FJModel / pipeline.py

set self.metric

7fb7cd0 almost 3 years ago

3.48 kB

from typing import Dict, List, Any
from PIL import Image

import os
import json
import numpy as np
from fastai.learner import load_learner


class OrdinalRegressionMetric(Metric):
    def __init__(self):
        super().__init__()
        self.total = 0
        self.count = 0

    def accumulate(self, learn):
        # Get predictions and targets
        preds, targs = learn.pred, learn.y

        # Your custom logic to convert predictions and targets to numeric values
        preds_numeric = torch.argmax(preds, dim=1)
        targs_numeric = targs

        #print("preds_numeric: ",preds_numeric)
        #print("targs_numeric: ",targs_numeric)

        # Calculate the metric (modify this based on your specific needs)
        squared_diff = torch.sum(torch.sqrt((preds_numeric - targs_numeric)**2))

        # Normalize by the maximum possible difference
        max_diff = torch.sqrt((torch.max(targs_numeric) - torch.min(targs_numeric))**2)

        #print("squared_diff: ",squared_diff)
        #print("max_diff: ",max_diff)

        # Update the metric value
        self.total += squared_diff
        #print("self.total: ",self.total)
        self.count += max_diff
        #print("self.count: ",self.count)
    @property
    def value(self):
        if self.count == 0:
            return 0.0  # or handle this case appropriately
        #print("self.total / self.count: ", (self.total / self.count))
        # Calculate the normalized metric value
        metric_value = 1/(self.total / self.count)
        return metric_value

class PreTrainedPipeline():
    def __init__(self, path=""):
        # IMPLEMENT_THIS
        # Preload all the elements you are going to need at inference.
        # For instance your model, processors, tokenizer that might be needed.
        # This function is only called once, so do all the heavy processing I/O here"""
        self.metric = OrdinalRegressionMetric()
        self.model = load_learner(os.path.join(path, "textfile3-2.pk1"))
        with open(os.path.join(path, "config.json")) as config:
            config = json.load(config)
        self.id2label = config["id2label"]

    def __call__(self, inputs: "Image.Image") -> List[Dict[str, Any]]:
        """
        Args:
            inputs (:obj:`PIL.Image`):
                The raw image representation as PIL.
                No transformation made whatsoever from the input. Make all necessary transformations here.
        Return:
            A :obj:`list`:. The list contains items that are dicts should be liked {"label": "XXX", "score": 0.82}
                It is preferred if the returned list is in decreasing `score` order
        """
        # IMPLEMENT_THIS
        # FastAI expects a np array, not a PIL Image.
        _, _, preds = self.model.predict(np.array(inputs))
        preds = preds.tolist()
        labels = [
            {"label": str(self.id2label["0"]), "score": preds[0]},
            {"label": str(self.id2label["1"]), "score": preds[1]},
            {"label": str(self.id2label["2"]), "score": preds[2]},
            {"label": str(self.id2label["3"]), "score": preds[3]},
            {"label": str(self.id2label["4"]), "score": preds[4]},
            {"label": str(self.id2label["5"]), "score": preds[5]},
            {"label": str(self.id2label["6"]), "score": preds[6]},
            {"label": str(self.id2label["7"]), "score": preds[7]},
            {"label": str(self.id2label["8"]), "score": preds[8]},
        ]
        return labels

	from typing import Dict, List, Any
	from PIL import Image

	import os
	import json
	import numpy as np
	from fastai.learner import load_learner


	class OrdinalRegressionMetric(Metric):
	def __init__(self):
	super().__init__()
	self.total = 0
	self.count = 0

	def accumulate(self, learn):
	# Get predictions and targets
	preds, targs = learn.pred, learn.y

	# Your custom logic to convert predictions and targets to numeric values
	preds_numeric = torch.argmax(preds, dim=1)
	targs_numeric = targs

	#print("preds_numeric: ",preds_numeric)
	#print("targs_numeric: ",targs_numeric)

	# Calculate the metric (modify this based on your specific needs)
	squared_diff = torch.sum(torch.sqrt((preds_numeric - targs_numeric)**2))

	# Normalize by the maximum possible difference
	max_diff = torch.sqrt((torch.max(targs_numeric) - torch.min(targs_numeric))**2)

	#print("squared_diff: ",squared_diff)
	#print("max_diff: ",max_diff)

	# Update the metric value
	self.total += squared_diff
	#print("self.total: ",self.total)
	self.count += max_diff
	#print("self.count: ",self.count)
	@property
	def value(self):
	if self.count == 0:
	return 0.0 # or handle this case appropriately
	#print("self.total / self.count: ", (self.total / self.count))
	# Calculate the normalized metric value
	metric_value = 1/(self.total / self.count)
	return metric_value

	class PreTrainedPipeline():
	def __init__(self, path=""):
	# IMPLEMENT_THIS
	# Preload all the elements you are going to need at inference.
	# For instance your model, processors, tokenizer that might be needed.
	# This function is only called once, so do all the heavy processing I/O here"""
	self.metric = OrdinalRegressionMetric()
	self.model = load_learner(os.path.join(path, "textfile3-2.pk1"))
	with open(os.path.join(path, "config.json")) as config:
	config = json.load(config)
	self.id2label = config["id2label"]

	def __call__(self, inputs: "Image.Image") -> List[Dict[str, Any]]:
	"""
	Args:
	inputs (:obj:`PIL.Image`):
	The raw image representation as PIL.
	No transformation made whatsoever from the input. Make all necessary transformations here.
	Return:
	A :obj:`list`:. The list contains items that are dicts should be liked {"label": "XXX", "score": 0.82}
	It is preferred if the returned list is in decreasing `score` order
	"""
	# IMPLEMENT_THIS
	# FastAI expects a np array, not a PIL Image.
	_, _, preds = self.model.predict(np.array(inputs))
	preds = preds.tolist()
	labels = [
	{"label": str(self.id2label["0"]), "score": preds[0]},
	{"label": str(self.id2label["1"]), "score": preds[1]},
	{"label": str(self.id2label["2"]), "score": preds[2]},
	{"label": str(self.id2label["3"]), "score": preds[3]},
	{"label": str(self.id2label["4"]), "score": preds[4]},
	{"label": str(self.id2label["5"]), "score": preds[5]},
	{"label": str(self.id2label["6"]), "score": preds[6]},
	{"label": str(self.id2label["7"]), "score": preds[7]},
	{"label": str(self.id2label["8"]), "score": preds[8]},
	]
	return labels