first try

2020-12-08 20:14:00 +01:00 · 2020-12-08 20:14:00 +01:00 · bb5d311df5
commit bb5d311df5
parent 242c10da9f
6 changed files with 975 additions and 0 deletions
--- a/dev-0/out.tsv
+++ b/dev-0/out.tsv
@ -0,0 +1,462 @@
 753233.5625
 717278.0
 761310.75
 737726.625
 581629.3125
 633398.5625
 734318.5
 742838.8125
 635378.125
 736022.625
 626772.6875
 726735.5
 755602.1875
 660751.5625
 752824.625
 737743.6875
 728132.8125
 1178018.125
 735153.5
 534563.3125
 741015.5
 903060.0
 739771.5
 954096.4375
 753063.125
 688970.6875
 977016.0
 737845.9375
 751359.125
 768007.75
 727502.3125
 730910.4375
 761072.25
 647391.75
 761072.25
 909365.0
 789121.0
 577675.875
 721061.0
 793636.75
 727894.25
 942082.875
 891472.375
 739243.25
 784162.1875
 577062.4375
 624591.5
 644341.5
 725832.375
 624216.5625
 620229.0625
 899992.625
 716937.125
 747951.0
 776238.3125
 729206.375
 636997.0
 704648.0
 687266.625
 526043.0
 707684.125
 582992.5625
 613992.25
 1065286.125
 802480.75
 657002.625
 845082.1875
 780805.1875
 752364.5
 682154.5
 705690.375
 715948.8125
 739430.75
 919589.3125
 725951.625
 985195.5
 764139.5
 970915.5
 730160.625
 943786.875
 895732.5
 538721.1875
 724094.1875
 725798.25
 899992.625
 738646.875
 532859.25
 754767.25
 712165.8125
 534563.3125
 794128.0
 777976.5
 594205.25
 942082.875
 886360.25
 911921.0625
 880225.625
 754750.1875
 737897.0625
 547684.5
 951114.375
 732614.5
 738749.125
 724043.125
 526434.9375
 525617.0
 952818.375
 724366.875
 724486.125
 733466.5
 732273.6875
 774173.5
 895851.8125
 752040.75
 724298.6875
 732273.6875
 735681.75
 738067.5
 720345.25
 570348.5
 709609.75
 939185.9375
 1076822.5
 824463.125
 744542.875
 701444.375
 717448.375
 663409.875
 728712.1875
 758175.3125
 642961.1875
 1082383.625
 945965.1875
 740282.75
 734829.75
 886258.0
 529451.125
 894999.75
 717278.0
 741475.5625
 636465.875
 787144.25
 580402.375
 609218.0
 764991.5625
 536437.75
 658297.6875
 682648.6875
 648260.8125
 971563.0
 897538.8125
 737266.5625
 992863.75
 736022.625
 967575.5
 717278.0
 907916.5
 763117.125
 727195.625
 663409.875
 735187.625
 675338.25
 735937.375
 730910.4375
 737845.9375
 708042.0
 978924.5625
 888916.3125
 763287.5
 613992.25
 724043.125
 552166.1875
 737743.6875
 753063.125
 705789.75
 554228.125
 951404.0
 710666.25
 739345.5
 730041.375
 698104.4375
 548195.75
 514114.625
 713733.5
 700953.125
 901696.75
 738646.875
 613259.5
 762946.6875
 742838.8125
 803523.125
 770069.625
 727502.3125
 595738.875
 728320.25
 771807.75
 732614.5
 745531.25
 759368.125
 776919.9375
 727502.3125
 880140.375
 708076.0625
 753063.125
 904116.5
 724094.1875
 742838.8125
 632736.875
 662285.1875
 723634.125
 754767.25
 778624.0
 677601.75
 534563.3125
 763287.5
 555898.0625
 743520.4375
 783054.5
 739788.5625
 732614.5
 740197.5
 546679.125
 503890.28125
 556443.375
 548076.4375
 734318.5
 589076.0625
 717278.0
 730041.375
 707735.25
 721435.875
 708416.875
 724094.1875
 629328.75
 762810.375
 583742.375
 734318.5
 1117450.125
 723412.5625
 634440.9375
 538244.0625
 526179.3125
 633827.4375
 705349.625
 554500.75
 733125.6875
 733636.875
 639893.875
 737556.25
 665284.3125
 743861.25
 723276.25
 708076.0625
 802855.6875
 729632.375
 733602.875
 739430.75
 642719.75
 697491.0
 663409.875
 619513.375
 528939.875
 594546.0625
 703798.875
 552967.125
 805888.875
 523060.90625
 612816.4375
 1039725.25
 710938.875
 707053.625
 919418.875
 1036317.125
 696249.9375
 747030.8125
 663665.5
 677059.375
 670226.125
 736022.625
 992522.9375
 912773.125
 947467.6875
 522634.90625
 747899.875
 592501.1875
 717278.0
 640626.625
 521459.09375
 761583.4375
 718641.25
 1005473.75
 749229.0
 734318.5
 920373.1875
 908683.375
 700595.25
 728081.6875
 740282.75
 735698.8125
 636145.0
 923849.4375
 792648.375
 961338.6875
 804627.875
 717278.0
 645176.5
 553188.625
 520930.84375
 573092.0
 544872.8125
 733926.625
 522345.21875
 706181.6875
 720305.375
 570348.5
 717278.0
 743690.875
 772370.125
 616088.25
 768399.625
 904440.25
 976675.1875
 580572.8125
 532859.25
 544787.625
 713921.0
 1184289.0
 726411.75
 910217.0
 886019.375
 732614.5
 704548.6875
 956888.1875
 760901.8125
 599147.0
 1111295.625
 753063.125
 917033.25
 735085.375
 892051.75
 773136.9375
 949188.75
 557482.875
 729206.375
 606324.0
 783906.5625
 606324.0
 570246.25
 745585.25
 732819.0
 728422.5
 738578.6875
 533336.375
 742458.125
 611197.5625
 538244.0625
 717295.0
 616497.1875
 712063.5625
 651481.5
 760561.0
 599147.0
 669183.75
 906808.875
 753233.5625
 560124.125
 746076.5
 1146211.75
 570246.25
 734301.5
 739430.75
 708842.875
 724946.25
 845593.375
 580061.5625
 722816.125
 627624.6875
 906808.875
 726343.5625
 660001.75
 541652.1875
 915138.8125
 615642.25
 733994.75
 636806.625
 895831.875
 558249.6875
 703884.125
 740569.5
 704071.5625
 540646.75
 539573.1875
 771807.75
 574438.1875
 556716.0
 636806.625
 675338.25
 574659.75
 895050.875
 749859.5
 743486.375
 733276.1875
 1011949.125
 969518.125
 936459.5
 1255859.375
 714173.6875
 566940.375
 725713.0625
 654889.5625
 807490.75
 613157.25
 580794.3125
 634492.0
 797453.8125
 598499.5
 992673.375
 744031.625
 580760.25
 591700.3125
 895542.1875
 724043.125
 595298.75
 764395.125
 731114.875
 546662.0625
 738698.0
 738698.0
 733279.0625
 738425.3125
 796669.9375
 795920.1875
 644324.4375
 689822.75
 663069.0625
 787076.125
 992352.5
 617380.4375
 835519.5
 736056.6875
 559749.25
 630893.5625
 580095.6875
 638019.4375
 654395.4375
 675048.5625
 868899.125
 763696.5
 714074.375
 730893.375
 713852.8125
 585667.9375
 713852.8125
 733347.25
 574012.1875
 776545.0625
 597408.875
 776545.0625
 557107.9375
 744815.5
 1171488.75
 775045.5
--- a/BIN
+++ b/BIN
--- a/model.pkl
+++ b/model.pkl
--- a/predict.py
+++ b/predict.py
@ -0,0 +1,37 @@
 import pickle
 import sys
 import torch
 import pandas as pd
 def read_data_file(filepath):
    df = pd.read_csv(filepath, sep='\t', header=None, index_col=None)
    dataframe = df.iloc[:, [7,10]]
    dataframe.columns = ['biggy','type']
    #print(dataframe.size[0])
    # for x in range(len(dataframe)):
    #     dataframe['biggy'].loc[x] = dataframe['biggy'].loc[x].replace(" ","")
        #such dumb solution, well, but at least it works
    dataframe['bias'] = 1
    dataframe['biggy'] = dataframe['biggy'].astype(float)
    return dataframe
 def dataframe_to_arrays(dataframe):
    dataframe1 = dataframe.copy(deep=True)
    dataframe1["type"] = dataframe1["type"].astype('category').cat.codes
    return dataframe1
 PREDICT_FILE_PATH = 'test-A/in.tsv'
 def main():
    w = pickle.load(open('model.pkl', 'rb'))
    data = read_data_file(PREDICT_FILE_PATH)
    data = dataframe_to_arrays(data)
    for index, row in data.iterrows():
        #print(row[0], row[1])
        x = torch.tensor([float(row[0]), float(row[1]), 1])
        y = x @ w
        print(y.item())
 main()
--- a/test-A/out.tsv
+++ b/test-A/out.tsv
@ -0,0 +1,418 @@
 758158.25
 761583.4375
 739686.3125
 639161.125
 728115.75
 810830.625
 545639.625
 880205.6875
 1147683.0
 735784.0
 756454.25
 758175.3125
 735000.125
 717278.0
 734318.5
 736022.625
 806911.3125
 834196.125
 703645.5
 939049.625
 778624.0
 705349.625
 646369.3125
 718982.0
 618062.0625
 735698.8125
 760987.0
 1111420.75
 607704.25
 735681.75
 1005644.125
 944298.125
 1056765.75
 620621.0
 636145.0
 776238.3125
 922145.375
 747951.0
 725798.25
 681779.625
 761072.25
 563344.8125
 532007.1875
 754767.25
 715573.9375
 738646.875
 756641.6875
 713409.75
 704838.375
 628286.375
 753114.25
 610243.3125
 627454.3125
 757834.5
 734318.5
 946036.25
 665812.5625
 622410.25
 643881.375
 723412.5625
 620958.9375
 923849.4375
 736022.625
 976675.1875
 718078.875
 653185.5
 701069.5
 560771.6875
 917033.25
 719322.875
 893176.4375
 575460.625
 941742.0
 933835.25
 730399.25
 759027.375
 724435.0
 910217.0
 735698.8125
 746246.9375
 816215.5
 797147.125
 674264.6875
 760731.375
 707053.625
 745105.25
 724230.5
 757323.3125
 735341.0
 717278.0
 746570.6875
 894829.375
 757152.875
 917033.25
 549899.8125
 1070398.25
 828893.625
 942082.875
 534733.6875
 530643.9375
 742838.8125
 737726.625
 741134.75
 753063.125
 724605.375
 731080.8125
 759879.375
 622969.75
 622066.5625
 543083.5625
 1022684.6875
 746332.125
 703901.125
 629328.75
 625730.3125
 703815.9375
 543236.9375
 743350.0
 820563.75
 539505.0625
 822608.625
 726548.0
 730586.625
 736908.6875
 681029.8125
 886019.375
 628647.125
 737845.9375
 752824.625
 754767.25
 936118.625
 566122.4375
 717278.0
 718982.0
 760390.625
 983340.9375
 730910.4375
 737726.625
 558420.0625
 716235.625
 561811.125
 710666.25
 761242.625
 762265.0625
 800776.75
 544787.625
 728183.9375
 1022684.6875
 759879.375
 738220.8125
 705042.875
 725798.25
 712268.0625
 766695.625
 532859.25
 732614.5
 1072085.25
 1051210.625
 736942.75
 825826.375
 766695.625
 537971.375
 734318.5
 657871.6875
 767240.875
 595298.75
 753693.625
 747951.0
 742838.8125
 754750.1875
 518715.5625
 774023.0625
 738527.5625
 718982.0
 533285.25
 727604.5625
 549899.8125
 639893.875
 677042.3125
 626448.875
 717278.0
 946002.1875
 778624.0
 743179.625
 718198.1875
 721026.875
 683057.625
 982449.0625
 742838.8125
 765673.1875
 841012.375
 558300.8125
 551603.875
 1031205.0
 706917.3125
 570518.875
 725798.25
 920441.375
 727672.75
 741134.75
 676340.75
 570348.5
 722185.625
 733125.6875
 597613.375
 744491.75
 696249.9375
 906519.1875
 721776.6875
 1299764.25
 735903.3125
 644665.25
 541209.125
 950091.9375
 720686.125
 997123.875
 961338.6875
 696249.9375
 708076.0625
 897907.9375
 736022.625
 919027.0
 906808.875
 703901.125
 651481.5
 720856.5
 748479.25
 1029500.9375
 727502.3125
 727502.3125
 736022.625
 772540.5
 716425.9375
 725798.25
 1139844.375
 654634.0
 968683.1875
 1203314.625
 656593.625
 741305.125
 573756.5625
 583980.9375
 736022.625
 729325.625
 578868.75
 663921.0625
 613992.25
 717278.0
 726070.875
 794471.75
 849683.125
 705349.625
 648260.8125
 722731.0
 961338.6875
 697831.8125
 669203.6875
 781827.625
 654634.0
 512580.96875
 721231.375
 722407.1875
 784639.3125
 757152.875
 934585.0
 727246.6875
 732273.6875
 766695.625
 690674.75
 768399.625
 766695.625
 727195.625
 802480.75
 1147512.625
 975823.125
 751359.125
 713529.0625
 745735.75
 578868.75
 642961.1875
 555011.9375
 688970.6875
 764991.5625
 738016.375
 696249.9375
 549899.8125
 549899.8125
 580402.375
 551603.875
 583980.9375
 715642.125
 708451.0
 744525.875
 776511.0
 816624.4375
 643642.8125
 604429.625
 534563.3125
 734318.5
 538823.4375
 760561.0
 712847.4375
 629158.375
 724366.875
 722901.375
 644665.25
 624878.25
 710291.375
 659831.375
 736022.625
 543697.0
 561828.1875
 580572.8125
 535347.125
 549559.0
 745122.25
 623174.1875
 798050.25
 646164.8125
 734318.5
 522242.96875
 639945.0
 553307.875
 556716.0
 540033.3125
 712336.25
 651481.5
 583980.9375
 756982.5
 624196.625
 672782.1875
 715573.9375
 1073826.25
 585565.6875
 595057.25
 620808.4375
 724435.0
 741305.125
 674486.25
 522242.96875
 754920.5625
 910029.5625
 750847.875
 559544.75
 825826.375
 786803.5
 732273.6875
 591700.3125
 562492.75
 922176.5625
 724094.1875
 757834.5
 1077595.25
 788848.375
 686605.0
 733602.875
 1009802.0625
 720686.125
 727502.3125
 1090847.0
 732392.9375
 530030.5
 740776.875
 751427.25
 675338.25
 735153.5
 769081.25
 1038021.25
 551603.875
 560635.375
 734318.5
 714426.4375
 530473.5625
 879714.375
 546474.625
 695786.9375
 737726.625
 540646.75
 649862.625
 538295.1875
 845474.125
 714892.3125
 1046030.25
 568184.3125
 644648.1875
 985826.0
 734744.5625
 725661.9375
 1041278.875
 1041278.875
 1099898.375
 771194.3125
 1002767.1875
 554807.5
 745991.3125
 731609.125
 736022.625
 892903.75
 733449.5
 663409.875
 736806.5
 782424.0625
 743588.625
 743588.625
 1152573.625
 721538.125
 709453.4375
 1077725.75
 769047.1875
 751785.125
 736976.875
 520981.96875
 713835.75
 712114.6875
 736124.875
 783242.0
 593574.75
 764718.875
--- a/train.py
+++ b/train.py
@ -0,0 +1,58 @@
 import torch
 import torch.nn as nn
 import numpy as np
 import pandas as pd
 import sys
 import torch.nn.functional as F
 import pickle
 from torch.utils.data import TensorDataset, DataLoader
 TRAIN_FILE_PATH = 'train/train.tsv'
 #prepare data methods
 def read_data_file(filepath):
    df = pd.read_csv(filepath, sep='\t', header=None, index_col=None)
    dataframe = df.iloc[:, [0,8,11]]
    dataframe.columns = ['price','biggy','type']
    #print(dataframe.size[0])
    for x in range(len(dataframe)):
        dataframe['biggy'].loc[x] = dataframe['biggy'].loc[x].replace(" ","")
        #such dumb solution, well, but at least it works
    dataframe['bias'] = 1
    dataframe['biggy'] = dataframe['biggy'].astype(float)
    return dataframe
 def dataframe_to_arrays(dataframe):
    dataframe1 = dataframe.copy(deep=True)
    dataframe1["type"] = dataframe1["type"].astype('category').cat.codes
    inputs_array = dataframe1[input_cols].to_numpy()
    targets_array = dataframe1[output_cols].to_numpy()
    return inputs_array, targets_array
 data =  read_data_file(TRAIN_FILE_PATH)
 input_cols = data.columns.values[1:]
 output_cols = data.columns.values[:1]
 inputs_array_training, targets_array_training = dataframe_to_arrays(data)
 inputs_training = torch.from_numpy(inputs_array_training).type(torch.float32)
 targets_training = torch.from_numpy(targets_array_training).type(torch.float32)
 print(inputs_training)
 w = torch.tensor([7201.61492633873, 1,7201.500], requires_grad=True)
 learning_rate = torch.tensor(0.000000000005)
 print("training started")
 for i in range(10000):
    y_predicted = inputs_training @ w
    cost = torch.sum((y_predicted - targets_training) ** 2)
    cost.backward()
    with torch.no_grad():
        w -= learning_rate * w.grad
    w.requires_grad = True
 print(w)
 pickle.dump(w, open('model.pkl', 'wb'))