3.1.1 A quick review of FL input and output interface

Overall

Before

After

load dataset from local
training epochs
save model at local

import flavor # step 1
load dataset from os.environ["INPUT_PATH"] # step 4
SetEvent("TrainInitDone") # step 5
epoch=-1
while True: # step 2
epoch+=1
  WaitEvent("TrainStarted") # step 5
  load model from aggregator by os.environ["GLOBAL_MODEL_PATH"] # step 4
  Prepare an output dictionary # step 3
  SaveInfoJson(dictionary) # step 5
training 1 epoch # step 2
  save model to os.environ["LOCAL_MODEL_PATH"] # step 4
  SetEvent("TrainFinished") # step 5
SetEvent("ProcessFinished") # step 5 (Optional in most cases)

Input interface

INPUT_PATH with examples

Before	After
# import cv2 img = cv2.imread("./dataset/img1.jpg")	# import cv2 img = cv2.imread(os.environ["INPUT_PATH"]+"/dataset/img1.jpg")
# import pandas as pd df = pd.read_csv("./data.csv")	# import pandas as pd df = pd.read_csv(os.environ["INPUT_PATH"]+"/data.csv")

LOCAL_MODEL_PATH with an example

Pytorch	Tensorflow Keras
weights={"state_dict":model.state_dict()}	weights = {"state_dict": {str(key): val for key,val in enumerate(model.get_weights())}}
torch.save(weights, os.environ["LOCAL_MODEL_PATH"])	with open(os.environ["LOCAL_MODEL_PATH"], "wb") as F: pickle.dump(weights, F, protocol=pickle.HIGHEST_PROTOCOL)

GLOBAL_MODEL_PATH with an example

if epoch != 0 or os.path.exists(os.environ["GLOBAL_MODEL_PATH"]):

Pytorch	Tensorflow Keras
model.load_state_dict( torch.load( os.environ["GLOBAL_MODEL_PATH"])["state_dict"])	with open(os.environ["GLOBAL_MODEL_PATH"], "rb") as F: weights = pickle.load(F) model.set_weights(list(weights["state_dict"].values()))

Output interface

Prepare a Json file that has the following format and save it to OUTPUT_PATH

metadata: dict # Must
- epoch: int
- datasetSize: int # The number of training data.
- importance: float # Epoch weight. Set equal weights 1.0 in default
metrics: dict # These parameters are required ,you can add any other keys (str) with values (float).
- precision: float
- basic/confusion_tp: float
- basic/confusion_fp: float
- basic/confusion_fn: float
- basic/confusion_tn: float