rm files

training_dir/params.yaml

-runtime:
-  all_reduce_alg: null
-  batchnorm_spatial_persistent: false
-  dataset_num_private_threads: null
-  default_shard_dim: -1
-  distribution_strategy: mirrored
-  enable_xla: false
-  gpu_thread_mode: null
-  loss_scale: dynamic
-  mixed_precision_dtype: float16
-  num_cores_per_replica: 1
-  num_gpus: 2
-  num_packs: 1
-  per_gpu_thread_count: 0
-  run_eagerly: false
-  task_index: -1
-  tpu: null
-  tpu_enable_xla_dynamic_padder: null
-  worker_hosts: null
-task:
-  evaluation:
-    top_k: 5
-  gradient_clip_norm: 0.0
-  init_checkpoint: ''
-  logging_dir: null
-  losses:
-    l2_weight_decay: 0.0005
-    label_smoothing: 0.0
-    one_hot: true
-  model:
-    add_head_batch_norm: false
-    backbone:
-      darknet:
-        depth_scale: 1.0
-        dilate: false
-        max_level: 5
-        min_level: 3
-        model_id: darknet53
-        use_reorg_input: false
-        use_separable_conv: false
-        width_scale: 1.0
-      type: darknet
-    dropout_rate: 0.0
-    input_size: [256, 256, 3]
-    kernel_initializer: VarianceScaling
-    norm_activation:
-      activation: mish
-      norm_epsilon: 0.001
-      norm_momentum: 0.99
-      use_sync_bn: true
-    num_classes: 1001
-  name: null
-  train_data:
-    aug_policy: null
-    aug_rand_hflip: true
-    aug_type: null
-    block_length: 1
-    cache: false
-    color_jitter: 0.0
-    cycle_length: 10
-    decode_jpeg_only: true
-    decoder:
-      simple_decoder:
-        mask_binarize_threshold: null
-        regenerate_source_id: false
-      type: simple_decoder
-    deterministic: null
-    drop_remainder: true
-    dtype: float16
-    enable_tf_data_service: false
-    file_type: tfrecord
-    global_batch_size: 16
-    image_field_key: image/encoded
-    input_path: ''
-    is_multilabel: false
-    is_training: true
-    label_field_key: image/class/label
-    mixup_and_cutmix: null
-    randaug_magnitude: 10
-    random_erasing: null
-    seed: null
-    sharding: true
-    shuffle_buffer_size: 100
-    tf_data_service_address: null
-    tf_data_service_job_name: null
-    tfds_as_supervised: false
-    tfds_data_dir: ~/tensorflow_datasets/
-    tfds_name: imagenet2012
-    tfds_skip_decoding_feature: ''
-    tfds_split: train
-  validation_data:
-    aug_policy: null
-    aug_rand_hflip: true
-    aug_type: null
-    block_length: 1
-    cache: false
-    color_jitter: 0.0
-    cycle_length: 10
-    decode_jpeg_only: true
-    decoder:
-      simple_decoder:
-        mask_binarize_threshold: null
-        regenerate_source_id: false
-      type: simple_decoder
-    deterministic: null
-    drop_remainder: false
-    dtype: float16
-    enable_tf_data_service: false
-    file_type: tfrecord
-    global_batch_size: 16
-    image_field_key: image/encoded
-    input_path: ''
-    is_multilabel: false
-    is_training: true
-    label_field_key: image/class/label
-    mixup_and_cutmix: null
-    randaug_magnitude: 10
-    random_erasing: null
-    seed: null
-    sharding: true
-    shuffle_buffer_size: 100
-    tf_data_service_address: null
-    tf_data_service_job_name: null
-    tfds_as_supervised: false
-    tfds_data_dir: ~/tensorflow_datasets/
-    tfds_name: imagenet2012
-    tfds_skip_decoding_feature: ''
-    tfds_split: validation
-trainer:
-  allow_tpu_summary: false
-  best_checkpoint_eval_metric: ''
-  best_checkpoint_export_subdir: ''
-  best_checkpoint_metric_comp: higher
-  checkpoint_interval: 10000
-  continuous_eval_timeout: 3600
-  eval_tf_function: true
-  eval_tf_while_loop: false
-  loss_upper_bound: 1000000.0
-  max_to_keep: 5
-  optimizer_config:
-    ema: null
-    learning_rate:
-      polynomial:
-        cycle: false
-        decay_steps: 6392000
-        end_learning_rate: 1.25e-05
-        initial_learning_rate: 0.0125
-        name: PolynomialDecay
-        offset: 0
-        power: 4.0
-      type: polynomial
-    optimizer:
-      sgd:
-        clipnorm: null
-        clipvalue: null
-        decay: 0.0
-        global_clipnorm: null
-        momentum: 0.9
-        name: SGD
-        nesterov: false
-      type: sgd
-    warmup:
-      linear:
-        name: linear
-        warmup_learning_rate: 0
-        warmup_steps: 8000
-      type: linear
-  recovery_begin_steps: 0
-  recovery_max_trials: 0
-  steps_per_loop: 10000
-  summary_interval: 10000
-  train_steps: 6400000
-  train_tf_function: true
-  train_tf_while_loop: true
-  validation_interval: 10000
-  validation_steps: 3200
-  validation_summary_subdir: validation