image_classification/config_file_examples_pt/quantization_config.yaml

operation_mode: quantization

general:
  workers: 4 
  no_prefetcher: true
  display_figures: False 

model: 
  framework: 'torch'
  model_name: 'mobilenetv2_w035_pt' 
  pretrained: True
  pretrained_dataset: "imagenet"
  input_shape: [3, 224, 224]

quantization:
   quantizer: Onnx_quantizer
   quantization_type: PTQ
   quantization_input_type: uint8
   quantization_output_type: float
   export_dir: quantized_models

dataset:
  dataset_name: "imagenet" # options "flowers102", "food101", "imagenet"
  class_names: ''
  classes_file_path: ./datasets/deployment_labels_imagenet.txt
  #data_dir: "/local/datasets/" # there shud be imagenet folder inside this directory # can also be used for quantization
  num_classes: 1000 # change according to your dataset
  #train_split: "train"
  #val_split: "val"
  quantization_split: 0.01
  quantization_path: "/local/datasets/ic_imagenet_2012/val/"
  
preprocessing: 
  rescaling:
    scale: 1/255.0 # TODO scale node is already present under data_augmentation
    offset: 0
  resizing:
    interpolation: nearest # nearest 'Image resize interpolation type (overrides model)'
    aspect_ratio: fit
  color_mode: rgb
  # mean: [0.485, 0.456, 0.406] # 'Override mean pixel value of dataset'
  # std: [0.229, 0.224, 0.225] # 'Override std deviation of dataset'

mlflow:
  uri: ./pt/src/experiments_outputs/mlruns

hydra:
  run:
    dir: ./pt/src/experiments_outputs/${now:%Y_%m_%d_%H_%M_%S}