utils_optical_flow.py

import os
import pickle
import numpy as np
import pandas as pd
import random
import torch
import math
from torch.autograd import Variable
from helper import *

class DataLoader():

    def __init__(self,f_prefix, batch_size=4, seq_length=10, num_of_validation = 0, forcePreProcess=False, infer=False, generate = False):
        '''
        Initialiser function for the DataLoader class
        params:
        batch_size : Size of the mini-batch
        seq_length : Sequence length to be considered
        num_of_validation : number of validation dataset will be used
        infer : flag for test mode
        generate : flag for data generation mode
        forcePreProcess : Flag to forcefully preprocess the data again from csv files
        '''
        # test_files_path = 'data/test/IRVLab/'
        # train_files_path = 'data/train/IRVLab/'

        # for (_, _, filenames) in os.walk(test_files_path):
        #     base_test_dataset = filenames
        #     break
        # for (_, _, filenames) in os.walk(train_files_path):
        #     base_train_dataset = filenames

        # base test files
        base_test_dataset=  [ '/mnt/data1/vdd_optical_flow_labels/test/IRVLab/pool_swimmer_004_A.txt'#pool_flipper_001_A.txt'#, '/data/test/IRVLab/pool_flipper_003_A_0009.txt'#/data/test/biwi/biwi_eth.txt',
                        # '/data/test/crowds/crowds_zara01.txt',
                        # '/data/test/crowds/uni_examples.txt',
                        # '/data/test/stanford/coupa_0.txt',
                        #  '/data/test/stanford/coupa_1.txt', '/data/test/stanford/gates_2.txt','/data/test/stanford/hyang_0.txt','/data/test/stanford/hyang_1.txt','/data/test/stanford/hyang_3.txt','/data/test/stanford/hyang_8.txt',
                        #   '/data/test/stanford/little_0.txt','/data/test/stanford/little_1.txt','/data/test/stanford/little_2.txt','/data/test/stanford/little_3.txt','/data/test/stanford/nexus_5.txt','/data/test/stanford/nexus_6.txt',
                        #   '/data/test/stanford/quad_0.txt','/data/test/stanford/quad_1.txt','/data/test/stanford/quad_2.txt','/data/test/stanford/quad_3.txt'
                          ]
        #base train files
        base_train_dataset = ['/mnt/data1/vdd_optical_flow_labels/train/IRVLab/pool_swimmer_001_A.txt'#pool_flipper_003_A_0001.txt'#, '/data/test/IRVLab/pool_flipper_003_A_0004.txt'#/data/train/biwi/biwi_hotel.txt',
                        #'/data/train/crowds/arxiepiskopi1.txt','/data/train/crowds/crowds_zara02.txt',
                        #'/data/train/crowds/crowds_zara03.txt','/data/train/crowds/students001.txt','/data/train/crowds/students003.txt',
                        #'/data/train/mot/PETS09-S2L1.txt',
                        #'/data/train/stanford/bookstore_0.txt','/data/train/stanford/bookstore_1.txt','/data/train/stanford/bookstore_2.txt','/data/train/stanford/bookstore_3.txt','/data/train/stanford/coupa_3.txt','/data/train/stanford/deathCircle_0.txt','/data/train/stanford/deathCircle_1.txt','/data/train/stanford/deathCircle_2.txt','/data/train/stanford/deathCircle_3.txt',
                        #'/data/train/stanford/deathCircle_4.txt','/data/train/stanford/gates_0.txt','/data/train/stanford/gates_1.txt','/data/train/stanford/gates_3.txt','/data/train/stanford/gates_4.txt','/data/train/stanford/gates_5.txt','/data/train/stanford/gates_6.txt','/data/train/stanford/gates_7.txt','/data/train/stanford/gates_8.txt','/data/train/stanford/hyang_4.txt',
                        #'/data/train/stanford/hyang_5.txt','/data/train/stanford/hyang_6.txt','/data/train/stanford/hyang_9.txt','/data/train/stanford/nexus_0.txt','/data/train/stanford/nexus_1.txt','/data/train/stanford/nexus_2.txt','/data/train/stanford/nexus_3.txt','/data/train/stanford/nexus_4.txt','/data/train/stanford/nexus_7.txt','/data/train/stanford/nexus_8.txt','/data/train/stanford/nexus_9.txt'
                        ]
        # dimensions of each file set
        self.dataset_dimensions = {'IRVLab':[320,240]}

        # List of data directories where raw data resides
        self.base_train_path = '/mnt/data1/vdd_optical_flow_labels/train/'
        self.base_test_path = '/mnt/data1/vdd_optical_flow_labels/test/'
        self.base_validation_path = '/mnt/data1/vdd_optical_flow_labels/valid/'

        # check infer flag, if true choose test directory as base directory
        if infer is False:
            self.base_data_dirs = base_train_dataset
        else:
            self.base_data_dirs = base_test_dataset

        # get all files using python os and base directories
        self.train_dataset = self.get_dataset_path(self.base_train_path, f_prefix)
        self.test_dataset = self.get_dataset_path(self.base_test_path, f_prefix)
        self.validation_dataset = self.get_dataset_path(self.base_validation_path, f_prefix)


        # if generate mode, use directly train base files
        if generate:
            self.train_dataset = [os.path.join(f_prefix, dataset[1:]) for dataset in base_train_dataset]

        #request of use of validation dataset
        if num_of_validation>0:
            self.additional_validation = True
        else:
            self.additional_validation = False

        # check validation dataset availibility and clip the reuqested number if it is bigger than available validation dataset
        if self.additional_validation:
            if len(self.validation_dataset) == 0:
                print("There is no validation dataset.Aborted.")
                self.additional_validation = False
            else:
                num_of_validation = np.clip(num_of_validation, 0, len(self.validation_dataset))
                self.validation_dataset = random.sample(self.validation_dataset, num_of_validation)

        # if not infer mode, use train dataset
        if infer is False:
            self.data_dirs = self.train_dataset
        else:
            # use validation dataset
            if self.additional_validation:
                self.data_dirs = self.validation_dataset
            # use test dataset
            else:
                self.data_dirs = self.test_dataset


        self.infer = infer
        self.generate = generate

        # Number of datasets
        self.numDatasets = len(self.data_dirs)

        # array for keepinng target ped ids for each sequence
        self.target_ids = []

        # Data directory where the pre-processed pickle file resides
        self.train_data_dir = os.path.join(f_prefix, self.base_train_path)
        self.test_data_dir = os.path.join(f_prefix, self.base_test_path)
        self.val_data_dir = os.path.join(f_prefix, self.base_validation_path)


        # Store the arguments
        self.batch_size = batch_size
        self.seq_length = seq_length
        self.orig_seq_lenght = seq_length

        # Validation arguments
        self.val_fraction = 0

        # Define the path in which the process data would be stored
        self.data_file_tr = os.path.join(self.train_data_dir, "trajectories_train.cpkl")
        self.data_file_te = os.path.join(self.base_test_path, "trajectories_test.cpkl")
        self.data_file_vl = os.path.join(self.val_data_dir, "trajectories_val.cpkl")


        # for creating a dict key: folder names, values: files in this folder
        self.create_folder_file_dict()

        if self.additional_validation:
        # If the file doesn't exist or forcePreProcess is true
            if not(os.path.exists(self.data_file_vl)) or forcePreProcess:
                print("Creating pre-processed validation data from raw data")
                # Preprocess the data from the csv files of the datasets
                # Note that this data is processed in frames
                self.frame_preprocess(self.validation_dataset, self.data_file_vl, self.additional_validation)

        if self.infer:
        # if infer mode, and no additional files -> test preprocessing
            if not self.additional_validation:
                if not(os.path.exists(self.data_file_te)) or forcePreProcess:
                    print("Creating pre-processed test data from raw data")
                    # Preprocess the data from the csv files of the datasets
                    # Note that this data is processed in frames
                    print("Working on directory: ", self.data_file_te)
                    self.frame_preprocess(self.data_dirs, self.data_file_te)
            # if infer mode, and there are additional validation files -> validation dataset visualization
            else:
                print("Validation visualization file will be created")

        # if not infer mode
        else:
            # If the file doesn't exist or forcePreProcess is true -> training pre-process
            if not(os.path.exists(self.data_file_tr)) or forcePreProcess:
                print("Creating pre-processed training data from raw data")
                # Preprocess the data from the csv files of the datasets
                # Note that this data is processed in frames
                self.frame_preprocess(self.data_dirs, self.data_file_tr)

        if self.infer:
            # Load the processed data from the pickle file
            if not self.additional_validation: #test mode
                #print("Called test")
                self.load_preprocessed(self.data_file_te)
            else:  # validation mode
                #print("Called validation")
                self.load_preprocessed(self.data_file_vl, True)

        else: # training mode
            #print("Called train")
            self.load_preprocessed(self.data_file_tr)

        # Reset all the data pointers of the dataloader object
        self.reset_batch_pointer(valid=False)
        self.reset_batch_pointer(valid=True)

    def frame_preprocess(self, data_dirs, data_file, validation_set = False):
        '''
        Function that will pre-process the pixel_pos.csv files of each dataset
        into data with occupancy grid that can be used
        params:
        data_dirs : List of directories where raw data resides
        data_file : The file into which all the pre-processed data needs to be stored
        validation_set: true when a dataset is in validation set
        '''
        # all_frame_data would be a list of list of numpy arrays corresponding to each dataset
        # Each numpy array will correspond to a frame and would be of size (numPeds, 3) each row
        # containing pedID, x, y
        all_frame_data = []
        # Validation frame data
        valid_frame_data = []
        # frameList_data would be a list of lists corresponding to each dataset
        # Each list would contain the frameIds of all the frames in the dataset
        frameList_data = []
        valid_numPeds_data= []
        # numPeds_data would be a list of lists corresponding to each dataset
        # Each list would contain the number of pedestrians in each frame in the dataset
        numPeds_data = []


        #each list includes ped ids of this frame
        pedsList_data = []
        valid_pedsList_data = []
        # target ped ids for each sequence
        target_ids = []
        orig_data = []



        # Index of the current dataset
        dataset_index = 0

        # For each dataset
        for directory in data_dirs:

            # Load the data from the txt file
            print("Now processing: ", directory)
            column_names = ['frame_num','ped_id','xmin','ymin','xmax','ymax', 'xFlow', 'yFlow']

            # if training mode, read train file to pandas dataframe and process
            if self.infer is False:
                df = pd.read_csv(directory, dtype={'frame_num':'int','ped_id':'int', 'xFlow':'float', 'yFlow':'float'}, delimiter = ' ',  header=None, names=column_names)
                self.target_ids = np.array(df.drop_duplicates(subset={'ped_id'}, keep='first', inplace=False)['ped_id'])


            else:
                # if validation mode, read validation file to pandas dataframe and process
                if self.additional_validation:
                    df = pd.read_csv(directory, dtype={'frame_num':'int','ped_id':'int', 'xFlow':'float', 'yFlow':'float' }, delimiter = ' ',  header=None, names=column_names)
                    self.target_ids = np.array(df.drop_duplicates(subset={'ped_id'}, keep='first', inplace=False)['ped_id'])

                # if test mode, read test file to pandas dataframe and process
                else:
                    column_names = ['frame_num','ped_id','xmin','ymin','xmax','ymax','xFlow', 'yFlow']
                    df = pd.read_csv(directory, dtype={'frame_num':'int','ped_id':'int', 'xFlow':'float', 'yFlow':'float'}, delimiter = ' ',  header=None, names=column_names, converters = {c:lambda x: float('nan') if x == '?' else float(x) for c in ['xmin','ymin','xmax','ymax']})
                    #self.target_ids = np.array(df[df['ymin'].isnull()].drop_duplicates(subset={'ped_id'}, keep='first', inplace=False)['ped_id'])
                    self.target_ids = np.array(df.drop_duplicates(subset={'ped_id'}, keep='first', inplace=False)['ped_id'])
            # convert pandas -> numpy array
            data = np.array(df)
            #print("DATA",data)

            # keep original copy of file
            orig_data.append(data)

            #swap x and y points (in txt file it is like -> y,x)
            data = np.swapaxes(data,0,1)

            frameList = []
             # get frame numbers
            for frameNum in data[0,:].tolist():
                if frameNum not in frameList:
                    frameList.append(frameNum)
            #frameList = data[0, :]
            # Number of frames
            numFrames = len(frameList)
            #print("FRAME NUMS",numFrames)

            # Add the list of frameIDs to the frameList_data
            frameList_data.append(frameList)

            # Initialize the list of numPeds for the current dataset
            numPeds_data.append([])
            valid_numPeds_data.append([])
            # Initialize the list of numPeds for the current dataset
            numPeds_data.append([])
            valid_numPeds_data.append([])

            # Initialize the list of numpy arrays for the current dataset
            all_frame_data.append([])
            # Initialize the list of numpy arrays for the current dataset
            valid_frame_data.append([])

            # list of peds for each frame
            pedsList_data.append([])
            valid_pedsList_data.append([])

            target_ids.append(self.target_ids)



            for ind, frame in enumerate(frameList):


                # Extract all pedestrians in current frame
                pedsInFrame = data[: , data[0, :] == frame]
                #print("peds in %d: %s"%(frame,str(pedsInFrame)))



                # Extract peds list
                pedsList = pedsInFrame[1, :].tolist()

                # Add number of peds in the current frame to the stored data


                # Initialize the row of the numpy array
                pedsWithPos = []

                # For each ped in the current frame
                for ped in pedsList:
                    # Extract their x and y positions
                    current_xmin = pedsInFrame[2, pedsInFrame[1, :] == ped][0]
                    current_ymin = pedsInFrame[3, pedsInFrame[1, :] == ped][0]
                    current_xmax = pedsInFrame[4, pedsInFrame[1, :] == ped][0]
                    current_ymax = pedsInFrame[5, pedsInFrame[1, :] == ped][0]
                    xFlow = pedsInFrame[6, pedsInFrame[1, :] == ped][0]
                    yFlow = pedsInFrame[7, pedsInFrame[1, :] == ped][0]
                    #print(current_xmin,current_ymin,current_xmax,current_ymax)

                    # Add their pedID, x, y to the row of the numpy array
                    pedsWithPos.append([ped, current_xmin, current_ymin, current_xmax, current_ymax, xFlow, yFlow])

                # At inference time, data generation and if dataset is a validation dataset, no validation data
                if (ind >= numFrames * self.val_fraction) or (self.infer) or (self.generate) or (validation_set):
                    # Add the details of all the peds in the current frame to all_frame_data
                    all_frame_data[dataset_index].append(np.array(pedsWithPos))
                    pedsList_data[dataset_index].append(pedsList)
                    numPeds_data[dataset_index].append(len(pedsList))


                else:
                    valid_frame_data[dataset_index].append(np.array(pedsWithPos))
                    valid_pedsList_data[dataset_index].append(pedsList)
                    valid_numPeds_data[dataset_index].append(len(pedsList))



            dataset_index += 1
        # Save the arrays in the pickle file
        f = open(data_file, "wb")
        pickle.dump((all_frame_data, frameList_data, numPeds_data, valid_numPeds_data, valid_frame_data, pedsList_data, valid_pedsList_data, target_ids, orig_data), f, protocol=2)
        f.close()
        #print("Done here")


    def load_preprocessed(self, data_file, validation_set = False):
        '''
        Function to load the pre-processed data into the DataLoader object
        params:
        data_file : the path to the pickled data file
        validation_set : flag for validation dataset
        '''
        # Load data from the pickled file
        if(validation_set):
            print("Loading validaton datasets: ", data_file)
        else:
            print("Loading train or test dataset: ", data_file)
        print("DATA FILE************",data_file)
        f = open(data_file, 'rb')
        self.raw_data = pickle.load(f)
        #print(len(self.raw_data))

        f.close()

        # Get all the data from the pickle file
        self.data = self.raw_data[0]
        self.frameList = self.raw_data[1]
        self.numPedsList = self.raw_data[2]
        self.valid_numPedsList = self.raw_data[3]
        self.valid_data = self.raw_data[4]
        self.pedsList = self.raw_data[5]
        self.valid_pedsList = self.raw_data[6]
        self.target_ids = self.raw_data[7]
        self.orig_data = self.raw_data[8]

        counter = 0
        valid_counter = 0
        print('Sequence size(frame) ------>',self.seq_length)
        print('One batch size (frame)--->-', self.batch_size*self.seq_length)

        # For each dataset
        for dataset in range(len(self.data)):
            # get the frame data for the current dataset
            all_frame_data = self.data[dataset]
            valid_frame_data = self.valid_data[dataset]
            dataset_name = self.data_dirs[dataset].split('/')[-1]
            # calculate number of sequence
            num_seq_in_dataset = int(len(all_frame_data) / (self.seq_length+1))
            num_valid_seq_in_dataset = int(len(valid_frame_data) / (self.seq_length))
            if not validation_set:
                print('Training data from training dataset(name, # frame, #sequence)--> ', dataset_name, ':', len(all_frame_data),':', (num_seq_in_dataset))
                print('Validation data from training dataset(name, # frame, #sequence)--> ', dataset_name, ':', len(valid_frame_data),':', (num_valid_seq_in_dataset))
            else:
                print('Validation data from validation dataset(name, # frame, #sequence)--> ', dataset_name, ':', len(all_frame_data),':', (num_seq_in_dataset))

            # Increment the counter with the number of sequences in the current dataset
            counter += num_seq_in_dataset
            valid_counter += num_valid_seq_in_dataset

        # Calculate the number of batches
        self.num_batches = int(counter/self.batch_size)
        # print('***************************************')
        # print('Num Batches',self.num_batches)
        # print('Counter',counter)
        # print('Batch Size',self.batch_size)
        # print('***************************************')
        self.valid_num_batches = int(valid_counter/self.batch_size)


        if not validation_set:
            print('Total number of training batches:', self.num_batches)
            print('Total number of validation batches:', self.valid_num_batches)
        else:
            print('Total number of validation batches:', self.num_batches)

        # self.valid_num_batches = self.valid_num_batches * 2

    def adjustFramesForOpticalFlow(self, seqFrames):
        """
        seqFrames: Shape (sequenceLength+1, numDivers, 7)
        return transformed frames (sequenceLength+1, numDivers,5)
        """
        orig_boxes = seqFrames[:,:,0:4]
        # Ignore the transform of the first frame since we don't care about it's transform with the previous frame
        # Need only 0-th index of numDivers, since the transform for each diver is the same
        transforms = np.reshape(seqFrames[1:,0,4:], (orig_boxes.shape[0]-1,2))
        return self.transformedBoxes(orig_boxes, transforms)
    
    def transformedBoxes(self, orig_boxes, transforms):
        """
        Returns the boxes as seen from the frame of reference of the last frame
        orig_boxes: original bounding boxes - each in its own frame of reference. Shape (sequenceLength+1, numDivers, 5)
        transforms: The transform gives the deviation of the current frame with respect to the previous frame. shape: (seqLength+1, 2)
        """
        transformedFrames = np.zeros_like(orig_boxes, dtype=float)
        currIdx = transforms.shape[0] - 1
        currDeltaSum = np.zeros((2), dtype=float)
        # Last frame unchanged
        transformedFrames[currIdx+1,:,:] = orig_boxes[-1,:,:]
    
        while currIdx >= 0:
            currDeltaSum += transforms[currIdx]
            concatenatedTransform = np.concatenate((currDeltaSum, currDeltaSum))
            for i in range(orig_boxes.shape[1]):
                if (False in (orig_boxes[currIdx,i] == 0)):
                    transformedFrames[currIdx, i] = orig_boxes[currIdx,i] + concatenatedTransform
            #transformedFrames[currIdx,:] = orig_boxes[currIdx,:] + concatenatedTransform
            currIdx -= 1
        return transformedFrames

    def next_batch(self):
        '''
        Function to get the next batch of points
        '''
        # Source data
        x_batch = []
        # Target data
        y_batch = []
        # Dataset data
        d = []

        # pedlist per sequence
        numPedsList_batch = []

        # pedlist per sequence
        PedsList_batch = []

        #return target_id
        target_ids = []

        # Iteration index
        i = 0
        #print("BATCH_SIZE",self.batch_size)
        while i < self.batch_size:
            # Extract the frame data of the current dataset
            frame_data = self.data[self.dataset_pointer]
            numPedsList = self.numPedsList[self.dataset_pointer]
            pedsList = self.pedsList[self.dataset_pointer]
            # Get the frame pointer for the current dataset
            idx = self.frame_pointer
            # While there is still seq_length number of frames left in the current dataset
            # Instead of returning an x array of seq length, we will return an array of len seq_length + 1
            # of this, x[:-1] becomes the training instance, x[1:] becomes the target
            # That is, the output sequence is expected to be the future prediction of the input sequence
            if idx + self.seq_length + 1 <= len(frame_data):
                # All the data in this sequence
                seq_source_frame_data = frame_data[idx:idx+self.seq_length+1]
                seq_numPedsList = numPedsList[idx:idx+self.seq_length+1]
                seq_PedsList = pedsList[idx:idx+self.seq_length+1]
                seq_target_frame_data = frame_data[idx+1:idx+self.seq_length+1]

                # Number of unique peds in this sequence of frames
                x_batch.append(seq_source_frame_data)
                y_batch.append(seq_target_frame_data)
                numPedsList_batch.append(seq_numPedsList)
                PedsList_batch.append(seq_PedsList)
                # get correct target ped id for the sequence
                #print("******************")

                #print(self.frame_pointer,self.seq_length)
                #print("TARGET_IDS PRIVATE",self.target_ids)

                #print(self.target_ids[self.dataset_pointer][math.floor((self.frame_pointer)/self.seq_length)])
                #target_ids.append(self.target_ids[self.dataset_pointer][math.floor((self.frame_pointer)/self.seq_length)])
                self.frame_pointer += (self.seq_length + 1)

                d.append(self.dataset_pointer)
                i += 1

            else:
                # Not enough frames left
                # Increment the dataset pointer and set the frame_pointer to zero
                self.tick_batch_pointer(valid=False)

        #print("TARGET IDS IN NEXT BATCH",target_ids)
        #pedSeq = []
        #for pedSeq in PedsList_batch

        return np.array(x_batch), y_batch, d, numPedsList_batch, np.array(PedsList_batch), target_ids


    def next_valid_batch(self):
        '''
        Function to get the next Validation batch of points
        '''
        # Source data
        x_batch = []
        # Target data
        y_batch = []
        # Dataset data
        d = []

        # pedlist per sequence
        numPedsList_batch = []

         # pedlist per sequence
        PedsList_batch = []
        target_ids = []


        # Iteration index
        i = 0
        while i < self.batch_size:
            # Extract the frame data of the current dataset
            frame_data = self.valid_data[self.valid_dataset_pointer]
            numPedsList = self.valid_numPedsList[self.valid_dataset_pointer]
            pedsList = self.valid_pedsList[self.valid_dataset_pointer]


            # Get the frame pointer for the current dataset
            idx = self.valid_frame_pointer
            # While there is still seq_length number of frames left in the current dataset
            if idx + self.seq_length < len(frame_data):
                # All the data in this sequence
                # seq_frame_data = frame_data[idx:idx+self.seq_length+1]
                seq_source_frame_data = frame_data[idx:idx+self.seq_length]
                seq_numPedsList=numPedsList[idx:idx+self.seq_length]
                seq_PedsList = pedsList[idx:idx+self.seq_length]
                seq_target_frame_data = frame_data[idx+1:idx+self.seq_length+1]

                # Number of unique peds in this sequence of frames
                x_batch.append(seq_source_frame_data)
                y_batch.append(seq_target_frame_data)
                numPedsList_batch.append(seq_numPedsList)
                PedsList_batch.append(seq_PedsList)
                # get correct target ped id for the sequence
                target_ids.append(self.target_ids[self.dataset_pointer][math.floor((self.valid_frame_pointer)/self.seq_length)])
                self.valid_frame_pointer += self.seq_length

                d.append(self.valid_dataset_pointer)
                i += 1

            else:
                # Not enough frames left
                # Increment the dataset pointer and set the frame_pointer to zero
                self.tick_batch_pointer(valid=True)

        return np.array(x_batch), y_batch, d, numPedsList_batch, PedsList_batch, target_ids


    def tick_batch_pointer(self, valid=False):
        '''
        Advance the dataset pointer
        '''

        if not valid:

            # Go to the next dataset
            self.dataset_pointer += 1
            # Set the frame pointer to zero for the current dataset
            self.frame_pointer = 0
            # If all datasets are done, then go to the first one again
            if self.dataset_pointer >= len(self.data):
                print("OUT OF DATA GOING TO ORIGINAL")
                self.dataset_pointer = 0
            print("*******************")
            print("now processing: %s"% self.get_file_name())
        else:
            # Go to the next dataset
            self.valid_dataset_pointer += 1
            # Set the frame pointer to zero for the current dataset
            self.valid_frame_pointer = 0
            # If all datasets are done, then go to the first one again
            if self.valid_dataset_pointer >= len(self.valid_data):
                self.valid_dataset_pointer = 0
            print("*******************")
            print("now processing: %s"% self.get_file_name(pointer_type = 'valid'))

    def reset_batch_pointer(self, valid=False):
        '''
        Reset all pointers
        '''
        if not valid:
            # Go to the first frame of the first dataset
            self.dataset_pointer = 0
            self.frame_pointer = 0
        else:
            self.valid_dataset_pointer = 0
            self.valid_frame_pointer = 0

    def switch_to_dataset_type(self, train = False, load_data = True):
        # function for switching between train and validation datasets during training session
        print('--------------------------------------------------------------------------')
        if not train: # if train mode, switch to validation mode
            if self.additional_validation:
                print("Dataset type switching: training ----> validation")
                self.orig_seq_lenght, self.seq_length = self.seq_length, self.orig_seq_lenght
                self.data_dirs = self.validation_dataset
                self.numDatasets = len(self.data_dirs)
    
                if load_data:
                    self.load_preprocessed(self.data_file_vl, True)
                    self.reset_batch_pointer(valid=False)
            else:
                print("There is no validation dataset.Aborted.")
                return
        else:# if validation mode, switch to train mode
            print("Dataset type switching: validation -----> training")
            self.orig_seq_lenght, self.seq_length = self.seq_length, self.orig_seq_lenght
            self.data_dirs = self.train_dataset
            self.numDatasets = len(self.data_dirs)
            if load_data:
                self.load_preprocessed(self.data_file_tr)
                self.reset_batch_pointer(valid=False)
                self.reset_batch_pointer(valid=True)


    def convert_proper_array(self, x_seq, num_pedlist, pedlist):
        #converter function to appropriate format. Instead of direcly use ped ids, we are mapping ped ids to
        #array indices using a lookup table for each sequence -> speed
        #output: seq_lenght (real sequence lenght+1)*max_ped_id+1 (biggest id number in the sequence)*2 (x,y)
        num_inputs = 6
        #get unique ids from sequence
        unique_ids = pd.unique(np.concatenate(pedlist).ravel().tolist()).astype(int)
        # create a lookup table which maps ped ids -> array indices
        lookup_table = dict(zip(unique_ids, range(0, len(unique_ids))))

        seq_data = np.zeros(shape=(x_seq.shape[0], len(lookup_table), num_inputs))


        # create new structure of array
        for ind, frame in enumerate(x_seq):
            #print("FRAME",frame[:,1:5])
            corr_index = [lookup_table[x] for x in frame[:, 0]]
            #print("CORR_INDEX",corr_index)
            #print("SEQ DATA SHAPE", seq_data[ind, corr_index,:].shape)
            #print("FRAME DATA SHAPE",frame[:,1:5].shape)
            seq_data[ind, corr_index,:] = frame[:,1:(num_inputs+1)]

        x_seq = self.adjustFramesForOpticalFlow(seq_data)
        return_arr = Variable(torch.from_numpy(np.array(x_seq)).float())

        return return_arr, lookup_table

    def add_element_to_dict(self, dict, key, value):
        # helper function to add a element to dictionary
        dict.setdefault(key, [])
        dict[key].append(value)

    def get_dataset_path(self, base_path, f_prefix):
        # get all datasets from given set of directories
        dataset = []
        dir_names = unique_list(self.get_all_directory_namelist())
        for dir_ in dir_names:
            dir_path = os.path.join(f_prefix, base_path, dir_)
            file_names = get_all_file_names(dir_path)
            [dataset.append(os.path.join(dir_path, file_name)) for file_name in file_names]
        return dataset

    def get_file_name(self, offset=0, pointer_type = 'train'):
        #return file name of processing or pointing by dataset pointer
        if pointer_type == 'train':
            return self.data_dirs[self.dataset_pointer+offset].split('/')[-1]

        elif pointer_type == 'valid':
            return self.data_dirs[self.valid_dataset_pointer+offset].split('/')[-1]

    def create_folder_file_dict(self):
        # create a helper dictionary folder name:file name
        self.folder_file_dict = {}
        for dir_ in self.base_data_dirs:
            folder_name = dir_.split('/')[-2]
            file_name = dir_.split('/')[-1]
            self.add_element_to_dict(self.folder_file_dict, folder_name, file_name)


    def get_directory_name(self, offset=0):
        #return folder name of file of processing or pointing by dataset pointer
        folder_name = self.data_dirs[self.dataset_pointer+offset].split('/')[-2]
        return folder_name

    def get_directory_name_with_pointer(self, pointer_index):
        # get directory name using pointer index
        folder_name = self.data_dirs[pointer_index].split('/')[-2]
        return folder_name

    def get_all_directory_namelist(self):
        #return all directory names in this collection of dataset
        folder_list = [data_dir.split('/')[-2] for data_dir in (self.base_data_dirs)]
        return folder_list

    def get_file_path(self, base, prefix, model_name ='', offset=0):
        #return file path of file of processing or pointing by dataset pointer
        folder_name = self.data_dirs[self.dataset_pointer+offset].split('/')[-2]
        base_folder_name=os.path.join(prefix, base, model_name, folder_name)
        return base_folder_name

    def get_base_file_name(self, key):
        # return file name using folder- file dictionary
        return self.folder_file_dict[key]

    def get_len_of_dataset(self):
        # return the number of dataset in the mode
        return len(self.data)

    def clean_test_data(self, x_seq, target_id, obs_lenght, predicted_lenght):
        #remove (pedid, x , y) array if x or y is nan for each frame in observed part (for test mode)
        for frame_num in range(obs_lenght):
            nan_elements_index = np.where(np.isnan(x_seq[frame_num][:, 4]))

            try:
                x_seq[frame_num] = np.delete(x_seq[frame_num], nan_elements_index[0], axis=0)
            except ValueError:
                print("an error has been occured")
                pass

        for frame_num in range(obs_lenght, obs_lenght+predicted_lenght):
            nan_elements_index = x_seq[frame_num][:, 0] != target_id

            try:
                x_seq[frame_num] = x_seq[frame_num][~nan_elements_index]

            except ValueError:
                pass


    def clean_ped_list(self, x_seq, pedlist_seq, target_id, obs_lenght, predicted_lenght):
        # remove peds from pedlist after test cleaning
        target_id_arr = [target_id]
        for frame_num in range(obs_lenght+predicted_lenght):
            pedlist_seq[frame_num] = x_seq[frame_num][:,0]

    def write_to_file(self, data, base, f_prefix, model_name):
        # write all files as txt format
        self.reset_batch_pointer()
        for file in range(self.numDatasets):
            path = self.get_file_path(f_prefix, base, model_name, file)
            file_name = self.get_file_name(file)
            self.write_dataset(data[file], file_name, path)

    def write_dataset(self, dataset_seq, file_name, path):
        # write a file in txt format
        print("Writing to file  path: %s, file_name: %s"%(path, file_name))
        out = np.concatenate(dataset_seq, axis = 0)
        #np.savetxt(os.path.join(path, file_name), out, fmt = "%1d %1.1f %.3f %.3f", newline='\n')
        np.savetxt(os.path.join(path, file_name), out, fmt = "%1d %1.1f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f %.3f", newline='\n')

    def write_to_plot_file(self, data, path):
        # write plot file for further visualization in pkl format
        self.reset_batch_pointer()
        print("Length of data:", len(data))
        print("Dataloader.numDatasets:", self.numDatasets)
        print("Now starting loop")
        for file in range(self.numDatasets):
            print("In iteration:", file)
            print("Length of data currently:", len(data))
            file_name = self.get_file_name(file)
            file_name = file_name.split('.')[0] + '.pkl'
            print("Writing to plot file  path: %s, file_name: %s"%(path, file_name))
            with open(os.path.join(path, file_name), 'wb') as f:
                pickle.dump(data[file], f)

    def get_frame_sequence(self, frame_lenght):
        #begin and end of predicted fram numbers in this seq.
        begin_fr = (self.frame_pointer - frame_lenght)
        end_fr = (self.frame_pointer)
        #frame_number = self.orig_data[self.dataset_pointer][begin_fr:end_fr, 0].transpose()
        frameNum = self.frameList[self.dataset_pointer][begin_fr]#.transpose()
        frame_number = np.reshape([float(i) for i in range(int(frameNum),int(frameNum) + frame_lenght)],(frame_lenght))
        return frame_number

    def get_id_sequence(self, frame_lenght):
        #begin and end of predicted fram numbers in this seq.
        begin_fr = (self.frame_pointer - frame_lenght)
        end_fr = (self.frame_pointer)
        id_number = self.orig_data[self.dataset_pointer][begin_fr:end_fr, 1].transpose()
        id_number = [int(i) for i in id_number]
        return id_number

    def get_dataset_dimension(self, file_name):
        # return dataset dimension using dataset file name
        return self.dataset_dimensions[file_name]