configuration groups ==
-
-      Compose your configuration from those groups (For example, append hydra/job_logging=disabled
-      to command line)
-
-
-      $HYDRA_CONFIG_GROUPS
-
-
-      Use ''--cfg hydra'' to Show the Hydra config.
-
-      '
-    hydra_help: ???
-  hydra_logging:
-    version: 1
-    formatters:
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(purple)sHYDRA%(reset)s] %(message)s'
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-    root:
-      level: INFO
-      handlers:
-      - console
-    disable_existing_loggers: false
-  job_logging:
-    version: 1
-    formatters:
-      simple:
-        format: '[%(asctime)s][%(name)s][%(levelname)s] - %(message)s'
-      colorlog:
-        (): colorlog.ColoredFormatter
-        format: '[%(cyan)s%(asctime)s%(reset)s][%(blue)s%(name)s%(reset)s][%(log_color)s%(levelname)s%(reset)s]
-          - %(message)s'
-        log_colors:
-          DEBUG: purple
-          INFO: green
-          WARNING: yellow
-          ERROR: red
-          CRITICAL: red
-    handlers:
-      console:
-        class: logging.StreamHandler
-        formatter: colorlog
-        stream: ext://sys.stdout
-      file:
-        class: logging.FileHandler
-        formatter: simple
-        filename: ${hydra.runtime.output_dir}/${task_name}.log
-    root:
-      level: INFO
-      handlers:
-      - console
-      - file
-    disable_existing_loggers: false
-  env: {}
-  mode: RUN
-  searchpath: []
-  callbacks: {}
-  output_subdir: .hydra
-  overrides:
-    hydra:
-    - hydra.mode=RUN
-    task: []
-  job:
-    name: hydra_test2
-    chdir: null
-    override_dirname: ''
-    id: ???
-    num: ???
-    config_name: train
-    env_set: {}
-    env_copy: []
-    config:
-      override_dirname:
-        kv_sep: '='
-        item_sep: ','
-        exclude_keys: []
-  runtime:
-    version: 1.3.2
-    version_base: '1.1'
-    cwd: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws
-    config_sources:
-    - path: hydra.conf
-      schema: pkg
-      provider: hydra
-    - path: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/configs
-      schema: file
-      provider: main
-    - path: hydra_plugins.hydra_colorlog.conf
-      schema: pkg
-      provider: hydra-colorlog
-    - path: ''
-      schema: structured
-      provider: schema
-    output_dir: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/train/runs/2024-11-08_15-37-45
-    choices:
-      debug: null
-      experiment: catdog_experiment
-      hydra: default
-      paths: catdog
-      trainer: default
-      logger: default
-      callbacks: default
-      model: catdog_classifier
-      data: catdog
-      hydra/env: default
-      hydra/callbacks: null
-      hydra/job_logging: colorlog
-      hydra/hydra_logging: colorlog
-      hydra/hydra_help: default
-      hydra/help: default
-      hydra/sweeper: basic
-      hydra/launcher: basic
-      hydra/output: default
-  verbose: false

logs/train/runs/2024-11-08_15-37-45/.hydra/overrides.yaml

@@ -1 +0,0 @@
-[]

notebooks/training_lightning.ipynb

@@ -0,0 +1,889 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/javascript": "IPython.notebook.set_autosave_interval(300000)"
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Autosaving every 300 seconds\n"
+     ]
+    }
+   ],
+   "source": [
+    "%autosave 300\n",
+    "%load_ext autoreload\n",
+    "%autoreload 2\n",
+    "%reload_ext autoreload\n",
+    "%config Completer.use_jedi = False"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "/mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws\n"
+     ]
+    }
+   ],
+   "source": [
+    "\n",
+    "import os\n",
+    "\n",
+    "os.chdir(\"..\")\n",
+    "print(os.getcwd())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/anaconda/envs/emlo_env/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os\n",
+    "import shutil\n",
+    "from pathlib import Path\n",
+    "import torch\n",
+    "import lightning as L\n",
+    "from lightning.pytorch.loggers import Logger\n",
+    "from typing import List\n",
+    "from src.datamodules.catdog_datamodule import CatDogImageDataModule\n",
+    "from src.utils.logging_utils import setup_logger, task_wrapper\n",
+    "from loguru import logger\n",
+    "from dotenv import load_dotenv, find_dotenv\n",
+    "import rootutils\n",
+    "import hydra\n",
+    "from omegaconf import DictConfig, OmegaConf\n",
+    "from lightning.pytorch.callbacks import (\n",
+    "    ModelCheckpoint,\n",
+    "    EarlyStopping,\n",
+    "    RichModelSummary,\n",
+    "    RichProgressBar,\n",
+    ")\n",
+    "from lightning.pytorch.loggers import TensorBoardLogger, CSVLogger"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:17.572\u001b[0m | \u001b[31m\u001b[1mERROR   \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m9\u001b[0m - \u001b[31m\u001b[1mname '__file__' is not defined\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Load environment variables\n",
+    "load_dotenv(find_dotenv(\".env\"))\n",
+    "\n",
+    "# Setup root directory\n",
+    "try:\n",
+    "    root = rootutils.setup_root(__file__, indicator=\".project-root\")\n",
+    "\n",
+    "except Exception as e:\n",
+    "    logger.error(e)\n",
+    "    root = Path(os.getcwd())\n",
+    "    os.environ[\"PROJECT_ROOT\"] = str(root)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def load_checkpoint_if_available(ckpt_path: str) -> str:\n",
+    "    \"\"\"Check if the specified checkpoint exists and return the valid checkpoint path.\"\"\"\n",
+    "    if ckpt_path and Path(ckpt_path).exists():\n",
+    "        logger.info(f\"Checkpoint found: {ckpt_path}\")\n",
+    "        return ckpt_path\n",
+    "    else:\n",
+    "        logger.warning(\n",
+    "            f\"No checkpoint found at {ckpt_path}. Using current model weights.\"\n",
+    "        )\n",
+    "        return None\n",
+    "\n",
+    "\n",
+    "def clear_checkpoint_directory(ckpt_dir: str):\n",
+    "    \"\"\"Clear all contents of the checkpoint directory without deleting the directory itself.\"\"\"\n",
+    "    ckpt_dir_path = Path(ckpt_dir)\n",
+    "    if ckpt_dir_path.exists() and ckpt_dir_path.is_dir():\n",
+    "        logger.info(f\"Clearing checkpoint directory: {ckpt_dir}\")\n",
+    "        # Iterate over all files and directories in the checkpoint directory and remove them\n",
+    "        for item in ckpt_dir_path.iterdir():\n",
+    "            try:\n",
+    "                if item.is_file() or item.is_symlink():\n",
+    "                    item.unlink()  # Remove file or symlink\n",
+    "                elif item.is_dir():\n",
+    "                    shutil.rmtree(item)  # Remove directory\n",
+    "            except Exception as e:\n",
+    "                logger.error(f\"Failed to delete {item}: {e}\")\n",
+    "        logger.info(f\"Checkpoint directory cleared: {ckpt_dir}\")\n",
+    "    else:\n",
+    "        logger.info(\n",
+    "            f\"Checkpoint directory does not exist. Creating directory: {ckpt_dir}\"\n",
+    "        )\n",
+    "        os.makedirs(ckpt_dir_path, exist_ok=True)\n",
+    "\n",
+    "\n",
+    "@task_wrapper\n",
+    "def train_module(\n",
+    "    cfg: DictConfig,\n",
+    "    data_module: L.LightningDataModule,\n",
+    "    model: L.LightningModule,\n",
+    "    trainer: L.Trainer,\n",
+    "):\n",
+    "    \"\"\"Train the model using the provided Trainer and DataModule.\"\"\"\n",
+    "    logger.info(\"Training the model\")\n",
+    "    trainer.fit(model, data_module)\n",
+    "    train_metrics = trainer.callback_metrics\n",
+    "    try:\n",
+    "        logger.info(\n",
+    "            f\"Training completed with the following metrics- train_acc: {train_metrics['train_acc'].item()} and val_acc: {train_metrics['val_acc'].item()}\"\n",
+    "        )\n",
+    "    except KeyError:\n",
+    "        logger.info(f\"Training completed with the following metrics:{train_metrics}\")\n",
+    "\n",
+    "    return train_metrics\n",
+    "\n",
+    "\n",
+    "@task_wrapper\n",
+    "def run_test_module(\n",
+    "    cfg: DictConfig,\n",
+    "    datamodule: L.LightningDataModule,\n",
+    "    model: L.LightningModule,\n",
+    "    trainer: L.Trainer,\n",
+    "):\n",
+    "    \"\"\"Test the model using the best checkpoint or the current model weights.\"\"\"\n",
+    "    logger.info(\"Testing the model\")\n",
+    "    datamodule.setup(stage=\"test\")\n",
+    "\n",
+    "    ckpt_path = load_checkpoint_if_available(cfg.ckpt_path)\n",
+    "\n",
+    "    # If no checkpoint is available, Lightning will use current model weights\n",
+    "    test_metrics = trainer.test(model, datamodule, ckpt_path=ckpt_path)\n",
+    "    logger.info(f\"Test metrics:\\n{test_metrics}\")\n",
+    "\n",
+    "    return test_metrics[0] if test_metrics else {}"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/tmp/ipykernel_487789/541470590.py:8: UserWarning: \n",
+      "The version_base parameter is not specified.\n",
+      "Please specify a compatability version level, or None.\n",
+      "Will assume defaults for version 1.1\n",
+      "  with hydra.initialize(config_path=\"../configs\"):\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Full Configuration:\n",
+      "task_name: train\n",
+      "tags:\n",
+      "- dev\n",
+      "train: true\n",
+      "test: false\n",
+      "ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt\n",
+      "seed: 42\n",
+      "name: catdog_experiment\n",
+      "data:\n",
+      "  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule\n",
+      "  data_dir: ${paths.data_dir}\n",
+      "  url: ${paths.data_url}\n",
+      "  num_workers: 8\n",
+      "  batch_size: 64\n",
+      "  train_val_split:\n",
+      "  - 0.8\n",
+      "  - 0.2\n",
+      "  pin_memory: true\n",
+      "  image_size: 160\n",
+      "model:\n",
+      "  _target_: src.models.catdog_model.ViTTinyClassifier\n",
+      "  img_size: 160\n",
+      "  patch_size: 16\n",
+      "  num_classes: 2\n",
+      "  embed_dim: 64\n",
+      "  depth: 6\n",
+      "  num_heads: 2\n",
+      "  mlp_ratio: 3\n",
+      "  pre_norm: false\n",
+      "  lr: 0.001\n",
+      "  weight_decay: 1.0e-05\n",
+      "  factor: 0.1\n",
+      "  patience: 10\n",
+      "  min_lr: 1.0e-06\n",
+      "callbacks:\n",
+      "  model_checkpoint:\n",
+      "    dirpath: ${paths.ckpt_dir}\n",
+      "    filename: best-checkpoint\n",
+      "    monitor: val_acc\n",
+      "    verbose: true\n",
+      "    save_last: true\n",
+      "    save_top_k: 1\n",
+      "    mode: max\n",
+      "    auto_insert_metric_name: false\n",
+      "    save_weights_only: false\n",
+      "    every_n_train_steps: null\n",
+      "    train_time_interval: null\n",
+      "    every_n_epochs: null\n",
+      "    save_on_train_epoch_end: null\n",
+      "  early_stopping:\n",
+      "    monitor: val_acc\n",
+      "    min_delta: 0.0\n",
+      "    patience: 10\n",
+      "    verbose: true\n",
+      "    mode: max\n",
+      "    strict: true\n",
+      "    check_finite: true\n",
+      "    stopping_threshold: null\n",
+      "    divergence_threshold: null\n",
+      "    check_on_train_epoch_end: null\n",
+      "  rich_model_summary:\n",
+      "    max_depth: 1\n",
+      "  rich_progress_bar:\n",
+      "    refresh_rate: 1\n",
+      "logger:\n",
+      "  csv:\n",
+      "    save_dir: ${paths.output_dir}\n",
+      "    name: csv/\n",
+      "    prefix: ''\n",
+      "  tensorboard:\n",
+      "    save_dir: ${paths.output_dir}/tensorboard/\n",
+      "    name: null\n",
+      "    log_graph: false\n",
+      "    default_hp_metric: true\n",
+      "    prefix: ''\n",
+      "trainer:\n",
+      "  _target_: lightning.Trainer\n",
+      "  default_root_dir: ${paths.output_dir}\n",
+      "  min_epochs: 1\n",
+      "  max_epochs: 6\n",
+      "  accelerator: auto\n",
+      "  devices: auto\n",
+      "  deterministic: true\n",
+      "  log_every_n_steps: 10\n",
+      "  fast_dev_run: false\n",
+      "paths:\n",
+      "  root_dir: ${oc.env:PROJECT_ROOT}\n",
+      "  data_dir: ${paths.root_dir}/data/\n",
+      "  log_dir: ${paths.root_dir}/logs/\n",
+      "  ckpt_dir: ${paths.root_dir}/checkpoints\n",
+      "  artifact_dir: ${paths.root_dir}/artifacts/\n",
+      "  data_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip\n",
+      "  output_dir: ${hydra:runtime.output_dir}\n",
+      "  work_dir: ${hydra:runtime.cwd}\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "import hydra\n",
+    "from omegaconf import DictConfig, OmegaConf\n",
+    "\n",
+    "\n",
+    "# Function to load the configuration as an object without using the @hydra.main decorator\n",
+    "def load_config() -> DictConfig:\n",
+    "    # Initialize the configuration context (e.g., \"../configs\" directory)\n",
+    "    with hydra.initialize(config_path=\"../configs\"):\n",
+    "        # Compose the configuration object with a specific config name (e.g., \"train\")\n",
+    "        cfg = hydra.compose(config_name=\"train\")\n",
+    "    return cfg\n",
+    "\n",
+    "\n",
+    "# Load the configuration\n",
+    "cfg = load_config()\n",
+    "\n",
+    "# Print the entire configuration for reference\n",
+    "print(\"Full Configuration:\")\n",
+    "print(OmegaConf.to_yaml(cfg))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:23\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m8\u001b[0m - \u001b[1mWhole Config:\n",
+      "task_name: train\n",
+      "tags:\n",
+      "- dev\n",
+      "train: true\n",
+      "test: false\n",
+      "ckpt_path: ${paths.ckpt_dir}/best-checkpoint.ckpt\n",
+      "seed: 42\n",
+      "name: catdog_experiment\n",
+      "data:\n",
+      "  _target_: src.datamodules.catdog_datamodule.CatDogImageDataModule\n",
+      "  data_dir: ${paths.data_dir}\n",
+      "  url: ${paths.data_url}\n",
+      "  num_workers: 8\n",
+      "  batch_size: 64\n",
+      "  train_val_split:\n",
+      "  - 0.8\n",
+      "  - 0.2\n",
+      "  pin_memory: true\n",
+      "  image_size: 160\n",
+      "model:\n",
+      "  _target_: src.models.catdog_model.ViTTinyClassifier\n",
+      "  img_size: 160\n",
+      "  patch_size: 16\n",
+      "  num_classes: 2\n",
+      "  embed_dim: 64\n",
+      "  depth: 6\n",
+      "  num_heads: 2\n",
+      "  mlp_ratio: 3\n",
+      "  pre_norm: false\n",
+      "  lr: 0.001\n",
+      "  weight_decay: 1.0e-05\n",
+      "  factor: 0.1\n",
+      "  patience: 10\n",
+      "  min_lr: 1.0e-06\n",
+      "callbacks:\n",
+      "  model_checkpoint:\n",
+      "    dirpath: ${paths.ckpt_dir}\n",
+      "    filename: best-checkpoint\n",
+      "    monitor: val_acc\n",
+      "    verbose: true\n",
+      "    save_last: true\n",
+      "    save_top_k: 1\n",
+      "    mode: max\n",
+      "    auto_insert_metric_name: false\n",
+      "    save_weights_only: false\n",
+      "    every_n_train_steps: null\n",
+      "    train_time_interval: null\n",
+      "    every_n_epochs: null\n",
+      "    save_on_train_epoch_end: null\n",
+      "  early_stopping:\n",
+      "    monitor: val_acc\n",
+      "    min_delta: 0.0\n",
+      "    patience: 10\n",
+      "    verbose: true\n",
+      "    mode: max\n",
+      "    strict: true\n",
+      "    check_finite: true\n",
+      "    stopping_threshold: null\n",
+      "    divergence_threshold: null\n",
+      "    check_on_train_epoch_end: null\n",
+      "  rich_model_summary:\n",
+      "    max_depth: 1\n",
+      "  rich_progress_bar:\n",
+      "    refresh_rate: 1\n",
+      "logger:\n",
+      "  csv:\n",
+      "    save_dir: ${paths.output_dir}\n",
+      "    name: csv/\n",
+      "    prefix: ''\n",
+      "  tensorboard:\n",
+      "    save_dir: ${paths.output_dir}/tensorboard/\n",
+      "    name: null\n",
+      "    log_graph: false\n",
+      "    default_hp_metric: true\n",
+      "    prefix: ''\n",
+      "trainer:\n",
+      "  _target_: lightning.Trainer\n",
+      "  default_root_dir: ${paths.output_dir}\n",
+      "  min_epochs: 1\n",
+      "  max_epochs: 6\n",
+      "  accelerator: auto\n",
+      "  devices: auto\n",
+      "  deterministic: true\n",
+      "  log_every_n_steps: 10\n",
+      "  fast_dev_run: false\n",
+      "paths:\n",
+      "  root_dir: ${oc.env:PROJECT_ROOT}\n",
+      "  data_dir: ${paths.root_dir}/data/\n",
+      "  log_dir: ${paths.root_dir}/logs/\n",
+      "  ckpt_dir: ${paths.root_dir}/checkpoints\n",
+      "  artifact_dir: ${paths.root_dir}/artifacts/\n",
+      "  data_url: https://download.pytorch.org/tutorials/cats_and_dogs_filtered.zip\n",
+      "  output_dir: ${hydra:runtime.output_dir}\n",
+      "  work_dir: ${hydra:runtime.cwd}\n",
+      "\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Initialize logger\n",
+    "if cfg.task_name == \"train\":\n",
+    "    log_path = Path(cfg.paths.log_dir) / \"train.log\"\n",
+    "else:\n",
+    "    log_path = Path(cfg.paths.log_dir) / \"eval.log\"\n",
+    "setup_logger(log_path)\n",
+    "\n",
+    "logger.info(f\"Whole Config:\\n{OmegaConf.to_yaml(cfg)}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m3\u001b[0m - \u001b[1mRoot directory: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws\u001b[0m\n",
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m5\u001b[0m - \u001b[1mCurrent working directory: ['.dvc', '.dvcignore', '.env', '.git', '.github', '.gitignore', '.project-root', 'aws', 'basic_setup.md', 'configs', 'data', 'data.dvc', 'docker-compose.yaml', 'Dockerfile', 'ec2_runner_setup.md', 'logs', 'main.py', 'notebooks', 'poetry.lock', 'pyproject.toml', 'README.md', 'setup_aws_ci.md', 'src', 'tests', 'todo.md']\u001b[0m\n",
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m8\u001b[0m - \u001b[1mCheckpoint directory: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/checkpoints\u001b[0m\n",
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m12\u001b[0m - \u001b[1mData directory: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/data/\u001b[0m\n",
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m16\u001b[0m - \u001b[1mLog directory: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/logs/\u001b[0m\n",
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m20\u001b[0m - \u001b[1mArtifact directory: /mnt/batch/tasks/shared/LS_root/mounts/clusters/soutrik-vm-dev/code/Users/Soutrik.Chowdhury/pytorch-template-aws/artifacts/\u001b[0m\n",
+      "\u001b[32m2024-11-08 18:25:25\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m28\u001b[0m - \u001b[1mExperiment name: catdog_experiment\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "# the path to the checkpoint directory\n",
+    "root_dir = cfg.paths.root_dir\n",
+    "logger.info(f\"Root directory: {root_dir}\")\n",
+    "\n",
+    "logger.info(f\"Current working directory: {os.listdir(root_dir)}\")\n",
+    "\n",
+    "ckpt_dir = cfg.paths.ckpt_dir\n",
+    "logger.info(f\"Checkpoint directory: {ckpt_dir}\")\n",
+    "\n",
+    "# the path to the data directory\n",
+    "data_dir = cfg.paths.data_dir\n",
+    "logger.info(f\"Data directory: {data_dir}\")\n",
+    "\n",
+    "# the path to the log directory\n",
+    "log_dir = cfg.paths.log_dir\n",
+    "logger.info(f\"Log directory: {log_dir}\")\n",
+    "\n",
+    "# the path to the artifact directory\n",
+    "artifact_dir = cfg.paths.artifact_dir\n",
+    "logger.info(f\"Artifact directory: {artifact_dir}\")\n",
+    "\n",
+    "# output directory\n",
+    "# output_dir = cfg.paths.output_dir\n",
+    "# logger.info(f\"Output directory: {output_dir}\")\n",
+    "\n",
+    "# name of the experiment\n",
+    "experiment_name = cfg.name\n",
+    "logger.info(f\"Experiment name: {experiment_name}\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:28\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m2\u001b[0m - \u001b[1mInstantiating datamodule <src.datamodules.catdog_datamodule.CatDogImageDataModule>\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Initialize DataModule\n",
+    "logger.info(f\"Instantiating datamodule <{cfg.data._target_}>\")\n",
+    "datamodule: L.LightningDataModule = hydra.utils.instantiate(cfg.data)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:28\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m2\u001b[0m - \u001b[1mNo GPU available\u001b[0m\n",
+      "Seed set to 42\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "42"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Check for GPU availability\n",
+    "logger.info(\"GPU available\" if torch.cuda.is_available() else \"No GPU available\")\n",
+    "\n",
+    "# Set seed for reproducibility\n",
+    "L.seed_everything(cfg.seed, workers=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:29\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m2\u001b[0m - \u001b[1mInstantiating model <src.models.catdog_model.ViTTinyClassifier>\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Initialize model\n",
+    "logger.info(f\"Instantiating model <{cfg.model._target_}>\")\n",
+    "model: L.LightningModule = hydra.utils.instantiate(cfg.model)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[32m2024-11-08 18:25:30\u001b[0m | \u001b[1mINFO    \u001b[0m | \u001b[36m__main__\u001b[0m:\u001b[36m<module>\u001b[0m:\u001b[36m1\u001b[0m - \u001b[1mModel summary:\n",
+      "ViTTinyClassifier(\n",
+      "  (model): VisionTransformer(\n",
+      "    (patch_embed): PatchEmbed(\n",
+      "      (proj): Conv2d(3, 64, kernel_size=(16, 16), stride=(16, 16))\n",
+      "      (norm): Identity()\n",
+      "    )\n",
+      "    (pos_drop): Dropout(p=0.0, inplace=False)\n",
+      "    (patch_drop): Identity()\n",
+      "    (norm_pre): Identity()\n",
+      "    (blocks): Sequential(\n",
+      "      (0): Block(\n",
+      "        (norm1): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (attn): Attention(\n",
+      "          (qkv): Linear(in_features=64, out_features=192, bias=False)\n",
+      "          (q_norm): Identity()\n",
+      "          (k_norm): Identity()\n",
+      "          (attn_drop): Dropout(p=0.0, inplace=False)\n",
+      "          (proj): Linear(in_features=64, out_features=64, bias=True)\n",
+      "          (proj_drop): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls1): Identity()\n",
+      "        (drop_path1): Identity()\n",
+      "        (norm2): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (mlp): Mlp(\n",
+      "          (fc1): Linear(in_features=64, out_features=192, bias=True)\n",
+      "          (act): GELU(approximate='none')\n",
+      "          (drop1): Dropout(p=0.0, inplace=False)\n",
+      "          (norm): Identity()\n",
+      "          (fc2): Linear(in_features=192, out_features=64, bias=True)\n",
+      "          (drop2): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls2): Identity()\n",
+      "        (drop_path2): Identity()\n",
+      "      )\n",
+      "      (1): Block(\n",
+      "        (norm1): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (attn): Attention(\n",
+      "          (qkv): Linear(in_features=64, out_features=192, bias=False)\n",
+      "          (q_norm): Identity()\n",
+      "          (k_norm): Identity()\n",
+      "          (attn_drop): Dropout(p=0.0, inplace=False)\n",
+      "          (proj): Linear(in_features=64, out_features=64, bias=True)\n",
+      "          (proj_drop): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls1): Identity()\n",
+      "        (drop_path1): Identity()\n",
+      "        (norm2): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (mlp): Mlp(\n",
+      "          (fc1): Linear(in_features=64, out_features=192, bias=True)\n",
+      "          (act): GELU(approximate='none')\n",
+      "          (drop1): Dropout(p=0.0, inplace=False)\n",
+      "          (norm): Identity()\n",
+      "          (fc2): Linear(in_features=192, out_features=64, bias=True)\n",
+      "          (drop2): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls2): Identity()\n",
+      "        (drop_path2): Identity()\n",
+      "      )\n",
+      "      (2): Block(\n",
+      "        (norm1): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (attn): Attention(\n",
+      "          (qkv): Linear(in_features=64, out_features=192, bias=False)\n",
+      "          (q_norm): Identity()\n",
+      "          (k_norm): Identity()\n",
+      "          (attn_drop): Dropout(p=0.0, inplace=False)\n",
+      "          (proj): Linear(in_features=64, out_features=64, bias=True)\n",
+      "          (proj_drop): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls1): Identity()\n",
+      "        (drop_path1): Identity()\n",
+      "        (norm2): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (mlp): Mlp(\n",
+      "          (fc1): Linear(in_features=64, out_features=192, bias=True)\n",
+      "          (act): GELU(approximate='none')\n",
+      "          (drop1): Dropout(p=0.0, inplace=False)\n",
+      "          (norm): Identity()\n",
+      "          (fc2): Linear(in_features=192, out_features=64, bias=True)\n",
+      "          (drop2): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls2): Identity()\n",
+      "        (drop_path2): Identity()\n",
+      "      )\n",
+      "      (3): Block(\n",
+      "        (norm1): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (attn): Attention(\n",
+      "          (qkv): Linear(in_features=64, out_features=192, bias=False)\n",
+      "          (q_norm): Identity()\n",
+      "          (k_norm): Identity()\n",
+      "          (attn_drop): Dropout(p=0.0, inplace=False)\n",
+      "          (proj): Linear(in_features=64, out_features=64, bias=True)\n",
+      "          (proj_drop): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls1): Identity()\n",
+      "        (drop_path1): Identity()\n",
+      "        (norm2): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (mlp): Mlp(\n",
+      "          (fc1): Linear(in_features=64, out_features=192, bias=True)\n",
+      "          (act): GELU(approximate='none')\n",
+      "          (drop1): Dropout(p=0.0, inplace=False)\n",
+      "          (norm): Identity()\n",
+      "          (fc2): Linear(in_features=192, out_features=64, bias=True)\n",
+      "          (drop2): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls2): Identity()\n",
+      "        (drop_path2): Identity()\n",
+      "      )\n",
+      "      (4): Block(\n",
+      "        (norm1): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (attn): Attention(\n",
+      "          (qkv): Linear(in_features=64, out_features=192, bias=False)\n",
+      "          (q_norm): Identity()\n",
+      "          (k_norm): Identity()\n",
+      "          (attn_drop): Dropout(p=0.0, inplace=False)\n",
+      "          (proj): Linear(in_features=64, out_features=64, bias=True)\n",
+      "          (proj_drop): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls1): Identity()\n",
+      "        (drop_path1): Identity()\n",
+      "        (norm2): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (mlp): Mlp(\n",
+      "          (fc1): Linear(in_features=64, out_features=192, bias=True)\n",
+      "          (act): GELU(approximate='none')\n",
+      "          (drop1): Dropout(p=0.0, inplace=False)\n",
+      "          (norm): Identity()\n",
+      "          (fc2): Linear(in_features=192, out_features=64, bias=True)\n",
+      "          (drop2): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls2): Identity()\n",
+      "        (drop_path2): Identity()\n",
+      "      )\n",
+      "      (5): Block(\n",
+      "        (norm1): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (attn): Attention(\n",
+      "          (qkv): Linear(in_features=64, out_features=192, bias=False)\n",
+      "          (q_norm): Identity()\n",
+      "          (k_norm): Identity()\n",
+      "          (attn_drop): Dropout(p=0.0, inplace=False)\n",
+      "          (proj): Linear(in_features=64, out_features=64, bias=True)\n",
+      "          (proj_drop): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls1): Identity()\n",
+      "        (drop_path1): Identity()\n",
+      "        (norm2): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "        (mlp): Mlp(\n",
+      "          (fc1): Linear(in_features=64, out_features=192, bias=True)\n",
+      "          (act): GELU(approximate='none')\n",
+      "          (drop1): Dropout(p=0.0, inplace=False)\n",
+      "          (norm): Identity()\n",
+      "          (fc2): Linear(in_features=192, out_features=64, bias=True)\n",
+      "          (drop2): Dropout(p=0.0, inplace=False)\n",
+      "        )\n",
+      "        (ls2): Identity()\n",
+      "        (drop_path2): Identity()\n",
+      "      )\n",
+      "    )\n",
+      "    (norm): LayerNorm((64,), eps=1e-06, elementwise_affine=True)\n",
+      "    (fc_norm): Identity()\n",
+      "    (head_drop): Dropout(p=0.0, inplace=False)\n",
+      "    (head): Linear(in_features=64, out_features=2, bias=True)\n",
+      "  )\n",
+      "  (train_metrics): ModuleDict(\n",
+      "    (accuracy): MulticlassAccuracy()\n",
+      "    (precision): MulticlassPrecision()\n",
+      "    (recall): MulticlassRecall()\n",
+      "    (f1): MulticlassF1Score()\n",
+      "  )\n",
+      "  (val_metrics): ModuleDict(\n",
+      "    (accuracy): MulticlassAccuracy()\n",
+      "    (precision): MulticlassPrecision()\n",
+      "    (recall): MulticlassRecall()\n",
+      "    (f1): MulticlassF1Score()\n",
+      "  )\n",
+      "  (test_metrics): ModuleDict(\n",
+      "    (accuracy): MulticlassAccuracy()\n",
+      "    (precision): MulticlassPrecision()\n",
+      "    (recall): MulticlassRecall()\n",
+      "    (f1): MulticlassF1Score()\n",
+      "  )\n",
+      "  (criterion): CrossEntropyLoss()\n",
+      ")\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "logger.info(f\"Model summary:\\n{model}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def initialize_callbacks(cfg: DictConfig) -> List[L.Callback]:\n",
+    "    \"\"\"Initialize the callbacks based on the configuration.\"\"\"\n",
+    "    if not cfg:\n",
+    "        logger.warning(\"No callback configs found! Skipping..\")\n",
+    "        return callbacks\n",
+    "\n",
+    "    if not isinstance(cfg, DictConfig):\n",
+    "        raise TypeError(\"Callbacks config must be a DictConfig!\")\n",
+    "    callbacks = []\n",
+    "\n",
+    "    # Initialize the model checkpoint callback\n",
+    "    model_checkpoint = ModelCheckpoint(**cfg.callbacks.model_checkpoint)\n",
+    "    callbacks.append(model_checkpoint)\n",
+    "\n",
+    "    # Initialize the early stopping callback\n",
+    "    early_stopping = EarlyStopping(**cfg.callbacks.early_stopping)\n",
+    "    callbacks.append(early_stopping)\n",
+    "\n",
+    "    # Initialize the rich model summary callback\n",
+    "    model_summary = RichModelSummary(**cfg.callbacks.rich_model_summary)\n",
+    "    callbacks.append(model_summary)\n",
+    "\n",
+    "    # Initialize the rich progress bar callback\n",
+    "    progress_bar = RichProgressBar(**cfg.callbacks.rich_progress_bar)\n",
+    "    callbacks.append(progress_bar)\n",
+    "\n",
+    "    return callbacks\n",
+    "\n",
+    "\n",
+    "def initialize_logger(cfg: DictConfig) -> Logger:\n",
+    "    \"\"\"Initialize the logger based on the configuration.\"\"\"\n",
+    "    if not cfg:\n",
+    "        logger.warning(\"No logger configs found! Skipping..\")\n",
+    "        return None\n",
+    "\n",
+    "    if not isinstance(cfg, DictConfig):\n",
+    "        raise TypeError(\"Logger config must be a DictConfig!\")\n",
+    "\n",
+    "    loggers = []\n",
+    "\n",
+    "    # Initialize the TensorBoard logger\n",
+    "    tensorboard_logger = TensorBoardLogger(**cfg.loggers.tensorboard)\n",
+    "    loggers.append(tensorboard_logger)\n",
+    "\n",
+    "    # Initialize the CSV logger\n",
+    "    csv_logger = CSVLogger(**cfg.loggers.csv)\n",
+    "    loggers.append(csv_logger)\n",
+    "\n",
+    "    return loggers"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "########################################## End of the script ##########################################"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "emlo_env",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.15"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

src/datamodules/catdog_datamodule.py

@@ -48,7 +48,8 @@ class CatDogImageDataModule(L.LightningDataModule):
 
     def setup(self, stage: Optional[str] = None):
         """Set up the train, validation, and test datasets."""
-        transform = transforms.Compose(
+
+        train_transform = transforms.Compose(
             [
                 transforms.Resize((self.image_size, self.image_size)),
                 transforms.RandomHorizontalFlip(),
@@ -59,11 +60,21 @@ class CatDogImageDataModule(L.LightningDataModule):
             ]
         )
 
+        test_transform = transforms.Compose(
+            [
+                transforms.Resize((self.image_size, self.image_size)),
+                transforms.ToTensor(),
+                transforms.Normalize(
+                    mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]
+                ),
+            ]
+        )
+
         train_path = self.data_dir / "cats_and_dogs_filtered" / "train"
         test_path = self.data_dir / "cats_and_dogs_filtered" / "validation"
 
         if stage == "fit" or stage is None:
-            full_train_dataset = ImageFolder(root=train_path, transform=transform)
+            full_train_dataset = ImageFolder(root=train_path, transform=train_transform)
             train_size = int(self.train_val_split[0] * len(full_train_dataset))
             val_size = len(full_train_dataset) - train_size
             self.train_dataset, self.val_dataset = random_split(
@@ -74,7 +85,7 @@ class CatDogImageDataModule(L.LightningDataModule):
             )
 
         if stage == "test" or stage is None:
-            self.test_dataset = ImageFolder(root=test_path, transform=transform)
+            self.test_dataset = ImageFolder(root=test_path, transform=test_transform)
             logger.info(f"Test dataset size: {len(self.test_dataset)} images.")
 
     def _create_dataloader(self, dataset, shuffle: bool = False) -> DataLoader:
@@ -123,7 +134,7 @@ if __name__ == "__main__":
             train_val_split=cfg.data.train_val_split,
             pin_memory=cfg.data.pin_memory,
             image_size=cfg.data.image_size,
-            url=cfg.data.dataset_url,
+            url=cfg.data.url,
         )
         datamodule.prepare_data()
         datamodule.setup()

src/models/catdog_model.py

@@ -9,7 +9,7 @@ class ViTTinyClassifier(L.LightningModule):
     def __init__(
         self,
         img_size: int = 224,
-        num_classes: int = 2,
+        num_classes: int = 2,  # Should be 2 for binary classification
         embed_dim: int = 64,
         depth: int = 6,
         num_heads: int = 2,
@@ -40,16 +40,12 @@ class ViTTinyClassifier(L.LightningModule):
             global_pool="token",
         )
 
-        # Metrics for multi-class classification
+        # Metrics for binary classification
         metrics = {
-            "accuracy": Accuracy(task="multiclass", num_classes=num_classes),
-            "precision": Precision(
-                task="multiclass", num_classes=num_classes, average="macro"
-            ),
-            "recall": Recall(
-                task="multiclass", num_classes=num_classes, average="macro"
-            ),
-            "f1": F1Score(task="multiclass", num_classes=num_classes, average="macro"),
+            "acc": Accuracy(task="binary"),
+            "precision": Precision(task="binary"),
+            "recall": Recall(task="binary"),
+            "f1": F1Score(task="binary"),
         }
 
         # Initialize metrics for each stage

src/train.py

@@ -0,0 +1,184 @@
+import os
+import shutil
+from pathlib import Path
+from typing import List
+import torch
+import lightning as L
+from lightning.pytorch.loggers import Logger, TensorBoardLogger, CSVLogger
+from lightning.pytorch.callbacks import (
+    ModelCheckpoint,
+    EarlyStopping,
+    RichModelSummary,
+    RichProgressBar,
+)
+from dotenv import load_dotenv, find_dotenv
+import hydra
+from omegaconf import DictConfig, OmegaConf
+from src.datamodules.catdog_datamodule import CatDogImageDataModule
+from src.utils.logging_utils import setup_logger, task_wrapper
+from loguru import logger
+import rootutils
+
+# Load environment variables
+load_dotenv(find_dotenv(".env"))
+
+# Setup root directory
+try:
+    root = rootutils.setup_root(__file__, indicator=".project-root")
+except Exception as e:
+    root = os.getcwd()
+
+os.environ.setdefault("PROJECT_ROOT", str(root))
+
+
+def initialize_callbacks(cfg: DictConfig) -> List[L.Callback]:
+    """Initialize callbacks based on configuration."""
+    callback_classes = {
+        "model_checkpoint": ModelCheckpoint,
+        "early_stopping": EarlyStopping,
+        "rich_model_summary": RichModelSummary,
+        "rich_progress_bar": RichProgressBar,
+    }
+    return [callback_classes[name](**params) for name, params in cfg.callbacks.items()]
+
+
+def initialize_loggers(cfg: DictConfig) -> List[Logger]:
+    """Initialize loggers based on configuration."""
+    logger_classes = {
+        "tensorboard": TensorBoardLogger,
+        "csv": CSVLogger,
+    }
+    return [logger_classes[name](**params) for name, params in cfg.logger.items()]
+
+
+def load_checkpoint_if_available(ckpt_path: str) -> str:
+    """Return the checkpoint path if available, else None."""
+    if ckpt_path and Path(ckpt_path).exists():
+        logger.info(f"Using checkpoint: {ckpt_path}")
+        return ckpt_path
+    logger.warning(f"Checkpoint not found at {ckpt_path}. Using current model weights.")
+    return None
+
+
+def clear_checkpoint_directory(ckpt_dir: str):
+    """Clear checkpoint directory contents without removing the directory."""
+    ckpt_dir_path = Path(ckpt_dir)
+    if not ckpt_dir_path.exists():
+        logger.info(f"Creating checkpoint directory: {ckpt_dir}")
+        ckpt_dir_path.mkdir(parents=True, exist_ok=True)
+    else:
+        logger.info(f"Clearing checkpoint directory: {ckpt_dir}")
+        for item in ckpt_dir_path.iterdir():
+            try:
+                item.unlink() if item.is_file() else shutil.rmtree(item)
+            except Exception as e:
+                logger.error(f"Failed to delete {item}: {e}")
+
+
+@task_wrapper
+def train_module(
+    data_module: L.LightningDataModule, model: L.LightningModule, trainer: L.Trainer
+):
+    """Train the model and log metrics."""
+    logger.info("Starting training")
+    trainer.fit(model, data_module)
+    train_metrics = trainer.callback_metrics
+    train_acc = train_metrics.get("train_acc")
+    val_acc = train_metrics.get("val_acc")
+    logger.info(
+        f"Training completed. Metrics - train_acc: {train_acc}, val_acc: {val_acc}"
+    )
+    return train_metrics
+
+
+@task_wrapper
+def run_test_module(
+    cfg: DictConfig,
+    datamodule: L.LightningDataModule,
+    model: L.LightningModule,
+    trainer: L.Trainer,
+):
+    """Test the model using the best checkpoint or current model weights."""
+    logger.info("Starting testing")
+    datamodule.setup(stage="test")
+    test_metrics = trainer.test(
+        model, datamodule, ckpt_path=load_checkpoint_if_available(cfg.ckpt_path)
+    )
+    logger.info(f"Test metrics: {test_metrics}")
+    return test_metrics[0] if test_metrics else {}
+
+
+@hydra.main(config_path="../configs", config_name="train", version_base="1.1")
+def setup_run_trainer(cfg: DictConfig):
+    """Set up and run the Trainer for training and testing."""
+    # Display configuration
+    logger.info(f"Config:\n{OmegaConf.to_yaml(cfg)}")
+
+    # Initialize logger
+    log_path = Path(cfg.paths.log_dir) / (
+        "train.log" if cfg.task_name == "train" else "eval.log"
+    )
+    setup_logger(log_path)
+
+    # Display key paths
+    for path_name in [
+        "root_dir",
+        "data_dir",
+        "log_dir",
+        "ckpt_dir",
+        "artifact_dir",
+        "output_dir",
+    ]:
+        logger.info(
+            f"{path_name.replace('_', ' ').capitalize()}: {cfg.paths[path_name]}"
+        )
+
+    # Initialize DataModule and Model
+    logger.info(f"Instantiating datamodule <{cfg.data._target_}>")
+    datamodule: L.LightningDataModule = hydra.utils.instantiate(cfg.data)
+    logger.info(f"Instantiating model <{cfg.model._target_}>")
+    model: L.LightningModule = hydra.utils.instantiate(cfg.model)
+
+    # Check GPU availability and set seed for reproducibility
+    logger.info("GPU available" if torch.cuda.is_available() else "No GPU available")
+    L.seed_everything(cfg.seed, workers=True)
+
+    # Set up callbacks, loggers, and Trainer
+    callbacks = initialize_callbacks(cfg)
+    logger.info(f"Callbacks: {callbacks}")
+    loggers = initialize_loggers(cfg)
+    logger.info(f"Loggers: {loggers}")
+    trainer: L.Trainer = hydra.utils.instantiate(
+        cfg.trainer, callbacks=callbacks, logger=loggers
+    )
+
+    # Training phase
+    train_metrics = {}
+    if cfg.get("train"):
+        clear_checkpoint_directory(cfg.paths.ckpt_dir)
+        train_metrics = train_module(datamodule, model, trainer)
+        (Path(cfg.paths.ckpt_dir) / "train_done.flag").write_text(
+            "Training completed.\n"
+        )
+
+    # Testing phase
+    test_metrics = {}
+    if cfg.get("test"):
+        test_metrics = run_test_module(cfg, datamodule, model, trainer)
+
+    # Combine metrics and extract optimization metric
+    all_metrics = {**train_metrics, **test_metrics}
+    optimization_metric = all_metrics.get(cfg.get("optimization_metric"), 0.0)
+    (
+        logger.warning(
+            f"Optimization metric '{cfg.get('optimization_metric')}' not found. Defaulting to 0."
+        )
+        if optimization_metric == 0.0
+        else logger.info(f"Optimization metric: {optimization_metric}")
+    )
+
+    return optimization_metric
+
+
+if __name__ == "__main__":
+    setup_run_trainer()