mirror of
https://github.com/wassname/cryptokitties_genetics.git
synced 2026-08-13 12:10:27 +08:00
251 KiB
251 KiB
In [28]:
%matplotlib inline
import numpy as np
from matplotlib import pyplot as plt
import pandas as pd
import json
from tqdm import tqdm
import os
import datetime
import arrow
import timeIn [58]:
sales_data_file = '.cache/sales.json'
sales = json.load(open(sales_data_file))
len(sales)
sales['sales'][0]Out [58]:
{'blockNumber': 4688676,
'blocktimeStamp': 1512617298,
'id': 'log_9357a0df',
'rank': 1,
'returnValues': {'0': '18',
'1': '253336776620370370370',
'2': '0xA6d3fdf423BbC578dd4d41220078475371626B22'},
'soldPrice': 115197.04572803818}In [129]:
# convert to dataframe
df = pd.DataFrame(sales['sales'])
# convert to pandas timestamp
datetimes = df['blocktimeStamp'].apply(datetime.datetime.fromtimestamp)
df['date'] = pd.to_datetime(datetimes)
# grab some of the fields under return values for the dataframe
df2=pd.DataFrame.from_records(df['returnValues'].values)
df2.columns=['kitty_id','price_18eth','address']
df2['price_eth']=df2['price_18eth'].apply(lambda x:float(x)*1e-18)
df2['kitty_id'] = pd.to_numeric(df2['kitty_id'])
for col in ['kitty_id','price_eth']:
df[col] = df2[col]
# rename cols
df['soldPrice'] = df['soldPrice'].rename('soldPrice_USD')
df = df.rename(columns={"soldPrice":"sold_price_usd"})
df.index = df['kitty_id']
# drop uneeded columns
df = df.drop(['id', 'blockNumber', 'rank', 'returnValues', 'blocktimeStamp'], axis=1)
df_sales = df
df_salesOut [129]:
| sold_price_usd | date | kitty_id | price_eth | |
|---|---|---|---|---|
| kitty_id | ||||
| 18 | 1.151970e+05 | 2017-12-07 11:28:18 | 18 | 2.533368e+02 |
| 4 | 1.123156e+05 | 2017-12-07 03:41:57 | 4 | 2.470000e+02 |
| 1 | 1.144816e+05 | 2017-12-03 04:32:36 | 1 | 2.469255e+02 |
| 21 | 1.080061e+05 | 2017-12-08 17:31:03 | 21 | 2.375228e+02 |
| 22 | 1.023118e+05 | 2017-12-08 17:34:36 | 22 | 2.250000e+02 |
| 5 | 1.016005e+05 | 2017-12-06 00:45:01 | 5 | 2.220000e+02 |
| 7 | 8.767734e+04 | 2017-12-05 03:45:47 | 7 | 1.900468e+02 |
| 35 | 8.500053e+04 | 2017-12-06 15:18:02 | 35 | 1.888897e+02 |
| 87 | 8.142809e+04 | 2017-12-07 02:11:42 | 87 | 1.790734e+02 |
| 101 | 8.199325e+04 | 2017-12-04 11:28:49 | 101 | 1.757532e+02 |
| 30 | 7.792517e+04 | 2017-12-06 00:28:28 | 30 | 1.686849e+02 |
| 78 | 7.349530e+04 | 2017-12-05 14:49:17 | 78 | 1.568700e+02 |
| 14 | 7.048144e+04 | 2017-12-07 05:59:58 | 14 | 1.550000e+02 |
| 18 | 6.930015e+04 | 2017-12-06 10:37:44 | 18 | 1.540000e+02 |
| 19 | 6.930015e+04 | 2017-12-06 10:19:43 | 19 | 1.540000e+02 |
| 102 | 6.982621e+04 | 2017-12-08 16:18:33 | 102 | 1.535590e+02 |
| 2 | 7.016610e+04 | 2017-12-04 09:16:36 | 2 | 1.500000e+02 |
| 37 | 6.772737e+04 | 2017-12-05 16:14:40 | 37 | 1.432795e+02 |
| 23 | 6.293852e+04 | 2017-12-05 15:07:48 | 23 | 1.338756e+02 |
| 38 | 6.033274e+04 | 2017-12-05 20:59:48 | 38 | 1.300000e+02 |
| 102 | 5.991045e+04 | 2017-12-04 11:43:16 | 102 | 1.284185e+02 |
| 93 | 5.671367e+04 | 2017-12-06 15:26:53 | 93 | 1.260301e+02 |
| 52 | 5.657864e+04 | 2017-12-06 14:01:35 | 52 | 1.257300e+02 |
| 62 | 5.203606e+04 | 2017-12-06 15:23:07 | 62 | 1.156354e+02 |
| 40 | 5.229268e+04 | 2017-12-07 13:51:53 | 40 | 1.150000e+02 |
| 43 | 4.501718e+04 | 2017-12-07 23:51:39 | 43 | 9.900000e+01 |
| 55 | 4.461293e+04 | 2017-12-07 12:20:39 | 55 | 9.811100e+01 |
| 27 | 4.335030e+04 | 2017-12-07 21:27:04 | 27 | 9.533427e+01 |
| 31 | 4.450235e+04 | 2017-12-05 05:47:30 | 31 | 9.526291e+01 |
| 40 | 4.251421e+04 | 2017-12-08 18:02:20 | 40 | 9.349556e+01 |
| ... | ... | ... | ... | ... |
| 3464 | 4.130450e-01 | 2017-11-24 13:31:30 | 3464 | 1.000000e-03 |
| 4391 | 4.661040e-01 | 2017-11-25 12:22:40 | 4391 | 1.000000e-03 |
| 6067 | 4.601970e-01 | 2017-11-26 11:11:44 | 6067 | 1.000000e-03 |
| 3629 | 4.733770e-01 | 2017-11-27 11:13:18 | 3629 | 1.000000e-03 |
| 6112 | 4.932540e-01 | 2017-11-27 14:11:17 | 6112 | 1.000000e-03 |
| 8311 | 4.803550e-01 | 2017-11-28 07:53:38 | 8311 | 1.000000e-03 |
| 8882 | 4.815180e-01 | 2017-11-28 14:11:43 | 8882 | 1.000000e-03 |
| 9110 | 4.815180e-01 | 2017-11-28 14:12:43 | 9110 | 1.000000e-03 |
| 6853 | 4.886860e-01 | 2017-11-29 10:00:49 | 6853 | 1.000000e-03 |
| 9199 | 4.886860e-01 | 2017-11-29 10:06:22 | 9199 | 1.000000e-03 |
| 7946 | 4.886860e-01 | 2017-11-29 10:19:55 | 7946 | 1.000000e-03 |
| 4137 | 4.842060e-01 | 2017-11-29 11:12:36 | 4137 | 1.000000e-03 |
| 8296 | 4.842060e-01 | 2017-11-29 11:15:34 | 8296 | 1.000000e-03 |
| 11172 | 4.611260e-01 | 2017-11-30 09:38:46 | 11172 | 1.000000e-03 |
| 13314 | 4.342620e-01 | 2017-11-30 20:23:40 | 13314 | 1.000000e-03 |
| 13771 | 4.715570e-01 | 2017-12-02 05:52:01 | 13771 | 1.000000e-03 |
| 54364 | 4.547190e-01 | 2017-12-07 02:45:38 | 54364 | 1.000000e-03 |
| 7545 | 4.471072e-01 | 2017-12-01 07:20:59 | 7545 | 9.999847e-04 |
| 113539 | 1.684144e-02 | 2017-12-07 22:52:26 | 113539 | 3.703704e-05 |
| 113534 | 1.294686e-02 | 2017-12-07 22:23:15 | 113534 | 2.847222e-05 |
| 109415 | 8.947017e-03 | 2017-12-07 04:45:23 | 109415 | 1.967593e-05 |
| 86900 | 7.916684e-03 | 2017-12-06 11:46:53 | 86900 | 1.759259e-05 |
| 43283 | 5.618865e-03 | 2017-12-06 05:03:29 | 43283 | 1.203704e-05 |
| 41574 | 5.365162e-03 | 2017-12-05 23:11:30 | 41574 | 1.157407e-05 |
| 30173 | 4.558781e-03 | 2017-12-04 07:13:30 | 30173 | 9.785880e-06 |
| 91591 | 3.526177e-03 | 2017-12-07 14:02:35 | 91591 | 7.754630e-06 |
| 136018 | 1.539413e-03 | 2017-12-08 05:03:21 | 136018 | 3.385417e-06 |
| 127036 | 3.947214e-04 | 2017-12-08 00:26:49 | 127036 | 8.680556e-07 |
| 111960 | 3.157771e-04 | 2017-12-08 11:45:07 | 111960 | 6.944444e-07 |
| 45652 | 1.023946e-13 | 2017-12-06 01:29:05 | 45652 | 2.220000e-16 |
78124 rows × 4 columns
In [109]:
genetics_file = '.cache/genes.json'
genetics = json.load(open(genetics_file))
len(genetics)Out [109]:
43488
In [496]:
# convert to dataframe
df = pd.DataFrame.from_dict(genetics).T
df.columns=['is_gestating', 'is_ready', 'cooldown_index', 'next_action_at', 'siring_with_id', 'birth_time', 'matron_id', 'sire_id', 'generation', 'genes']
df.index = pd.to_numeric(df.index)
df['generation'] = pd.to_numeric(df['generation'])
df['matron_id'] = pd.to_numeric(df['matron_id'])
df['sire_id'] = pd.to_numeric(df['sire_id'])
df['birth_time'] = pd.to_numeric(df['birth_time'])
df = df.sort_index()
df = df.drop(['is_gestating', 'is_ready', 'cooldown_index', 'next_action_at', 'siring_with_id'], axis=1)
df = df[df.genes!='0'] # remove rows with no genes
df = df[1:] # remove origin kitty
df_genetics = df
df_geneticsOut [496]:
| birth_time | matron_id | sire_id | generation | genes | |
|---|---|---|---|---|---|
| 1 | 1511417999 | 0 | 0 | 0 | 6268376211548016160889809226598771686091543863... |
| 2 | 1511417999 | 0 | 0 | 0 | 6233328247424174420738016520205540105237269755... |
| 3 | 1511417999 | 0 | 0 | 0 | 5163523354162354170567022901547386224918079227... |
| 4 | 1511417999 | 0 | 0 | 0 | 6268375141947334719316716288420757560178523965... |
| 5 | 1511417999 | 0 | 0 | 0 | 6233328806923846998926376260807366625937483650... |
| 6 | 1511417999 | 0 | 0 | 0 | 4613035485150908523120757036068930199538348135... |
| 7 | 1511417999 | 0 | 0 | 0 | 6233277698034429017103950567765524970954426879... |
| 8 | 1511418008 | 0 | 0 | 0 | 4559620020693848583707206074171681675830775819... |
| 9 | 1511418035 | 0 | 0 | 0 | 6233833779874278041852346338088492342734244547... |
| 10 | 1511418044 | 0 | 0 | 0 | 5163523337171367126011753839819537067276020035... |
| 11 | 1511418044 | 0 | 0 | 0 | 4611367880921070440157963212743285505842704654... |
| 12 | 1511418044 | 0 | 0 | 0 | 4578493054511227949035857584594486760104829763... |
| 13 | 1511418112 | 0 | 0 | 0 | 4560159221201941276348499729156645386396250750... |
| 14 | 1511418116 | 0 | 0 | 0 | 5129554889428984613286854531398112217255378602... |
| 15 | 1511418116 | 0 | 0 | 0 | 4613035452223835287331024509742580987888675330... |
| 16 | 1511418116 | 0 | 0 | 0 | 6233833252284378338724434974413836525523562295... |
| 17 | 1511418116 | 0 | 0 | 0 | 6269404553147363984118345854639337024260173607... |
| 18 | 1511418116 | 0 | 0 | 0 | 4560192904665667822832111046880560033060704840... |
| 19 | 1511418116 | 0 | 0 | 0 | 5129049373951532651003023567579923289793462567... |
| 20 | 1511418149 | 0 | 0 | 0 | 6269421452513848782205821543180235199856079283... |
| 21 | 1511418149 | 0 | 0 | 0 | 6216023906598010121394201352319905816623933883... |
| 22 | 1511418149 | 0 | 0 | 0 | 5129538055862746223357733958711974998244902552... |
| 23 | 1511418239 | 0 | 0 | 0 | 4613002328763362082249999317894476883917193731... |
| 24 | 1511418239 | 0 | 0 | 0 | 6233866409217284196160692456666556233901127417... |
| 25 | 1511418239 | 0 | 0 | 0 | 6216073403359021113389331319842603070273775510... |
| 26 | 1511418239 | 0 | 0 | 0 | 4611367355436383673714579266113856136150479633... |
| 27 | 1511418239 | 0 | 0 | 0 | 5112332628896149655482179525762049643336261414... |
| 28 | 1511418239 | 0 | 0 | 0 | 6234945333635884915515939895611296479031266622... |
| 29 | 1511418239 | 0 | 0 | 0 | 5111793972432994898422441895738492541345210887... |
| 30 | 1511418239 | 0 | 0 | 0 | 5164095758889984122005573356320380613355077493... |
| ... | ... | ... | ... | ... | ... |
| 45615 | 1512333845 | 16747 | 8707 | 15 | 6269453578415381943516255760284043326213056811... |
| 45616 | 1512333853 | 16752 | 14356 | 16 | 6274256848369521403290224743805255098953710821... |
| 45617 | 1512333853 | 16759 | 10781 | 18 | 6216579413460865674109906598893983985233123945... |
| 45618 | 1512333853 | 16779 | 3129 | 15 | 5113411010303687998100933201783764826643580312... |
| 45619 | 1512333870 | 16785 | 26791 | 7 | 5163523370084392770969021128124236334163811383... |
| 45620 | 1512333903 | 16812 | 30917 | 8 | 1125665612349183931157596564365780411787821413... |
| 45621 | 1512333941 | 16838 | 21772 | 20 | 5129015690991404686765294140461041275344784455... |
| 45622 | 1512333943 | 16881 | 22417 | 9 | 6233884335447604873748743543094761863725880936... |
| 45623 | 1512334078 | 16889 | 14801 | 17 | 4559621634831099400769715992700498680286764411... |
| 45624 | 1512334078 | 16906 | 22726 | 17 | 5164063062683703662909948084918540684004499716... |
| 45625 | 1512334078 | 17005 | 40401 | 18 | 4576909156825801026091186298516853686702166042... |
| 45626 | 1512334078 | 17007 | 18698 | 5 | 5112283131601826856222393066936544803756918176... |
| 45627 | 1512334078 | 17014 | 4026 | 16 | 5128997229134311199621080030231934548551440779... |
| 45628 | 1512334078 | 17048 | 11538 | 2 | 5130683853628148456670481052173151647647645194... |
| 45629 | 1512334078 | 17187 | 17460 | 17 | 5164095741922126726071764639350042066763227222... |
| 45630 | 1512334078 | 17189 | 30399 | 11 | 6267785376692115586123185654357364764310432318... |
| 45631 | 1512334078 | 17207 | 21691 | 11 | 5112283148576568425285464734611270849724426078... |
| 45632 | 1512334078 | 17228 | 30507 | 6 | 6785467971343760798530560564557499310539827456... |
| 45633 | 1512334078 | 17231 | 28310 | 15 | 6429545616359495633152585752099073668690663239... |
| 45634 | 1512334078 | 17241 | 18605 | 19 | 4611418102266473210778517038652661250506831269... |
| 45635 | 1512334078 | 17249 | 16165 | 17 | 5112282587556835590741740255473440867729208112... |
| 45636 | 1512334078 | 17250 | 14644 | 17 | 5320430242218365032649964898064686653005504838... |
| 45637 | 1512334078 | 17332 | 11538 | 2 | 5129587503499545887800398241586485522317051291... |
| 45638 | 1512334078 | 17358 | 26820 | 9 | 6234018569512224369657786035770197463952746524... |
| 45639 | 1512334078 | 17397 | 21492 | 14 | 4576874896703348387734864012261754610135465392... |
| 45640 | 1512334078 | 17442 | 22869 | 5 | 4611906011319878245491527305060291690203662556... |
| 45641 | 1512334078 | 17456 | 42818 | 14 | 6268376743958601894923265566941986372723849946... |
| 45642 | 1512334088 | 17495 | 28301 | 19 | 4768274754087843061098657370537325266142097512... |
| 45643 | 1512334088 | 17523 | 17920 | 12 | 4559604750943350203969277452538726732270344691... |
| 45644 | 1512334088 | 17553 | 34876 | 19 | 5165140456826221028155736310509814307546921341... |
43468 rows × 5 columns
In [518]:
def genestr_to_bits(x):
"""Gene data is a uint256 string, but I think the genes are it's bytes so lets convert to a bit array"""
bits = bin(int(x))[2:]
bitarray = [1 if b=='1' else 0 for b in bits]
bitarray = (256-len(bitarray))*[0] + bitarray # pad
return bitarrayIn [692]:
# merge and add parent genes
df = pd.merge(df_genetics, df_sales, how='inner', left_index=True, right_index=True)
df
# remove rows where we don't have the parent genetics
mask1 = df['sire_id'].apply(lambda x:int(x) in df_genetics.index)
mask2 = df['matron_id'].apply(lambda x:int(x) in df_genetics.index)
df = df[mask1*mask2]
# remove generation 0
df = df[df['generation']>0]
len(df)
df['sire_genes']=df['sire_id'].apply(lambda x:df_genetics.loc[x].genes).apply(genestr_to_bits)
df['sire_gen']=df['sire_id'].apply(lambda x:df_genetics.loc[x].generation)
df['matron_genes']=df['matron_id'].apply(lambda x:df_genetics.loc[x].genes).apply(genestr_to_bits)
df['matron_gen']=df['matron_id'].apply(lambda x:df_genetics.loc[x].generation)
df['genes']=df['genes'].apply(genestr_to_bits)
dfOut [692]:
/home/wassname/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/pandas/core/computation/expressions.py:183: UserWarning: evaluating in Python space because the '*' operator is not supported by numexpr for the bool dtype, use '&' instead unsupported[op_str]))
| birth_time | matron_id | sire_id | generation | genes | sold_price_usd | date | kitty_id | price_eth | sire_genes | sire_gen | matron_genes | matron_gen | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 3005 | 1511466911 | 1045 | 1003 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 75.475318 | 2017-12-02 05:08:56 | 3005 | 0.160056 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3007 | 1511466918 | 1044 | 1006 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 101.317700 | 2017-12-03 00:22:07 | 3007 | 0.220000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3008 | 1511466918 | 1097 | 1099 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5.555534 | 2017-11-24 08:25:32 | 3008 | 0.013552 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3010 | 1511467040 | 1041 | 1058 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 23.645100 | 2017-11-29 07:17:53 | 3010 | 0.050000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3010 | 1511467040 | 1041 | 1058 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 9.974763 | 2017-11-30 06:41:14 | 3010 | 0.022729 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3010 | 1511467040 | 1041 | 1058 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 4.200592 | 2017-11-24 04:15:04 | 3010 | 0.009996 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3011 | 1511467040 | 1093 | 1099 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 6.281243 | 2017-11-24 06:00:39 | 3011 | 0.014819 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3012 | 1511467189 | 1046 | 1087 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 478.842000 | 2017-12-04 00:15:11 | 3012 | 1.000000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3012 | 1511467189 | 1046 | 1087 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 399.712992 | 2017-12-05 09:53:38 | 3012 | 0.856569 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3012 | 1511467189 | 1046 | 1087 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 4.096416 | 2017-11-24 10:23:41 | 3012 | 0.009984 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3013 | 1511467349 | 1088 | 1019 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 231.375440 | 2017-12-05 00:18:59 | 3013 | 0.499733 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3017 | 1511467461 | 1078 | 1056 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 195.708110 | 2017-12-08 02:05:55 | 3017 | 0.430394 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3018 | 1511467461 | 1043 | 3003 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 7.777746 | 2017-11-24 09:43:28 | 3018 | 0.019250 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3018 | 1511467461 | 1043 | 3003 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 3.529732 | 2017-11-25 03:49:18 | 3018 | 0.007695 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3020 | 1511467461 | 1087 | 3006 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 124.010446 | 2017-12-05 03:36:28 | 3020 | 0.268801 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3020 | 1511467461 | 1087 | 3006 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 82.108207 | 2017-12-05 02:59:27 | 3020 | 0.177975 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3020 | 1511467461 | 1087 | 3006 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 8.249211 | 2017-11-24 11:04:56 | 3020 | 0.019992 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3021 | 1511467461 | 1093 | 3008 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5.974832 | 2017-11-25 13:08:18 | 3021 | 0.012687 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3022 | 1511467549 | 1062 | 1005 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 46.714600 | 2017-12-02 13:53:03 | 3022 | 0.100000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3024 | 1511467642 | 1077 | 1002 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 598.399623 | 2017-12-06 01:31:35 | 3024 | 1.297431 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3029 | 1511467718 | 1053 | 1010 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 551.445594 | 2017-12-04 05:27:15 | 3029 | 1.200295 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3029 | 1511467718 | 1053 | 1010 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 25.891600 | 2017-12-02 00:18:55 | 3029 | 0.056251 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3029 | 1511467718 | 1053 | 1010 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 18.916080 | 2017-11-29 07:21:59 | 3029 | 0.040000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3030 | 1511467718 | 1036 | 1047 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 4.121639 | 2017-11-24 11:28:35 | 3030 | 0.009989 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3031 | 1511467718 | 1098 | 1017 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 937.400000 | 2017-12-05 10:11:05 | 3031 | 2.000000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3031 | 1511467718 | 1098 | 1017 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 2.289915 | 2017-11-26 13:20:38 | 3031 | 0.005000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3032 | 1511467718 | 1054 | 1004 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 981.295931 | 2017-12-06 06:53:53 | 3032 | 2.180653 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3032 | 1511467718 | 1054 | 1004 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 314.984033 | 2017-12-06 06:13:25 | 3032 | 0.699963 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3032 | 1511467718 | 1054 | 1004 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 7.705208 | 2017-11-28 21:47:07 | 3032 | 0.016276 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| 3033 | 1511467823 | 1039 | 1051 | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 47.888720 | 2017-11-29 16:19:38 | 3033 | 0.099592 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 0 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 45592 | 1512333116 | 16162 | 28528 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 15.392169 | 2017-12-04 07:42:22 | 45592 | 0.033041 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 15 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 |
| 45594 | 1512333116 | 16176 | 38530 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 12.723501 | 2017-12-08 00:43:11 | 45594 | 0.027981 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 |
| 45597 | 1512333116 | 16262 | 5646 | 4 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 71.372034 | 2017-12-06 06:00:54 | 45597 | 0.158604 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 3 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 3 |
| 45597 | 1512333116 | 16262 | 5646 | 4 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 23.875589 | 2017-12-07 13:46:57 | 45597 | 0.052506 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 3 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 3 |
| 45598 | 1512333116 | 16274 | 27366 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 32.143957 | 2017-12-04 05:13:35 | 45598 | 0.069966 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 |
| 45598 | 1512333116 | 16274 | 27366 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 17.731081 | 2017-12-06 14:13:00 | 45598 | 0.039402 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 |
| 45599 | 1512333116 | 16331 | 6439 | 14 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 26.936386 | 2017-12-04 13:25:22 | 45599 | 0.057391 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11 |
| 45599 | 1512333116 | 16331 | 6439 | 14 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 22.709755 | 2017-12-05 21:59:24 | 45599 | 0.048933 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11 |
| 45599 | 1512333116 | 16331 | 6439 | 14 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13.582793 | 2017-12-07 06:29:50 | 45599 | 0.029871 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11 |
| 45600 | 1512333116 | 16364 | 13561 | 13 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 26.837560 | 2017-12-06 08:07:35 | 45600 | 0.059639 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 12 |
| 45601 | 1512333116 | 16365 | 40842 | 6 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 66.577622 | 2017-12-04 05:07:14 | 45601 | 0.144915 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5 |
| 45601 | 1512333116 | 16365 | 40842 | 6 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 30.555222 | 2017-12-07 02:10:26 | 45601 | 0.067196 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5 |
| 45602 | 1512333116 | 16386 | 19319 | 20 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 111.586185 | 2017-12-06 11:31:01 | 45602 | 0.247969 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 19 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 12 |
| 45604 | 1512333394 | 16392 | 32070 | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 17.635940 | 2017-12-08 12:44:17 | 45604 | 0.038784 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 15 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 8 |
| 45607 | 1512333492 | 25943 | 13947 | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 31.803504 | 2017-12-05 00:49:27 | 45607 | 0.068690 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 14 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 15 |
| 45609 | 1512333564 | 16557 | 35617 | 12 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 42.137717 | 2017-12-05 10:21:51 | 45609 | 0.089903 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11 |
| 45610 | 1512333654 | 16561 | 30199 | 21 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 26.655555 | 2017-12-05 22:01:17 | 45610 | 0.057435 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 20 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 15 |
| 45621 | 1512333941 | 16838 | 21772 | 20 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 89.230785 | 2017-12-04 04:58:55 | 45621 | 0.188659 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 19 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 14 |
| 45622 | 1512333943 | 16881 | 22417 | 9 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 11.356870 | 2017-12-07 19:04:46 | 45622 | 0.024976 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 8 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 5 |
| 45624 | 1512334078 | 16906 | 22726 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 38.217005 | 2017-12-04 21:52:03 | 45624 | 0.082567 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 12 |
| 45624 | 1512334078 | 16906 | 22726 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 23.142950 | 2017-12-04 21:34:21 | 45624 | 0.050000 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 12 |
| 45628 | 1512334078 | 17048 | 11538 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 163.355168 | 2017-12-05 16:15:09 | 45628 | 0.345583 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 |
| 45629 | 1512334078 | 17187 | 17460 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 32.375588 | 2017-12-04 16:55:59 | 45629 | 0.068417 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 6 |
| 45629 | 1512334078 | 17187 | 17460 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 18.923884 | 2017-12-04 16:07:52 | 45629 | 0.039991 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 6 |
| 45629 | 1512334078 | 17187 | 17460 | 17 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13.773179 | 2017-12-04 05:01:55 | 45629 | 0.029979 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 16 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 6 |
| 45630 | 1512334078 | 17189 | 30399 | 11 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 30.228184 | 2017-12-04 23:06:07 | 45630 | 0.064918 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 10 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 9 |
| 45630 | 1512334078 | 17189 | 30399 | 11 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 20.756527 | 2017-12-04 17:36:33 | 45630 | 0.044004 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 10 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 9 |
| 45630 | 1512334078 | 17189 | 30399 | 11 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 4.770418 | 2017-12-04 17:17:38 | 45630 | 0.010113 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 10 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 9 |
| 45631 | 1512334078 | 17207 | 21691 | 11 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 13.970090 | 2017-12-04 07:33:42 | 45631 | 0.029988 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 9 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 10 |
| 45637 | 1512334078 | 17332 | 11538 | 2 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 213.853079 | 2017-12-06 06:47:25 | 45637 | 0.475228 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 | [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ... | 1 |
27727 rows × 13 columns
In [674]:
# parent genes
X = np.array([df['sire_genes'], df['matron_genes']])
X = np.transpose(X, (1,2,0))
# child genes
Y = np.stack(df['genes'].values)
X.shape, Y.shapeOut [674]:
((27727, 256, 2), (27727, 256))
In [675]:
# split into test and train, val (& shuffle)
import sklearn.model_selection
X_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(X,Y, random_state=42, test_size=0.1)
X_train.shape, y_train.shapeOut [675]:
((24954, 256, 2), (24954, 256))
In [676]:
# # NOTE: there is ~50% overlap between test and train y values :( because of repeated breeding
# # For now I'll just leave it and try to get an accuracy higher than the overlap
# # check for overlap
# overlaps = []
# for y in tqdm(y_test[:1000]):
# overlaps.append(((y - y_train)==0).all(-1).sum()>0)
# overlaps = np.array(overlaps)
# print('overlap fraction', overlaps.sum()/len(overlaps))In [681]:
from sklearn.dummy import DummyClassifier
for strategy in ['stratified', 'prior', 'uniform', 'most_frequent']:
clf = DummyClassifier(strategy=strategy, random_state=0)
clf.fit(X_train.reshape((-1,512)), y_train)
acc = clf.score(X_test.reshape((-1,512)), y_test)
y_pred = clf.predict(X_test.reshape((-1,512)))
loss = sklearn.metrics.log_loss(y_test, y_pred)
print(strategy,'loss',loss,'accuracy',acc)stratified loss 1666.43792773 accuracy 0.0 prior loss 1166.88507952 accuracy 0.0 uniform loss 2205.66894511 accuracy 0.0 most_frequent loss 1166.88507952 accuracy 0.0
In [592]:
import kerasIn [595]:
# Simple model with two layers
model = keras.models.Sequential()
model.add(keras.layers.InputLayer((256,2)))
model.add(keras.layers.Flatten())
model.add(keras.layers.Dense(256, activation='elu'))
model.add(keras.layers.Dense(256, activation='sigmoid'))
model.compile(loss='categorical_crossentropy',
optimizer=keras.optimizers.Adam(lr=1e-3),
metrics=['accuracy'])
model.summary()_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_42 (InputLayer) (None, 256, 2) 0 _________________________________________________________________ flatten_38 (Flatten) (None, 512) 0 _________________________________________________________________ dense_40 (Dense) (None, 256) 131328 _________________________________________________________________ dense_41 (Dense) (None, 256) 65792 ================================================================= Total params: 197,120 Trainable params: 197,120 Non-trainable params: 0 _________________________________________________________________
In [596]:
history = model.fit(X_train, y_train, validation_split=0.2, epochs=500)Train on 19963 samples, validate on 4991 samples Epoch 1/500 19963/19963 [==============================] - 4s - loss: 509.6478 - acc: 0.2359 - val_loss: 505.6140 - val_acc: 0.0950 Epoch 2/500 19963/19963 [==============================] - 4s - loss: 503.6146 - acc: 0.0947 - val_loss: 503.8711 - val_acc: 0.1052 Epoch 3/500 19963/19963 [==============================] - 4s - loss: 502.3385 - acc: 0.1134 - val_loss: 503.3724 - val_acc: 0.2637 Epoch 4/500 19963/19963 [==============================] - 3s - loss: 501.8278 - acc: 0.2307 - val_loss: 503.1031 - val_acc: 0.3264 Epoch 5/500 19963/19963 [==============================] - 3s - loss: 501.5010 - acc: 0.3075 - val_loss: 503.0677 - val_acc: 0.3364 Epoch 6/500 19963/19963 [==============================] - 3s - loss: 501.2706 - acc: 0.3177 - val_loss: 502.9476 - val_acc: 0.3881 Epoch 7/500 19963/19963 [==============================] - 3s - loss: 501.1096 - acc: 0.2912 - val_loss: 502.9000 - val_acc: 0.2176 Epoch 8/500 19963/19963 [==============================] - 4s - loss: 500.9237 - acc: 0.2650 - val_loss: 502.9253 - val_acc: 0.1867 Epoch 9/500 19963/19963 [==============================] - 4s - loss: 500.7793 - acc: 0.2410 - val_loss: 502.9301 - val_acc: 0.2645 Epoch 10/500 19963/19963 [==============================] - 4s - loss: 500.6340 - acc: 0.2070 - val_loss: 502.9690 - val_acc: 0.1268 Epoch 11/500 19963/19963 [==============================] - 4s - loss: 500.5172 - acc: 0.2013 - val_loss: 502.9286 - val_acc: 0.2114 Epoch 12/500 19963/19963 [==============================] - 4s - loss: 500.3877 - acc: 0.1923 - val_loss: 502.9189 - val_acc: 0.2659 Epoch 13/500 19963/19963 [==============================] - 4s - loss: 500.2545 - acc: 0.2060 - val_loss: 503.0023 - val_acc: 0.2589 Epoch 14/500 19963/19963 [==============================] - 4s - loss: 500.1431 - acc: 0.1919 - val_loss: 502.9604 - val_acc: 0.2731 Epoch 15/500 19963/19963 [==============================] - 4s - loss: 500.0178 - acc: 0.1983 - val_loss: 502.9331 - val_acc: 0.2152 Epoch 16/500 19963/19963 [==============================] - 4s - loss: 499.9065 - acc: 0.1946 - val_loss: 503.0404 - val_acc: 0.1399 Epoch 17/500 19963/19963 [==============================] - 4s - loss: 499.7670 - acc: 0.2007 - val_loss: 503.0182 - val_acc: 0.1837 Epoch 18/500 19963/19963 [==============================] - 4s - loss: 499.6611 - acc: 0.1996 - val_loss: 503.0419 - val_acc: 0.2180 Epoch 19/500 19963/19963 [==============================] - 4s - loss: 499.5488 - acc: 0.1936 - val_loss: 503.0547 - val_acc: 0.1356 Epoch 20/500 19963/19963 [==============================] - 4s - loss: 499.4477 - acc: 0.1777 - val_loss: 503.0849 - val_acc: 0.1016 Epoch 21/500 19963/19963 [==============================] - 4s - loss: 499.3326 - acc: 0.1825 - val_loss: 503.1680 - val_acc: 0.1453 Epoch 22/500 19963/19963 [==============================] - 4s - loss: 499.2215 - acc: 0.1853 - val_loss: 503.2883 - val_acc: 0.1907 Epoch 23/500 19963/19963 [==============================] - 4s - loss: 499.1185 - acc: 0.1853 - val_loss: 503.2227 - val_acc: 0.2266 Epoch 24/500 19963/19963 [==============================] - 4s - loss: 499.0264 - acc: 0.1903 - val_loss: 503.2855 - val_acc: 0.1120 Epoch 25/500 19963/19963 [==============================] - 4s - loss: 498.9257 - acc: 0.1727 - val_loss: 503.2961 - val_acc: 0.1348 Epoch 26/500 19963/19963 [==============================] - 4s - loss: 498.8296 - acc: 0.1854 - val_loss: 503.4510 - val_acc: 0.1126 Epoch 27/500 19963/19963 [==============================] - 4s - loss: 498.7325 - acc: 0.1931 - val_loss: 503.4958 - val_acc: 0.2967 Epoch 28/500 19963/19963 [==============================] - 4s - loss: 498.6436 - acc: 0.1994 - val_loss: 503.6021 - val_acc: 0.2166 Epoch 29/500 19963/19963 [==============================] - 4s - loss: 498.5487 - acc: 0.1954 - val_loss: 503.6562 - val_acc: 0.2861 Epoch 30/500 19963/19963 [==============================] - 4s - loss: 498.4737 - acc: 0.2154 - val_loss: 503.6285 - val_acc: 0.2174 Epoch 31/500 19963/19963 [==============================] - 4s - loss: 498.3974 - acc: 0.2214 - val_loss: 503.6386 - val_acc: 0.2565 Epoch 32/500 19963/19963 [==============================] - 4s - loss: 498.3082 - acc: 0.2308 - val_loss: 503.7773 - val_acc: 0.2016 Epoch 33/500 19963/19963 [==============================] - 4s - loss: 498.2250 - acc: 0.2446 - val_loss: 503.8567 - val_acc: 0.2847 Epoch 34/500 19963/19963 [==============================] - 4s - loss: 498.1548 - acc: 0.2580 - val_loss: 503.8739 - val_acc: 0.2448 Epoch 35/500 19963/19963 [==============================] - 4s - loss: 498.0624 - acc: 0.2640 - val_loss: 504.0535 - val_acc: 0.2849 Epoch 36/500 19963/19963 [==============================] - 4s - loss: 497.9936 - acc: 0.2831 - val_loss: 503.9876 - val_acc: 0.3270 Epoch 37/500 19963/19963 [==============================] - 4s - loss: 497.9291 - acc: 0.3028 - val_loss: 503.9642 - val_acc: 0.2663 Epoch 38/500 19963/19963 [==============================] - 4s - loss: 497.8539 - acc: 0.3249 - val_loss: 504.2965 - val_acc: 0.3466 Epoch 39/500 19963/19963 [==============================] - 4s - loss: 497.7858 - acc: 0.3371 - val_loss: 504.2448 - val_acc: 0.3306 Epoch 40/500 19963/19963 [==============================] - 4s - loss: 497.7325 - acc: 0.3455 - val_loss: 504.3414 - val_acc: 0.4392 Epoch 41/500 19963/19963 [==============================] - 4s - loss: 497.6698 - acc: 0.3735 - val_loss: 504.2452 - val_acc: 0.3953 Epoch 42/500 19963/19963 [==============================] - 4s - loss: 497.5922 - acc: 0.3984 - val_loss: 504.4114 - val_acc: 0.4714 Epoch 43/500 19963/19963 [==============================] - 4s - loss: 497.5279 - acc: 0.4100 - val_loss: 504.5356 - val_acc: 0.4210 Epoch 44/500 19963/19963 [==============================] - 4s - loss: 497.4763 - acc: 0.4366 - val_loss: 504.4461 - val_acc: 0.4025 Epoch 45/500 19963/19963 [==============================] - 4s - loss: 497.4147 - acc: 0.4499 - val_loss: 504.6093 - val_acc: 0.3779 Epoch 46/500 19963/19963 [==============================] - 4s - loss: 497.3665 - acc: 0.4681 - val_loss: 504.5588 - val_acc: 0.4562 Epoch 47/500 19963/19963 [==============================] - 4s - loss: 497.3015 - acc: 0.4885 - val_loss: 504.7112 - val_acc: 0.5013 Epoch 48/500 19963/19963 [==============================] - 4s - loss: 497.2564 - acc: 0.4987 - val_loss: 504.8900 - val_acc: 0.4682 Epoch 49/500 19963/19963 [==============================] - 4s - loss: 497.2158 - acc: 0.5140 - val_loss: 504.7040 - val_acc: 0.4073 Epoch 50/500 19963/19963 [==============================] - 4s - loss: 497.1621 - acc: 0.5254 - val_loss: 504.9764 - val_acc: 0.5554 Epoch 51/500 19963/19963 [==============================] - 4s - loss: 497.1012 - acc: 0.5548 - val_loss: 505.1522 - val_acc: 0.5580 Epoch 52/500 19963/19963 [==============================] - 4s - loss: 497.0671 - acc: 0.5633 - val_loss: 505.0681 - val_acc: 0.5223 Epoch 53/500 19963/19963 [==============================] - 4s - loss: 497.0238 - acc: 0.5708 - val_loss: 505.1494 - val_acc: 0.4372 Epoch 54/500 19963/19963 [==============================] - 3s - loss: 496.9774 - acc: 0.5798 - val_loss: 505.1545 - val_acc: 0.6067 Epoch 55/500 19963/19963 [==============================] - 3s - loss: 496.9288 - acc: 0.5975 - val_loss: 505.2224 - val_acc: 0.5837 Epoch 56/500 19963/19963 [==============================] - 3s - loss: 496.8779 - acc: 0.6084 - val_loss: 505.4580 - val_acc: 0.4873 Epoch 57/500 19963/19963 [==============================] - 4s - loss: 496.8303 - acc: 0.6223 - val_loss: 505.2500 - val_acc: 0.6672 Epoch 58/500 19963/19963 [==============================] - 4s - loss: 496.8028 - acc: 0.6384 - val_loss: 505.4695 - val_acc: 0.6117 Epoch 59/500 19963/19963 [==============================] - 3s - loss: 496.7449 - acc: 0.6367 - val_loss: 505.4026 - val_acc: 0.7291 Epoch 60/500 19963/19963 [==============================] - 3s - loss: 496.7252 - acc: 0.6485 - val_loss: 505.3404 - val_acc: 0.6854 Epoch 61/500 19963/19963 [==============================] - 3s - loss: 496.6695 - acc: 0.6537 - val_loss: 505.6919 - val_acc: 0.6754 Epoch 62/500
19963/19963 [==============================] - 4s - loss: 496.6298 - acc: 0.6739 - val_loss: 505.6031 - val_acc: 0.5893 Epoch 63/500 19963/19963 [==============================] - 4s - loss: 496.5867 - acc: 0.6749 - val_loss: 505.6839 - val_acc: 0.6959 Epoch 64/500 19963/19963 [==============================] - 4s - loss: 496.5602 - acc: 0.6896 - val_loss: 505.7272 - val_acc: 0.6295 Epoch 65/500 19963/19963 [==============================] - 4s - loss: 496.5212 - acc: 0.6915 - val_loss: 505.9080 - val_acc: 0.7077 Epoch 66/500 19963/19963 [==============================] - 4s - loss: 496.4646 - acc: 0.7001 - val_loss: 505.9090 - val_acc: 0.6872 Epoch 67/500 19963/19963 [==============================] - 4s - loss: 496.4422 - acc: 0.7116 - val_loss: 506.0468 - val_acc: 0.6922 Epoch 68/500 19963/19963 [==============================] - 4s - loss: 496.4119 - acc: 0.7199 - val_loss: 505.9182 - val_acc: 0.7652 Epoch 69/500 19963/19963 [==============================] - 4s - loss: 496.3752 - acc: 0.7226 - val_loss: 505.8799 - val_acc: 0.6650 Epoch 70/500 19963/19963 [==============================] - 4s - loss: 496.3287 - acc: 0.7313 - val_loss: 506.0791 - val_acc: 0.7381 Epoch 71/500 19963/19963 [==============================] - 4s - loss: 496.2896 - acc: 0.7475 - val_loss: 506.2530 - val_acc: 0.7047 Epoch 72/500 19963/19963 [==============================] - 4s - loss: 496.2791 - acc: 0.7421 - val_loss: 506.3418 - val_acc: 0.7532 Epoch 73/500 19963/19963 [==============================] - 4s - loss: 496.2597 - acc: 0.7520 - val_loss: 506.1495 - val_acc: 0.6776 Epoch 74/500 19963/19963 [==============================] - 4s - loss: 496.2109 - acc: 0.7509 - val_loss: 506.2464 - val_acc: 0.7782 Epoch 75/500 19963/19963 [==============================] - 4s - loss: 496.1969 - acc: 0.7563 - val_loss: 506.3127 - val_acc: 0.7542 Epoch 76/500 19963/19963 [==============================] - 4s - loss: 496.1508 - acc: 0.7721 - val_loss: 506.4001 - val_acc: 0.8010 Epoch 77/500 19963/19963 [==============================] - 4s - loss: 496.1274 - acc: 0.7610 - val_loss: 506.4066 - val_acc: 0.7674 Epoch 78/500 19963/19963 [==============================] - 4s - loss: 496.0915 - acc: 0.7793 - val_loss: 506.6890 - val_acc: 0.8014 Epoch 79/500 19963/19963 [==============================] - 4s - loss: 496.0701 - acc: 0.7838 - val_loss: 506.4873 - val_acc: 0.8235 Epoch 80/500 19963/19963 [==============================] - 4s - loss: 496.0370 - acc: 0.7872 - val_loss: 506.5870 - val_acc: 0.7349 Epoch 81/500 19963/19963 [==============================] - 4s - loss: 496.0149 - acc: 0.7900 - val_loss: 506.7571 - val_acc: 0.7542 Epoch 82/500 19963/19963 [==============================] - 4s - loss: 495.9887 - acc: 0.7949 - val_loss: 506.6772 - val_acc: 0.7956 Epoch 83/500 19963/19963 [==============================] - 4s - loss: 495.9526 - acc: 0.7947 - val_loss: 506.9624 - val_acc: 0.7822 Epoch 84/500 19963/19963 [==============================] - 4s - loss: 495.9517 - acc: 0.7978 - val_loss: 506.7497 - val_acc: 0.8101 Epoch 85/500 19963/19963 [==============================] - 4s - loss: 495.9068 - acc: 0.8045 - val_loss: 506.8190 - val_acc: 0.8171 Epoch 86/500 19963/19963 [==============================] - 4s - loss: 495.8695 - acc: 0.8036 - val_loss: 506.9736 - val_acc: 0.8253 Epoch 87/500 19963/19963 [==============================] - 4s - loss: 495.8618 - acc: 0.8105 - val_loss: 507.1757 - val_acc: 0.7660 Epoch 88/500 19963/19963 [==============================] - 4s - loss: 495.8229 - acc: 0.8107 - val_loss: 506.8222 - val_acc: 0.8042 Epoch 89/500 19963/19963 [==============================] - 4s - loss: 495.8041 - acc: 0.8150 - val_loss: 506.9486 - val_acc: 0.8479 Epoch 90/500 19963/19963 [==============================] - 4s - loss: 495.7913 - acc: 0.8247 - val_loss: 507.3672 - val_acc: 0.8275 Epoch 91/500 19963/19963 [==============================] - 4s - loss: 495.7597 - acc: 0.8239 - val_loss: 506.9244 - val_acc: 0.8509 Epoch 92/500 19963/19963 [==============================] - 4s - loss: 495.7321 - acc: 0.8296 - val_loss: 507.0505 - val_acc: 0.8319 Epoch 93/500 19963/19963 [==============================] - 4s - loss: 495.7026 - acc: 0.8340 - val_loss: 507.1182 - val_acc: 0.8291 Epoch 94/500 19963/19963 [==============================] - 4s - loss: 495.6915 - acc: 0.8330 - val_loss: 507.2960 - val_acc: 0.7906 Epoch 95/500 19963/19963 [==============================] - 4s - loss: 495.6799 - acc: 0.8375 - val_loss: 507.3805 - val_acc: 0.7890 Epoch 96/500 19963/19963 [==============================] - 4s - loss: 495.6380 - acc: 0.8381 - val_loss: 507.5264 - val_acc: 0.8239 Epoch 97/500 19963/19963 [==============================] - 4s - loss: 495.6330 - acc: 0.8362 - val_loss: 507.4712 - val_acc: 0.8425 Epoch 98/500 19963/19963 [==============================] - 4s - loss: 495.5924 - acc: 0.8460 - val_loss: 507.7760 - val_acc: 0.8662 Epoch 99/500 19963/19963 [==============================] - 4s - loss: 495.5892 - acc: 0.8517 - val_loss: 507.5328 - val_acc: 0.8545 Epoch 100/500 19963/19963 [==============================] - 4s - loss: 495.5482 - acc: 0.8501 - val_loss: 507.7463 - val_acc: 0.8591 Epoch 101/500 19963/19963 [==============================] - 4s - loss: 495.5459 - acc: 0.8532 - val_loss: 507.5970 - val_acc: 0.8794 Epoch 102/500 19963/19963 [==============================] - 4s - loss: 495.5041 - acc: 0.8543 - val_loss: 507.7565 - val_acc: 0.8485 Epoch 103/500 19963/19963 [==============================] - 4s - loss: 495.4949 - acc: 0.8554 - val_loss: 507.9597 - val_acc: 0.8539 Epoch 104/500 19963/19963 [==============================] - 4s - loss: 495.4827 - acc: 0.8570 - val_loss: 507.9951 - val_acc: 0.8740 Epoch 105/500 19963/19963 [==============================] - 4s - loss: 495.4498 - acc: 0.8573 - val_loss: 507.9802 - val_acc: 0.7648 Epoch 106/500 19963/19963 [==============================] - 4s - loss: 495.4317 - acc: 0.8588 - val_loss: 507.7718 - val_acc: 0.8688 Epoch 107/500 19963/19963 [==============================] - 4s - loss: 495.4141 - acc: 0.8672 - val_loss: 508.1052 - val_acc: 0.8427 Epoch 108/500 19963/19963 [==============================] - 4s - loss: 495.4020 - acc: 0.8632 - val_loss: 508.1533 - val_acc: 0.8409 Epoch 109/500 19963/19963 [==============================] - 4s - loss: 495.3818 - acc: 0.8669 - val_loss: 508.2390 - val_acc: 0.8654 Epoch 110/500 19963/19963 [==============================] - 4s - loss: 495.3660 - acc: 0.8647 - val_loss: 508.1105 - val_acc: 0.8876 Epoch 111/500 19963/19963 [==============================] - 4s - loss: 495.3293 - acc: 0.8679 - val_loss: 508.4268 - val_acc: 0.8593 Epoch 112/500 19963/19963 [==============================] - 4s - loss: 495.3288 - acc: 0.8737 - val_loss: 508.0949 - val_acc: 0.8405 Epoch 113/500 19963/19963 [==============================] - 4s - loss: 495.2931 - acc: 0.8702 - val_loss: 508.1951 - val_acc: 0.8593 Epoch 114/500 19963/19963 [==============================] - 4s - loss: 495.2874 - acc: 0.8720 - val_loss: 508.2544 - val_acc: 0.8806 Epoch 115/500 19963/19963 [==============================] - 4s - loss: 495.2631 - acc: 0.8737 - val_loss: 507.9510 - val_acc: 0.8978 Epoch 116/500 19963/19963 [==============================] - 4s - loss: 495.2437 - acc: 0.8731 - val_loss: 508.0695 - val_acc: 0.8375 Epoch 117/500 19963/19963 [==============================] - 4s - loss: 495.2461 - acc: 0.8738 - val_loss: 508.4074 - val_acc: 0.8658 Epoch 118/500 19963/19963 [==============================] - 4s - loss: 495.2379 - acc: 0.8734 - val_loss: 508.4292 - val_acc: 0.8463 Epoch 119/500 19963/19963 [==============================] - 4s - loss: 495.2113 - acc: 0.8739 - val_loss: 508.9919 - val_acc: 0.8347 Epoch 120/500 19963/19963 [==============================] - 4s - loss: 495.1892 - acc: 0.8794 - val_loss: 508.5061 - val_acc: 0.8692 Epoch 121/500 19963/19963 [==============================] - 4s - loss: 495.1607 - acc: 0.8852 - val_loss: 508.4319 - val_acc: 0.8597 Epoch 122/500 19963/19963 [==============================] - 4s - loss: 495.1425 - acc: 0.8823 - val_loss: 508.6568 - val_acc: 0.8752 Epoch 123/500
19963/19963 [==============================] - 4s - loss: 495.1186 - acc: 0.8851 - val_loss: 508.7714 - val_acc: 0.8730 Epoch 124/500 19963/19963 [==============================] - 4s - loss: 495.1231 - acc: 0.8882 - val_loss: 508.4651 - val_acc: 0.8978 Epoch 125/500 19963/19963 [==============================] - 4s - loss: 495.1060 - acc: 0.8877 - val_loss: 509.0085 - val_acc: 0.9203 Epoch 126/500 19963/19963 [==============================] - 4s - loss: 495.0852 - acc: 0.8858 - val_loss: 508.9583 - val_acc: 0.8499 Epoch 127/500 19963/19963 [==============================] - 4s - loss: 495.0680 - acc: 0.8819 - val_loss: 508.7727 - val_acc: 0.8784 Epoch 128/500 19963/19963 [==============================] - 4s - loss: 495.0595 - acc: 0.8847 - val_loss: 508.9968 - val_acc: 0.8786 Epoch 129/500 19963/19963 [==============================] - 4s - loss: 495.0271 - acc: 0.8868 - val_loss: 508.9998 - val_acc: 0.8511 Epoch 130/500 19963/19963 [==============================] - 4s - loss: 495.0220 - acc: 0.8853 - val_loss: 509.0853 - val_acc: 0.8804 Epoch 131/500 19963/19963 [==============================] - 4s - loss: 495.0011 - acc: 0.8863 - val_loss: 509.2207 - val_acc: 0.8527 Epoch 132/500 19963/19963 [==============================] - 4s - loss: 495.0155 - acc: 0.8931 - val_loss: 509.0804 - val_acc: 0.8880 Epoch 133/500 19963/19963 [==============================] - 4s - loss: 494.9755 - acc: 0.8913 - val_loss: 509.0576 - val_acc: 0.9038 Epoch 134/500 19963/19963 [==============================] - 4s - loss: 494.9738 - acc: 0.8908 - val_loss: 509.2794 - val_acc: 0.8541 Epoch 135/500 19963/19963 [==============================] - 4s - loss: 494.9564 - acc: 0.8922 - val_loss: 509.6963 - val_acc: 0.8371 Epoch 136/500 19963/19963 [==============================] - 4s - loss: 494.9171 - acc: 0.8952 - val_loss: 509.2114 - val_acc: 0.8966 Epoch 137/500 19963/19963 [==============================] - 4s - loss: 494.9248 - acc: 0.8952 - val_loss: 509.4895 - val_acc: 0.8682 Epoch 138/500 19963/19963 [==============================] - 4s - loss: 494.9100 - acc: 0.8951 - val_loss: 509.4049 - val_acc: 0.8908 Epoch 139/500 19963/19963 [==============================] - 4s - loss: 494.8894 - acc: 0.8993 - val_loss: 509.3632 - val_acc: 0.8724 Epoch 140/500 19963/19963 [==============================] - 4s - loss: 494.8787 - acc: 0.8992 - val_loss: 509.5401 - val_acc: 0.8908 Epoch 141/500 19963/19963 [==============================] - 4s - loss: 494.8604 - acc: 0.8975 - val_loss: 509.8726 - val_acc: 0.9140 Epoch 142/500 19963/19963 [==============================] - 4s - loss: 494.8521 - acc: 0.9044 - val_loss: 509.3432 - val_acc: 0.8830 Epoch 143/500 19963/19963 [==============================] - 4s - loss: 494.8378 - acc: 0.9029 - val_loss: 509.8774 - val_acc: 0.8575 Epoch 144/500 19963/19963 [==============================] - 4s - loss: 494.8332 - acc: 0.9009 - val_loss: 509.9265 - val_acc: 0.8964 Epoch 145/500 19963/19963 [==============================] - 4s - loss: 494.8141 - acc: 0.9026 - val_loss: 509.7003 - val_acc: 0.9118 Epoch 146/500 19963/19963 [==============================] - 4s - loss: 494.8138 - acc: 0.9071 - val_loss: 509.8207 - val_acc: 0.8922 Epoch 147/500 19963/19963 [==============================] - 4s - loss: 494.7802 - acc: 0.9055 - val_loss: 509.6696 - val_acc: 0.8900 Epoch 148/500 19963/19963 [==============================] - 4s - loss: 494.7762 - acc: 0.9023 - val_loss: 509.7778 - val_acc: 0.9209 Epoch 149/500 19963/19963 [==============================] - 4s - loss: 494.7642 - acc: 0.9060 - val_loss: 509.9377 - val_acc: 0.9130 Epoch 150/500 19963/19963 [==============================] - 4s - loss: 494.7440 - acc: 0.9111 - val_loss: 510.2611 - val_acc: 0.8928 Epoch 151/500 19963/19963 [==============================] - 4s - loss: 494.7320 - acc: 0.9066 - val_loss: 509.9017 - val_acc: 0.9225 Epoch 152/500 19963/19963 [==============================] - 4s - loss: 494.7211 - acc: 0.9080 - val_loss: 510.2036 - val_acc: 0.9046 Epoch 153/500 19963/19963 [==============================] - 4s - loss: 494.7028 - acc: 0.9082 - val_loss: 509.9755 - val_acc: 0.9158 Epoch 154/500 19963/19963 [==============================] - 4s - loss: 494.6833 - acc: 0.9098 - val_loss: 509.9168 - val_acc: 0.8872 Epoch 155/500 19963/19963 [==============================] - 4s - loss: 494.6870 - acc: 0.9107 - val_loss: 509.8072 - val_acc: 0.8842 Epoch 156/500 19963/19963 [==============================] - 4s - loss: 494.6775 - acc: 0.9117 - val_loss: 510.2043 - val_acc: 0.9078 Epoch 157/500 19963/19963 [==============================] - 4s - loss: 494.6643 - acc: 0.9099 - val_loss: 510.3874 - val_acc: 0.9054 Epoch 158/500 19963/19963 [==============================] - 4s - loss: 494.6609 - acc: 0.9114 - val_loss: 510.2125 - val_acc: 0.8994 Epoch 159/500 19963/19963 [==============================] - 4s - loss: 494.6342 - acc: 0.9135 - val_loss: 510.7567 - val_acc: 0.8882 Epoch 160/500 19963/19963 [==============================] - 4s - loss: 494.6283 - acc: 0.9123 - val_loss: 510.7425 - val_acc: 0.9044 Epoch 161/500 19963/19963 [==============================] - 4s - loss: 494.6078 - acc: 0.9143 - val_loss: 510.8797 - val_acc: 0.9086 Epoch 162/500 19963/19963 [==============================] - 4s - loss: 494.5912 - acc: 0.9164 - val_loss: 510.6641 - val_acc: 0.9142 Epoch 163/500 19963/19963 [==============================] - 4s - loss: 494.5934 - acc: 0.9154 - val_loss: 510.5877 - val_acc: 0.9060 Epoch 164/500 19963/19963 [==============================] - 4s - loss: 494.5855 - acc: 0.9164 - val_loss: 510.8551 - val_acc: 0.9235 Epoch 165/500 19963/19963 [==============================] - 4s - loss: 494.5503 - acc: 0.9229 - val_loss: 510.7812 - val_acc: 0.9335 Epoch 166/500 19963/19963 [==============================] - 4s - loss: 494.5422 - acc: 0.9226 - val_loss: 510.6672 - val_acc: 0.9183 Epoch 167/500 19963/19963 [==============================] - 4s - loss: 494.5441 - acc: 0.9217 - val_loss: 510.9202 - val_acc: 0.9173 Epoch 168/500 19963/19963 [==============================] - 4s - loss: 494.5300 - acc: 0.9200 - val_loss: 510.8307 - val_acc: 0.8992 Epoch 169/500 19963/19963 [==============================] - 4s - loss: 494.5317 - acc: 0.9216 - val_loss: 510.9512 - val_acc: 0.9297 Epoch 170/500 19963/19963 [==============================] - 4s - loss: 494.5035 - acc: 0.9251 - val_loss: 511.2561 - val_acc: 0.9118 Epoch 171/500 19963/19963 [==============================] - 4s - loss: 494.4820 - acc: 0.9210 - val_loss: 510.9530 - val_acc: 0.9217 Epoch 172/500 19963/19963 [==============================] - 4s - loss: 494.4756 - acc: 0.9248 - val_loss: 510.6623 - val_acc: 0.9088 Epoch 173/500 19963/19963 [==============================] - 4s - loss: 494.4650 - acc: 0.9267 - val_loss: 510.8010 - val_acc: 0.9239 Epoch 174/500 19963/19963 [==============================] - 4s - loss: 494.4804 - acc: 0.9231 - val_loss: 511.1056 - val_acc: 0.9114 Epoch 175/500 19963/19963 [==============================] - 4s - loss: 494.4666 - acc: 0.9227 - val_loss: 511.2407 - val_acc: 0.9337 Epoch 176/500 19963/19963 [==============================] - 4s - loss: 494.4561 - acc: 0.9264 - val_loss: 511.2170 - val_acc: 0.9251 Epoch 177/500 19963/19963 [==============================] - 4s - loss: 494.4301 - acc: 0.9259 - val_loss: 511.1347 - val_acc: 0.9054 Epoch 178/500 19963/19963 [==============================] - 4s - loss: 494.4055 - acc: 0.9239 - val_loss: 511.1440 - val_acc: 0.8962 Epoch 179/500 19963/19963 [==============================] - 4s - loss: 494.4178 - acc: 0.9284 - val_loss: 511.8084 - val_acc: 0.9090 Epoch 180/500 19963/19963 [==============================] - 4s - loss: 494.3943 - acc: 0.9258 - val_loss: 511.4693 - val_acc: 0.9203 Epoch 181/500 19963/19963 [==============================] - 4s - loss: 494.3763 - acc: 0.9291 - val_loss: 511.4683 - val_acc: 0.9179 Epoch 182/500 19963/19963 [==============================] - 4s - loss: 494.3780 - acc: 0.9291 - val_loss: 511.8275 - val_acc: 0.9183 Epoch 183/500 19963/19963 [==============================] - 4s - loss: 494.3834 - acc: 0.9301 - val_loss: 511.3238 - val_acc: 0.9393 Epoch 184/500
19963/19963 [==============================] - 4s - loss: 494.3424 - acc: 0.9300 - val_loss: 512.0930 - val_acc: 0.9217 Epoch 185/500 19963/19963 [==============================] - 4s - loss: 494.3438 - acc: 0.9284 - val_loss: 510.9834 - val_acc: 0.9269 Epoch 186/500 19963/19963 [==============================] - 4s - loss: 494.3195 - acc: 0.9291 - val_loss: 511.9937 - val_acc: 0.9126 Epoch 187/500 19963/19963 [==============================] - 4s - loss: 494.3074 - acc: 0.9290 - val_loss: 511.9741 - val_acc: 0.9074 Epoch 188/500 19963/19963 [==============================] - 4s - loss: 494.3191 - acc: 0.9289 - val_loss: 511.4462 - val_acc: 0.9337 Epoch 189/500 19963/19963 [==============================] - 4s - loss: 494.2997 - acc: 0.9299 - val_loss: 511.5951 - val_acc: 0.9223 Epoch 190/500 19963/19963 [==============================] - 4s - loss: 494.2854 - acc: 0.9302 - val_loss: 511.7262 - val_acc: 0.9253 Epoch 191/500 19963/19963 [==============================] - 4s - loss: 494.2832 - acc: 0.9328 - val_loss: 511.9000 - val_acc: 0.9120 Epoch 192/500 19963/19963 [==============================] - 4s - loss: 494.2785 - acc: 0.9301 - val_loss: 512.1108 - val_acc: 0.9144 Epoch 193/500 19963/19963 [==============================] - 4s - loss: 494.2654 - acc: 0.9324 - val_loss: 511.8300 - val_acc: 0.9269 Epoch 194/500 19963/19963 [==============================] - 4s - loss: 494.2517 - acc: 0.9340 - val_loss: 511.8326 - val_acc: 0.9295 Epoch 195/500 19963/19963 [==============================] - 4s - loss: 494.2496 - acc: 0.9335 - val_loss: 511.6844 - val_acc: 0.9197 Epoch 196/500 19963/19963 [==============================] - 4s - loss: 494.2229 - acc: 0.9326 - val_loss: 511.9269 - val_acc: 0.9245 Epoch 197/500 19963/19963 [==============================] - 4s - loss: 494.2298 - acc: 0.9350 - val_loss: 512.0595 - val_acc: 0.9160 Epoch 198/500 19963/19963 [==============================] - 4s - loss: 494.2225 - acc: 0.9326 - val_loss: 512.3188 - val_acc: 0.9309 Epoch 199/500 19963/19963 [==============================] - 4s - loss: 494.1885 - acc: 0.9357 - val_loss: 512.5320 - val_acc: 0.9255 Epoch 200/500 19963/19963 [==============================] - 4s - loss: 494.1947 - acc: 0.9355 - val_loss: 512.3463 - val_acc: 0.9241 Epoch 201/500 19963/19963 [==============================] - 4s - loss: 494.1628 - acc: 0.9344 - val_loss: 512.0463 - val_acc: 0.9385 Epoch 202/500 19963/19963 [==============================] - 4s - loss: 494.1620 - acc: 0.9349 - val_loss: 512.0062 - val_acc: 0.9291 Epoch 203/500 19963/19963 [==============================] - 4s - loss: 494.1768 - acc: 0.9362 - val_loss: 512.4322 - val_acc: 0.9391 Epoch 204/500 19963/19963 [==============================] - 4s - loss: 494.1721 - acc: 0.9350 - val_loss: 512.4579 - val_acc: 0.9307 Epoch 205/500 19963/19963 [==============================] - 4s - loss: 494.1354 - acc: 0.9362 - val_loss: 512.4993 - val_acc: 0.9257 Epoch 206/500 19963/19963 [==============================] - 4s - loss: 494.1246 - acc: 0.9390 - val_loss: 512.5899 - val_acc: 0.9535 Epoch 207/500 19963/19963 [==============================] - 4s - loss: 494.1312 - acc: 0.9381 - val_loss: 512.6548 - val_acc: 0.9293 Epoch 208/500 19963/19963 [==============================] - 4s - loss: 494.1335 - acc: 0.9390 - val_loss: 512.5760 - val_acc: 0.8958 Epoch 209/500 19963/19963 [==============================] - 4s - loss: 494.1104 - acc: 0.9365 - val_loss: 513.0113 - val_acc: 0.9399 Epoch 210/500 19963/19963 [==============================] - 4s - loss: 494.0968 - acc: 0.9393 - val_loss: 513.2433 - val_acc: 0.9263 Epoch 211/500 19963/19963 [==============================] - 4s - loss: 494.0933 - acc: 0.9395 - val_loss: 512.7495 - val_acc: 0.9281 Epoch 212/500 19963/19963 [==============================] - 4s - loss: 494.1035 - acc: 0.9399 - val_loss: 512.7977 - val_acc: 0.9205 Epoch 213/500 19963/19963 [==============================] - 4s - loss: 494.0760 - acc: 0.9382 - val_loss: 512.5744 - val_acc: 0.9205 Epoch 214/500 19963/19963 [==============================] - 4s - loss: 494.0583 - acc: 0.9380 - val_loss: 512.7637 - val_acc: 0.9423 Epoch 215/500 19963/19963 [==============================] - 4s - loss: 494.0423 - acc: 0.9410 - val_loss: 512.8654 - val_acc: 0.9197 Epoch 216/500 19963/19963 [==============================] - 4s - loss: 494.0524 - acc: 0.9424 - val_loss: 512.8313 - val_acc: 0.9317 Epoch 217/500 19963/19963 [==============================] - 4s - loss: 494.0609 - acc: 0.9409 - val_loss: 513.4910 - val_acc: 0.9090 Epoch 218/500 19963/19963 [==============================] - 4s - loss: 494.0287 - acc: 0.9435 - val_loss: 512.5612 - val_acc: 0.9437 Epoch 219/500 19963/19963 [==============================] - 4s - loss: 494.0340 - acc: 0.9425 - val_loss: 513.4890 - val_acc: 0.9373 Epoch 220/500 19963/19963 [==============================] - 4s - loss: 494.0093 - acc: 0.9421 - val_loss: 513.5374 - val_acc: 0.9301 Epoch 221/500 19963/19963 [==============================] - 4s - loss: 494.0179 - acc: 0.9431 - val_loss: 512.9200 - val_acc: 0.9401 Epoch 222/500 19963/19963 [==============================] - 4s - loss: 493.9808 - acc: 0.9443 - val_loss: 513.4616 - val_acc: 0.9423 Epoch 223/500 19963/19963 [==============================] - 4s - loss: 493.9903 - acc: 0.9446 - val_loss: 513.4754 - val_acc: 0.9333 Epoch 224/500 19963/19963 [==============================] - 4s - loss: 493.9987 - acc: 0.9447 - val_loss: 513.8150 - val_acc: 0.9363 Epoch 225/500 19963/19963 [==============================] - 4s - loss: 493.9723 - acc: 0.9450 - val_loss: 513.1297 - val_acc: 0.9539 Epoch 226/500 19963/19963 [==============================] - 4s - loss: 493.9611 - acc: 0.9433 - val_loss: 513.7288 - val_acc: 0.9411 Epoch 227/500 19963/19963 [==============================] - 4s - loss: 493.9515 - acc: 0.9440 - val_loss: 513.4886 - val_acc: 0.9331 Epoch 228/500 19963/19963 [==============================] - 4s - loss: 493.9436 - acc: 0.9437 - val_loss: 513.0628 - val_acc: 0.9377 Epoch 229/500 19963/19963 [==============================] - 4s - loss: 493.9549 - acc: 0.9464 - val_loss: 513.8014 - val_acc: 0.9435 Epoch 230/500 19963/19963 [==============================] - 4s - loss: 493.9336 - acc: 0.9467 - val_loss: 513.6699 - val_acc: 0.9285 Epoch 231/500 19963/19963 [==============================] - 4s - loss: 493.9257 - acc: 0.9456 - val_loss: 514.0978 - val_acc: 0.9449 Epoch 232/500 19963/19963 [==============================] - 4s - loss: 493.9111 - acc: 0.9447 - val_loss: 513.4519 - val_acc: 0.9381 Epoch 233/500 19963/19963 [==============================] - 4s - loss: 493.8898 - acc: 0.9436 - val_loss: 513.4694 - val_acc: 0.9311 Epoch 234/500 19963/19963 [==============================] - 4s - loss: 493.8833 - acc: 0.9465 - val_loss: 513.9420 - val_acc: 0.9321 Epoch 235/500 19963/19963 [==============================] - 4s - loss: 493.8900 - acc: 0.9471 - val_loss: 513.5383 - val_acc: 0.9387 Epoch 236/500 19963/19963 [==============================] - 4s - loss: 493.8811 - acc: 0.9476 - val_loss: 514.1607 - val_acc: 0.9421 Epoch 237/500 19963/19963 [==============================] - 4s - loss: 493.8919 - acc: 0.9445 - val_loss: 514.2876 - val_acc: 0.9361 Epoch 238/500 19963/19963 [==============================] - 4s - loss: 493.8529 - acc: 0.9440 - val_loss: 513.8998 - val_acc: 0.9311 Epoch 239/500 19963/19963 [==============================] - 4s - loss: 493.8561 - acc: 0.9488 - val_loss: 513.5985 - val_acc: 0.9463 Epoch 240/500 19963/19963 [==============================] - 4s - loss: 493.8504 - acc: 0.9478 - val_loss: 513.8540 - val_acc: 0.9579 Epoch 241/500 19963/19963 [==============================] - 4s - loss: 493.8443 - acc: 0.9468 - val_loss: 513.8317 - val_acc: 0.9475 Epoch 242/500 19963/19963 [==============================] - 4s - loss: 493.8297 - acc: 0.9462 - val_loss: 514.6547 - val_acc: 0.9513 Epoch 243/500 19963/19963 [==============================] - 4s - loss: 493.8303 - acc: 0.9475 - val_loss: 514.4881 - val_acc: 0.9507 Epoch 244/500 19963/19963 [==============================] - 4s - loss: 493.8313 - acc: 0.9481 - val_loss: 514.7005 - val_acc: 0.9519 Epoch 245/500
19963/19963 [==============================] - 4s - loss: 493.8071 - acc: 0.9487 - val_loss: 514.3960 - val_acc: 0.9405 Epoch 246/500 19963/19963 [==============================] - 4s - loss: 493.8041 - acc: 0.9489 - val_loss: 514.0023 - val_acc: 0.9319 Epoch 247/500 19963/19963 [==============================] - 4s - loss: 493.7912 - acc: 0.9510 - val_loss: 514.8714 - val_acc: 0.9495 Epoch 248/500 19963/19963 [==============================] - 4s - loss: 493.8028 - acc: 0.9486 - val_loss: 514.3183 - val_acc: 0.9349 Epoch 249/500 19963/19963 [==============================] - 4s - loss: 493.7943 - acc: 0.9494 - val_loss: 514.6188 - val_acc: 0.9337 Epoch 250/500 19963/19963 [==============================] - 4s - loss: 493.7955 - acc: 0.9485 - val_loss: 514.5264 - val_acc: 0.9413 Epoch 251/500 19963/19963 [==============================] - 4s - loss: 493.7814 - acc: 0.9492 - val_loss: 514.4235 - val_acc: 0.9405 Epoch 252/500 19963/19963 [==============================] - 4s - loss: 493.7353 - acc: 0.9522 - val_loss: 514.3801 - val_acc: 0.9475 Epoch 253/500 19963/19963 [==============================] - 4s - loss: 493.7448 - acc: 0.9470 - val_loss: 514.2343 - val_acc: 0.9331 Epoch 254/500 19963/19963 [==============================] - 4s - loss: 493.7596 - acc: 0.9505 - val_loss: 514.6797 - val_acc: 0.9439 Epoch 255/500 19963/19963 [==============================] - 4s - loss: 493.7345 - acc: 0.9480 - val_loss: 515.0977 - val_acc: 0.9341 Epoch 256/500 19963/19963 [==============================] - 4s - loss: 493.7258 - acc: 0.9479 - val_loss: 514.2398 - val_acc: 0.9397 Epoch 257/500 19963/19963 [==============================] - 4s - loss: 493.7318 - acc: 0.9488 - val_loss: 514.9191 - val_acc: 0.9491 Epoch 258/500 19963/19963 [==============================] - 4s - loss: 493.7321 - acc: 0.9485 - val_loss: 515.0298 - val_acc: 0.9551 Epoch 259/500 19963/19963 [==============================] - 4s - loss: 493.7037 - acc: 0.9482 - val_loss: 514.5596 - val_acc: 0.9469 Epoch 260/500 19963/19963 [==============================] - 4s - loss: 493.6986 - acc: 0.9493 - val_loss: 514.8626 - val_acc: 0.9367 Epoch 261/500 19963/19963 [==============================] - 4s - loss: 493.6958 - acc: 0.9509 - val_loss: 515.1639 - val_acc: 0.9527 Epoch 262/500 19963/19963 [==============================] - 4s - loss: 493.6874 - acc: 0.9494 - val_loss: 515.0715 - val_acc: 0.9551 Epoch 263/500 19963/19963 [==============================] - 4s - loss: 493.6756 - acc: 0.9498 - val_loss: 514.1674 - val_acc: 0.9427 Epoch 264/500 19963/19963 [==============================] - 4s - loss: 493.6773 - acc: 0.9506 - val_loss: 515.0842 - val_acc: 0.9433 Epoch 265/500 19963/19963 [==============================] - 4s - loss: 493.6870 - acc: 0.9511 - val_loss: 515.6192 - val_acc: 0.9289 Epoch 266/500 19963/19963 [==============================] - 4s - loss: 493.6729 - acc: 0.9494 - val_loss: 515.6511 - val_acc: 0.9261 Epoch 267/500 19963/19963 [==============================] - 4s - loss: 493.6542 - acc: 0.9517 - val_loss: 515.4828 - val_acc: 0.9403 Epoch 268/500 19963/19963 [==============================] - 4s - loss: 493.6481 - acc: 0.9521 - val_loss: 515.6330 - val_acc: 0.9275 Epoch 269/500 19963/19963 [==============================] - 4s - loss: 493.6367 - acc: 0.9529 - val_loss: 515.9627 - val_acc: 0.9503 Epoch 270/500 19963/19963 [==============================] - 4s - loss: 493.6380 - acc: 0.9525 - val_loss: 515.7559 - val_acc: 0.9429 Epoch 271/500 19963/19963 [==============================] - 4s - loss: 493.6359 - acc: 0.9530 - val_loss: 515.1681 - val_acc: 0.9555 Epoch 272/500 19963/19963 [==============================] - 4s - loss: 493.6245 - acc: 0.9499 - val_loss: 515.8812 - val_acc: 0.9529 Epoch 273/500 19963/19963 [==============================] - 4s - loss: 493.6090 - acc: 0.9546 - val_loss: 515.5764 - val_acc: 0.9467 Epoch 274/500 19963/19963 [==============================] - 4s - loss: 493.6160 - acc: 0.9533 - val_loss: 515.7477 - val_acc: 0.9461 Epoch 275/500 19963/19963 [==============================] - 4s - loss: 493.6063 - acc: 0.9523 - val_loss: 515.3288 - val_acc: 0.9509 Epoch 276/500 19963/19963 [==============================] - 4s - loss: 493.5916 - acc: 0.9537 - val_loss: 515.5664 - val_acc: 0.9513 Epoch 277/500 19963/19963 [==============================] - 4s - loss: 493.5915 - acc: 0.9542 - val_loss: 516.1525 - val_acc: 0.9535 Epoch 278/500 19963/19963 [==============================] - 4s - loss: 493.5844 - acc: 0.9530 - val_loss: 515.4103 - val_acc: 0.9557 Epoch 279/500 19963/19963 [==============================] - 4s - loss: 493.5705 - acc: 0.9520 - val_loss: 515.9302 - val_acc: 0.9461 Epoch 280/500 19963/19963 [==============================] - 4s - loss: 493.5680 - acc: 0.9539 - val_loss: 515.8831 - val_acc: 0.9497 Epoch 281/500 19963/19963 [==============================] - 4s - loss: 493.5684 - acc: 0.9517 - val_loss: 515.8381 - val_acc: 0.9309 Epoch 282/500 19963/19963 [==============================] - 4s - loss: 493.5525 - acc: 0.9524 - val_loss: 516.3065 - val_acc: 0.9489 Epoch 283/500 19963/19963 [==============================] - 4s - loss: 493.5566 - acc: 0.9530 - val_loss: 516.4894 - val_acc: 0.9487 Epoch 284/500 19963/19963 [==============================] - 4s - loss: 493.5256 - acc: 0.9536 - val_loss: 517.0733 - val_acc: 0.9591 Epoch 285/500 19963/19963 [==============================] - 4s - loss: 493.5413 - acc: 0.9552 - val_loss: 515.8981 - val_acc: 0.9499 Epoch 286/500 19963/19963 [==============================] - 4s - loss: 493.5279 - acc: 0.9537 - val_loss: 515.9369 - val_acc: 0.9625 Epoch 287/500 19963/19963 [==============================] - 4s - loss: 493.5251 - acc: 0.9566 - val_loss: 515.2999 - val_acc: 0.9519 Epoch 288/500 19963/19963 [==============================] - 4s - loss: 493.5078 - acc: 0.9560 - val_loss: 515.9258 - val_acc: 0.9537 Epoch 289/500 19963/19963 [==============================] - 4s - loss: 493.5246 - acc: 0.9547 - val_loss: 515.7352 - val_acc: 0.9621 Epoch 290/500 19963/19963 [==============================] - 4s - loss: 493.5112 - acc: 0.9540 - val_loss: 516.4518 - val_acc: 0.9499 Epoch 291/500 19963/19963 [==============================] - 4s - loss: 493.4941 - acc: 0.9570 - val_loss: 516.4888 - val_acc: 0.9609 Epoch 292/500 19963/19963 [==============================] - 4s - loss: 493.4796 - acc: 0.9532 - val_loss: 516.6315 - val_acc: 0.9503 Epoch 293/500 19963/19963 [==============================] - 4s - loss: 493.4836 - acc: 0.9577 - val_loss: 516.3505 - val_acc: 0.9567 Epoch 294/500 19963/19963 [==============================] - 4s - loss: 493.4852 - acc: 0.9554 - val_loss: 516.0478 - val_acc: 0.9569 Epoch 295/500 19963/19963 [==============================] - 4s - loss: 493.4734 - acc: 0.9552 - val_loss: 516.5111 - val_acc: 0.9585 Epoch 296/500 19963/19963 [==============================] - 4s - loss: 493.4499 - acc: 0.9557 - val_loss: 515.8930 - val_acc: 0.9547 Epoch 297/500 19963/19963 [==============================] - 4s - loss: 493.4714 - acc: 0.9562 - val_loss: 516.2444 - val_acc: 0.9585 Epoch 298/500 19963/19963 [==============================] - 4s - loss: 493.4527 - acc: 0.9550 - val_loss: 517.4130 - val_acc: 0.9437 Epoch 299/500 19963/19963 [==============================] - 4s - loss: 493.4634 - acc: 0.9562 - val_loss: 516.6477 - val_acc: 0.9553 Epoch 300/500 19963/19963 [==============================] - 4s - loss: 493.4438 - acc: 0.9553 - val_loss: 516.2735 - val_acc: 0.9559 Epoch 301/500 19963/19963 [==============================] - 4s - loss: 493.4608 - acc: 0.9550 - val_loss: 516.3542 - val_acc: 0.9523 Epoch 302/500 19963/19963 [==============================] - 4s - loss: 493.4359 - acc: 0.9567 - val_loss: 516.4326 - val_acc: 0.9583 Epoch 303/500 19963/19963 [==============================] - 4s - loss: 493.4179 - acc: 0.9546 - val_loss: 516.1241 - val_acc: 0.9497 Epoch 304/500 19963/19963 [==============================] - 4s - loss: 493.4274 - acc: 0.9550 - val_loss: 517.7369 - val_acc: 0.9427 Epoch 305/500 19963/19963 [==============================] - 4s - loss: 493.4160 - acc: 0.9563 - val_loss: 516.8111 - val_acc: 0.9571 Epoch 306/500
19963/19963 [==============================] - 4s - loss: 493.4076 - acc: 0.9568 - val_loss: 516.4242 - val_acc: 0.9547 Epoch 307/500 19963/19963 [==============================] - 4s - loss: 493.3829 - acc: 0.9559 - val_loss: 516.8223 - val_acc: 0.9595 Epoch 308/500 19963/19963 [==============================] - 4s - loss: 493.4070 - acc: 0.9561 - val_loss: 516.5648 - val_acc: 0.9583 Epoch 309/500 19963/19963 [==============================] - 4s - loss: 493.3681 - acc: 0.9577 - val_loss: 516.8094 - val_acc: 0.9447 Epoch 310/500 19963/19963 [==============================] - 4s - loss: 493.3924 - acc: 0.9562 - val_loss: 517.5253 - val_acc: 0.9625 Epoch 311/500 19963/19963 [==============================] - 4s - loss: 493.3689 - acc: 0.9560 - val_loss: 517.4248 - val_acc: 0.9511 Epoch 312/500 19963/19963 [==============================] - 4s - loss: 493.3694 - acc: 0.9569 - val_loss: 516.7473 - val_acc: 0.9561 Epoch 313/500 19963/19963 [==============================] - 4s - loss: 493.3511 - acc: 0.9560 - val_loss: 517.6065 - val_acc: 0.9479 Epoch 314/500 19963/19963 [==============================] - 4s - loss: 493.3551 - acc: 0.9591 - val_loss: 516.5301 - val_acc: 0.9541 Epoch 315/500 19963/19963 [==============================] - 4s - loss: 493.3529 - acc: 0.9561 - val_loss: 517.5550 - val_acc: 0.9555 Epoch 316/500 19963/19963 [==============================] - 4s - loss: 493.3333 - acc: 0.9569 - val_loss: 518.2606 - val_acc: 0.9469 Epoch 317/500 19963/19963 [==============================] - 4s - loss: 493.3525 - acc: 0.9581 - val_loss: 516.8525 - val_acc: 0.9545 Epoch 318/500 19963/19963 [==============================] - 4s - loss: 493.3532 - acc: 0.9560 - val_loss: 517.1330 - val_acc: 0.9509 Epoch 319/500 19963/19963 [==============================] - 4s - loss: 493.3487 - acc: 0.9573 - val_loss: 517.4108 - val_acc: 0.9389 Epoch 320/500 19963/19963 [==============================] - 4s - loss: 493.3159 - acc: 0.9575 - val_loss: 517.6544 - val_acc: 0.9499 Epoch 321/500 19963/19963 [==============================] - 4s - loss: 493.3206 - acc: 0.9579 - val_loss: 517.5033 - val_acc: 0.9477 Epoch 322/500 19963/19963 [==============================] - 4s - loss: 493.2915 - acc: 0.9588 - val_loss: 518.1011 - val_acc: 0.9593 Epoch 323/500 19963/19963 [==============================] - 4s - loss: 493.3020 - acc: 0.9581 - val_loss: 517.1859 - val_acc: 0.9527 Epoch 324/500 19963/19963 [==============================] - 4s - loss: 493.2971 - acc: 0.9587 - val_loss: 517.7230 - val_acc: 0.9531 Epoch 325/500 19963/19963 [==============================] - 4s - loss: 493.3108 - acc: 0.9583 - val_loss: 517.5614 - val_acc: 0.9571 Epoch 326/500 19963/19963 [==============================] - 4s - loss: 493.2838 - acc: 0.9602 - val_loss: 517.8998 - val_acc: 0.9573 Epoch 327/500 19963/19963 [==============================] - 4s - loss: 493.3117 - acc: 0.9595 - val_loss: 517.8227 - val_acc: 0.9623 Epoch 328/500 19963/19963 [==============================] - 4s - loss: 493.2910 - acc: 0.9586 - val_loss: 517.5640 - val_acc: 0.9567 Epoch 329/500 19963/19963 [==============================] - 4s - loss: 493.2700 - acc: 0.9588 - val_loss: 517.6651 - val_acc: 0.9651 Epoch 330/500 19963/19963 [==============================] - 4s - loss: 493.2673 - acc: 0.9603 - val_loss: 517.6770 - val_acc: 0.9541 Epoch 331/500 19963/19963 [==============================] - 4s - loss: 493.2672 - acc: 0.9587 - val_loss: 518.3985 - val_acc: 0.9551 Epoch 332/500 19963/19963 [==============================] - 4s - loss: 493.2508 - acc: 0.9588 - val_loss: 517.5999 - val_acc: 0.9517 Epoch 333/500 19963/19963 [==============================] - 4s - loss: 493.2355 - acc: 0.9598 - val_loss: 517.2874 - val_acc: 0.9517 Epoch 334/500 19963/19963 [==============================] - 4s - loss: 493.2562 - acc: 0.9599 - val_loss: 517.8885 - val_acc: 0.9559 Epoch 335/500 19963/19963 [==============================] - 4s - loss: 493.2527 - acc: 0.9601 - val_loss: 517.7571 - val_acc: 0.9623 Epoch 336/500 19963/19963 [==============================] - 4s - loss: 493.2429 - acc: 0.9598 - val_loss: 517.5460 - val_acc: 0.9617 Epoch 337/500 19963/19963 [==============================] - 4s - loss: 493.2272 - acc: 0.9616 - val_loss: 518.1440 - val_acc: 0.9573 Epoch 338/500 19963/19963 [==============================] - 4s - loss: 493.2339 - acc: 0.9610 - val_loss: 517.9645 - val_acc: 0.9655 Epoch 339/500 19963/19963 [==============================] - 4s - loss: 493.2229 - acc: 0.9628 - val_loss: 518.5344 - val_acc: 0.9559 Epoch 340/500 19963/19963 [==============================] - 4s - loss: 493.2468 - acc: 0.9624 - val_loss: 517.8453 - val_acc: 0.9623 Epoch 341/500 19963/19963 [==============================] - 4s - loss: 493.1929 - acc: 0.9612 - val_loss: 517.8809 - val_acc: 0.9635 Epoch 342/500 19963/19963 [==============================] - 4s - loss: 493.2012 - acc: 0.9614 - val_loss: 518.1192 - val_acc: 0.9649 Epoch 343/500 19963/19963 [==============================] - 4s - loss: 493.2104 - acc: 0.9613 - val_loss: 518.0911 - val_acc: 0.9567 Epoch 344/500 19963/19963 [==============================] - 4s - loss: 493.1742 - acc: 0.9610 - val_loss: 518.3220 - val_acc: 0.9637 Epoch 345/500 19963/19963 [==============================] - 4s - loss: 493.1951 - acc: 0.9630 - val_loss: 518.3847 - val_acc: 0.9667 Epoch 346/500 19963/19963 [==============================] - 4s - loss: 493.1822 - acc: 0.9611 - val_loss: 519.2944 - val_acc: 0.9603 Epoch 347/500 19963/19963 [==============================] - 4s - loss: 493.1843 - acc: 0.9600 - val_loss: 518.2546 - val_acc: 0.9565 Epoch 348/500 19963/19963 [==============================] - 4s - loss: 493.1722 - acc: 0.9617 - val_loss: 518.4207 - val_acc: 0.9577 Epoch 349/500 19963/19963 [==============================] - 4s - loss: 493.1673 - acc: 0.9603 - val_loss: 518.4414 - val_acc: 0.9619 Epoch 350/500 19963/19963 [==============================] - 4s - loss: 493.1564 - acc: 0.9620 - val_loss: 518.0800 - val_acc: 0.9555 Epoch 351/500 19963/19963 [==============================] - 4s - loss: 493.1728 - acc: 0.9602 - val_loss: 518.5693 - val_acc: 0.9579 Epoch 352/500 19963/19963 [==============================] - 4s - loss: 493.1423 - acc: 0.9619 - val_loss: 519.8097 - val_acc: 0.9511 Epoch 353/500 19963/19963 [==============================] - 4s - loss: 493.1604 - acc: 0.9614 - val_loss: 519.2065 - val_acc: 0.9535 Epoch 354/500 19963/19963 [==============================] - 4s - loss: 493.1419 - acc: 0.9616 - val_loss: 518.6050 - val_acc: 0.9639 Epoch 355/500 19963/19963 [==============================] - 4s - loss: 493.1349 - acc: 0.9627 - val_loss: 519.0150 - val_acc: 0.9617 Epoch 356/500 19963/19963 [==============================] - 4s - loss: 493.1445 - acc: 0.9611 - val_loss: 519.0588 - val_acc: 0.9581 Epoch 357/500 19963/19963 [==============================] - 4s - loss: 493.1316 - acc: 0.9637 - val_loss: 518.0237 - val_acc: 0.9641 Epoch 358/500 19963/19963 [==============================] - 4s - loss: 493.1233 - acc: 0.9621 - val_loss: 519.0910 - val_acc: 0.9611 Epoch 359/500 19963/19963 [==============================] - 4s - loss: 493.1112 - acc: 0.9624 - val_loss: 518.6970 - val_acc: 0.9551 Epoch 360/500 19963/19963 [==============================] - 4s - loss: 493.1054 - acc: 0.9618 - val_loss: 519.2497 - val_acc: 0.9615 Epoch 361/500 19963/19963 [==============================] - 4s - loss: 493.0960 - acc: 0.9629 - val_loss: 518.2046 - val_acc: 0.9599 Epoch 362/500 19963/19963 [==============================] - 4s - loss: 493.1001 - acc: 0.9641 - val_loss: 519.1929 - val_acc: 0.9629 Epoch 363/500 19963/19963 [==============================] - 4s - loss: 493.1127 - acc: 0.9619 - val_loss: 519.1753 - val_acc: 0.9571 Epoch 364/500 19963/19963 [==============================] - 4s - loss: 493.0781 - acc: 0.9640 - val_loss: 519.4361 - val_acc: 0.9537 Epoch 365/500 19963/19963 [==============================] - 4s - loss: 493.0829 - acc: 0.9612 - val_loss: 518.9822 - val_acc: 0.9545 Epoch 366/500 19963/19963 [==============================] - 4s - loss: 493.0743 - acc: 0.9621 - val_loss: 519.2021 - val_acc: 0.9559 Epoch 367/500
19963/19963 [==============================] - 4s - loss: 493.0873 - acc: 0.9627 - val_loss: 518.8594 - val_acc: 0.9633 Epoch 368/500 19963/19963 [==============================] - 4s - loss: 493.0917 - acc: 0.9612 - val_loss: 519.3507 - val_acc: 0.9679 Epoch 369/500 19963/19963 [==============================] - 4s - loss: 493.0565 - acc: 0.9628 - val_loss: 519.6654 - val_acc: 0.9619 Epoch 370/500 19963/19963 [==============================] - 4s - loss: 493.0800 - acc: 0.9632 - val_loss: 518.6537 - val_acc: 0.9513 Epoch 371/500 19963/19963 [==============================] - 4s - loss: 493.0608 - acc: 0.9632 - val_loss: 518.9980 - val_acc: 0.9601 Epoch 372/500 19963/19963 [==============================] - 4s - loss: 493.0509 - acc: 0.9627 - val_loss: 519.6773 - val_acc: 0.9601 Epoch 373/500 19963/19963 [==============================] - 4s - loss: 493.0621 - acc: 0.9632 - val_loss: 519.5024 - val_acc: 0.9635 Epoch 374/500 19963/19963 [==============================] - 4s - loss: 493.0554 - acc: 0.9635 - val_loss: 519.4172 - val_acc: 0.9605 Epoch 375/500 19963/19963 [==============================] - 4s - loss: 493.0354 - acc: 0.9630 - val_loss: 519.8720 - val_acc: 0.9623 Epoch 376/500 19963/19963 [==============================] - 4s - loss: 493.0386 - acc: 0.9631 - val_loss: 519.8483 - val_acc: 0.9593 Epoch 377/500 19963/19963 [==============================] - 4s - loss: 493.0244 - acc: 0.9644 - val_loss: 519.3271 - val_acc: 0.9653 Epoch 378/500 19963/19963 [==============================] - 4s - loss: 493.0449 - acc: 0.9663 - val_loss: 519.2957 - val_acc: 0.9659 Epoch 379/500 19963/19963 [==============================] - 4s - loss: 493.0052 - acc: 0.9649 - val_loss: 519.4381 - val_acc: 0.9671 Epoch 380/500 19963/19963 [==============================] - 4s - loss: 493.0066 - acc: 0.9645 - val_loss: 519.8262 - val_acc: 0.9607 Epoch 381/500 19963/19963 [==============================] - 4s - loss: 493.0130 - acc: 0.9637 - val_loss: 519.1387 - val_acc: 0.9533 Epoch 382/500 19963/19963 [==============================] - 4s - loss: 493.0246 - acc: 0.9626 - val_loss: 520.1953 - val_acc: 0.9659 Epoch 383/500 19963/19963 [==============================] - 4s - loss: 492.9986 - acc: 0.9654 - val_loss: 519.1214 - val_acc: 0.9651 Epoch 384/500 19963/19963 [==============================] - 4s - loss: 492.9961 - acc: 0.9648 - val_loss: 519.6214 - val_acc: 0.9625 Epoch 385/500 19963/19963 [==============================] - 4s - loss: 492.9758 - acc: 0.9648 - val_loss: 520.2041 - val_acc: 0.9579 Epoch 386/500 19963/19963 [==============================] - 4s - loss: 493.0016 - acc: 0.9646 - val_loss: 520.5088 - val_acc: 0.9549 Epoch 387/500 19963/19963 [==============================] - 4s - loss: 492.9880 - acc: 0.9626 - val_loss: 520.1781 - val_acc: 0.9575 Epoch 388/500 19963/19963 [==============================] - 4s - loss: 492.9782 - acc: 0.9644 - val_loss: 520.0854 - val_acc: 0.9657 Epoch 389/500 19963/19963 [==============================] - 4s - loss: 492.9786 - acc: 0.9666 - val_loss: 520.1671 - val_acc: 0.9595 Epoch 390/500 19963/19963 [==============================] - 4s - loss: 492.9702 - acc: 0.9639 - val_loss: 519.9188 - val_acc: 0.9581 Epoch 391/500 19963/19963 [==============================] - 4s - loss: 492.9679 - acc: 0.9647 - val_loss: 520.0878 - val_acc: 0.9555 Epoch 392/500 19963/19963 [==============================] - 4s - loss: 492.9669 - acc: 0.9646 - val_loss: 519.9207 - val_acc: 0.9697 Epoch 393/500 19963/19963 [==============================] - 4s - loss: 492.9442 - acc: 0.9663 - val_loss: 520.3296 - val_acc: 0.9557 Epoch 394/500 19963/19963 [==============================] - 4s - loss: 492.9592 - acc: 0.9663 - val_loss: 519.4713 - val_acc: 0.9665 Epoch 395/500 19963/19963 [==============================] - 4s - loss: 492.9488 - acc: 0.9657 - val_loss: 520.2319 - val_acc: 0.9631 Epoch 396/500 19963/19963 [==============================] - 4s - loss: 492.9536 - acc: 0.9650 - val_loss: 520.2598 - val_acc: 0.9577 Epoch 397/500 19963/19963 [==============================] - 4s - loss: 492.9394 - acc: 0.9651 - val_loss: 519.9015 - val_acc: 0.9675 Epoch 398/500 19963/19963 [==============================] - 4s - loss: 492.9263 - acc: 0.9659 - val_loss: 520.8958 - val_acc: 0.9561 Epoch 399/500 19963/19963 [==============================] - 4s - loss: 492.9520 - acc: 0.9656 - val_loss: 519.7258 - val_acc: 0.9621 Epoch 400/500 19963/19963 [==============================] - 4s - loss: 492.9417 - acc: 0.9642 - val_loss: 520.5040 - val_acc: 0.9621 Epoch 401/500 19963/19963 [==============================] - 4s - loss: 492.9210 - acc: 0.9659 - val_loss: 521.3742 - val_acc: 0.9611 Epoch 402/500 19963/19963 [==============================] - 4s - loss: 492.9143 - acc: 0.9662 - val_loss: 519.5412 - val_acc: 0.9663 Epoch 403/500 19963/19963 [==============================] - 4s - loss: 492.8986 - acc: 0.9663 - val_loss: 519.5819 - val_acc: 0.9683 Epoch 404/500 19963/19963 [==============================] - 4s - loss: 492.8857 - acc: 0.9658 - val_loss: 519.6098 - val_acc: 0.9707 Epoch 405/500 19963/19963 [==============================] - 4s - loss: 492.9039 - acc: 0.9667 - val_loss: 520.1227 - val_acc: 0.9665 Epoch 406/500 19963/19963 [==============================] - 4s - loss: 492.9012 - acc: 0.9677 - val_loss: 520.5466 - val_acc: 0.9603 Epoch 407/500 19963/19963 [==============================] - 4s - loss: 492.8854 - acc: 0.9675 - val_loss: 520.2357 - val_acc: 0.9645 Epoch 408/500 19963/19963 [==============================] - 4s - loss: 492.9059 - acc: 0.9665 - val_loss: 520.1481 - val_acc: 0.9683 Epoch 409/500 19963/19963 [==============================] - 4s - loss: 492.8748 - acc: 0.9675 - val_loss: 520.3279 - val_acc: 0.9575 Epoch 410/500 19963/19963 [==============================] - 4s - loss: 492.8888 - acc: 0.9657 - val_loss: 521.0789 - val_acc: 0.9649 Epoch 411/500 19963/19963 [==============================] - 4s - loss: 492.8651 - acc: 0.9660 - val_loss: 520.5602 - val_acc: 0.9651 Epoch 412/500 19963/19963 [==============================] - 4s - loss: 492.8836 - acc: 0.9653 - val_loss: 520.3040 - val_acc: 0.9571 Epoch 413/500 19963/19963 [==============================] - 4s - loss: 492.8699 - acc: 0.9661 - val_loss: 520.6866 - val_acc: 0.9625 Epoch 414/500 19963/19963 [==============================] - 4s - loss: 492.8969 - acc: 0.9652 - val_loss: 520.2972 - val_acc: 0.9645 Epoch 415/500 19963/19963 [==============================] - 4s - loss: 492.8878 - acc: 0.9658 - val_loss: 521.0742 - val_acc: 0.9615 Epoch 416/500 19963/19963 [==============================] - 4s - loss: 492.8555 - acc: 0.9673 - val_loss: 520.4082 - val_acc: 0.9653 Epoch 417/500 19963/19963 [==============================] - 4s - loss: 492.8856 - acc: 0.9664 - val_loss: 520.9310 - val_acc: 0.9709 Epoch 418/500 19963/19963 [==============================] - 4s - loss: 492.8615 - acc: 0.9668 - val_loss: 520.9935 - val_acc: 0.9631 Epoch 419/500 19963/19963 [==============================] - 4s - loss: 492.8460 - acc: 0.9662 - val_loss: 520.5001 - val_acc: 0.9633 Epoch 420/500 19963/19963 [==============================] - 4s - loss: 492.8443 - acc: 0.9667 - val_loss: 521.3756 - val_acc: 0.9607 Epoch 421/500 19963/19963 [==============================] - 4s - loss: 492.8476 - acc: 0.9648 - val_loss: 522.1200 - val_acc: 0.9633 Epoch 422/500 19963/19963 [==============================] - 4s - loss: 492.8465 - acc: 0.9665 - val_loss: 520.3546 - val_acc: 0.9615 Epoch 423/500 19963/19963 [==============================] - 4s - loss: 492.8388 - acc: 0.9659 - val_loss: 520.3208 - val_acc: 0.9651 Epoch 424/500 19963/19963 [==============================] - 4s - loss: 492.8524 - acc: 0.9650 - val_loss: 521.0591 - val_acc: 0.9675 Epoch 425/500 19963/19963 [==============================] - 4s - loss: 492.8153 - acc: 0.9646 - val_loss: 521.6589 - val_acc: 0.9657 Epoch 426/500 19963/19963 [==============================] - 4s - loss: 492.8244 - acc: 0.9655 - val_loss: 521.4534 - val_acc: 0.9643 Epoch 427/500 19963/19963 [==============================] - 4s - loss: 492.8100 - acc: 0.9667 - val_loss: 520.9625 - val_acc: 0.9637 Epoch 428/500
19963/19963 [==============================] - 4s - loss: 492.8158 - acc: 0.9660 - val_loss: 521.3631 - val_acc: 0.9657 Epoch 429/500 19963/19963 [==============================] - 4s - loss: 492.7940 - acc: 0.9673 - val_loss: 521.3855 - val_acc: 0.9579 Epoch 430/500 19963/19963 [==============================] - 4s - loss: 492.8105 - acc: 0.9680 - val_loss: 521.1384 - val_acc: 0.9579 Epoch 431/500 19963/19963 [==============================] - 4s - loss: 492.7924 - acc: 0.9676 - val_loss: 521.3082 - val_acc: 0.9645 Epoch 432/500 19963/19963 [==============================] - 4s - loss: 492.8049 - acc: 0.9670 - val_loss: 520.9568 - val_acc: 0.9689 Epoch 433/500 19963/19963 [==============================] - 4s - loss: 492.8227 - acc: 0.9676 - val_loss: 520.5832 - val_acc: 0.9647 Epoch 434/500 19963/19963 [==============================] - 4s - loss: 492.7795 - acc: 0.9664 - val_loss: 521.5889 - val_acc: 0.9705 Epoch 435/500 19963/19963 [==============================] - 4s - loss: 492.7713 - acc: 0.9676 - val_loss: 522.3116 - val_acc: 0.9601 Epoch 436/500 19963/19963 [==============================] - 4s - loss: 492.7715 - acc: 0.9680 - val_loss: 521.8315 - val_acc: 0.9645 Epoch 437/500 19963/19963 [==============================] - 4s - loss: 492.7555 - acc: 0.9672 - val_loss: 521.4849 - val_acc: 0.9701 Epoch 438/500 19963/19963 [==============================] - 4s - loss: 492.7608 - acc: 0.9680 - val_loss: 522.6820 - val_acc: 0.9649 Epoch 439/500 19963/19963 [==============================] - 4s - loss: 492.7764 - acc: 0.9690 - val_loss: 522.1751 - val_acc: 0.9641 Epoch 440/500 19963/19963 [==============================] - 4s - loss: 492.7604 - acc: 0.9691 - val_loss: 521.6795 - val_acc: 0.9663 Epoch 441/500 19963/19963 [==============================] - 4s - loss: 492.7596 - acc: 0.9689 - val_loss: 521.3893 - val_acc: 0.9659 Epoch 442/500 19963/19963 [==============================] - 4s - loss: 492.7838 - acc: 0.9677 - val_loss: 521.7740 - val_acc: 0.9615 Epoch 443/500 19963/19963 [==============================] - 4s - loss: 492.7367 - acc: 0.9683 - val_loss: 521.2686 - val_acc: 0.9689 Epoch 444/500 19963/19963 [==============================] - 4s - loss: 492.7491 - acc: 0.9685 - val_loss: 521.5448 - val_acc: 0.9633 Epoch 445/500 19963/19963 [==============================] - 4s - loss: 492.7301 - acc: 0.9679 - val_loss: 520.7187 - val_acc: 0.9689 Epoch 446/500 19963/19963 [==============================] - 4s - loss: 492.7486 - acc: 0.9689 - val_loss: 521.4190 - val_acc: 0.9715 Epoch 447/500 19963/19963 [==============================] - 4s - loss: 492.7361 - acc: 0.9688 - val_loss: 521.9614 - val_acc: 0.9703 Epoch 448/500 19963/19963 [==============================] - 4s - loss: 492.7605 - acc: 0.9686 - val_loss: 523.0642 - val_acc: 0.9697 Epoch 449/500 19963/19963 [==============================] - 4s - loss: 492.7571 - acc: 0.9699 - val_loss: 521.7788 - val_acc: 0.9611 Epoch 450/500 19963/19963 [==============================] - 4s - loss: 492.7147 - acc: 0.9688 - val_loss: 521.6606 - val_acc: 0.9631 Epoch 451/500 19963/19963 [==============================] - 4s - loss: 492.7104 - acc: 0.9694 - val_loss: 521.7470 - val_acc: 0.9685 Epoch 452/500 19963/19963 [==============================] - 4s - loss: 492.7251 - acc: 0.9690 - val_loss: 522.2345 - val_acc: 0.9728 Epoch 453/500 19963/19963 [==============================] - 4s - loss: 492.6935 - acc: 0.9693 - val_loss: 522.9564 - val_acc: 0.9663 Epoch 454/500 19963/19963 [==============================] - 4s - loss: 492.7229 - acc: 0.9700 - val_loss: 521.8866 - val_acc: 0.9717 Epoch 455/500 19963/19963 [==============================] - 4s - loss: 492.7183 - acc: 0.9714 - val_loss: 521.3366 - val_acc: 0.9661 Epoch 456/500 19963/19963 [==============================] - 4s - loss: 492.7180 - acc: 0.9697 - val_loss: 521.5273 - val_acc: 0.9687 Epoch 457/500 19963/19963 [==============================] - 4s - loss: 492.7086 - acc: 0.9677 - val_loss: 522.6364 - val_acc: 0.9609 Epoch 458/500 19963/19963 [==============================] - 4s - loss: 492.7111 - acc: 0.9706 - val_loss: 522.3375 - val_acc: 0.9675 Epoch 459/500 19963/19963 [==============================] - 4s - loss: 492.7032 - acc: 0.9697 - val_loss: 522.8005 - val_acc: 0.9671 Epoch 460/500 19963/19963 [==============================] - 4s - loss: 492.6893 - acc: 0.9685 - val_loss: 522.3066 - val_acc: 0.9661 Epoch 461/500 19963/19963 [==============================] - 4s - loss: 492.6880 - acc: 0.9688 - val_loss: 522.6852 - val_acc: 0.9665 Epoch 462/500 19963/19963 [==============================] - 4s - loss: 492.6817 - acc: 0.9687 - val_loss: 521.9666 - val_acc: 0.9687 Epoch 463/500 19963/19963 [==============================] - 4s - loss: 492.6995 - acc: 0.9691 - val_loss: 522.7354 - val_acc: 0.9663 Epoch 464/500 19963/19963 [==============================] - 4s - loss: 492.6805 - acc: 0.9702 - val_loss: 521.8502 - val_acc: 0.9695 Epoch 465/500 19963/19963 [==============================] - 4s - loss: 492.6601 - acc: 0.9709 - val_loss: 522.4960 - val_acc: 0.9703 Epoch 466/500 19963/19963 [==============================] - 4s - loss: 492.6794 - acc: 0.9710 - val_loss: 522.4973 - val_acc: 0.9653 Epoch 467/500 19963/19963 [==============================] - 4s - loss: 492.6659 - acc: 0.9698 - val_loss: 521.7864 - val_acc: 0.9711 Epoch 468/500 19963/19963 [==============================] - 4s - loss: 492.6674 - acc: 0.9701 - val_loss: 522.8227 - val_acc: 0.9721 Epoch 469/500 19963/19963 [==============================] - 4s - loss: 492.6550 - acc: 0.9711 - val_loss: 522.3789 - val_acc: 0.9705 Epoch 470/500 19963/19963 [==============================] - 4s - loss: 492.6781 - acc: 0.9712 - val_loss: 522.4330 - val_acc: 0.9687 Epoch 471/500 19963/19963 [==============================] - 4s - loss: 492.6779 - acc: 0.9707 - val_loss: 523.5981 - val_acc: 0.9717 Epoch 472/500 19963/19963 [==============================] - 4s - loss: 492.6333 - acc: 0.9711 - val_loss: 522.8292 - val_acc: 0.9705 Epoch 473/500 19963/19963 [==============================] - 4s - loss: 492.6372 - acc: 0.9711 - val_loss: 523.0735 - val_acc: 0.9691 Epoch 474/500 19963/19963 [==============================] - 4s - loss: 492.6338 - acc: 0.9723 - val_loss: 522.8489 - val_acc: 0.9669 Epoch 475/500 19963/19963 [==============================] - 4s - loss: 492.6367 - acc: 0.9718 - val_loss: 522.4099 - val_acc: 0.9717 Epoch 476/500 19963/19963 [==============================] - 4s - loss: 492.6085 - acc: 0.9705 - val_loss: 522.8379 - val_acc: 0.9695 Epoch 477/500 19963/19963 [==============================] - 4s - loss: 492.6392 - acc: 0.9717 - val_loss: 523.2601 - val_acc: 0.9697 Epoch 478/500 19963/19963 [==============================] - 4s - loss: 492.6270 - acc: 0.9707 - val_loss: 523.0826 - val_acc: 0.9607 Epoch 479/500 19963/19963 [==============================] - 4s - loss: 492.6469 - acc: 0.9711 - val_loss: 522.1494 - val_acc: 0.9709 Epoch 480/500 19963/19963 [==============================] - 4s - loss: 492.6032 - acc: 0.9717 - val_loss: 523.0109 - val_acc: 0.9760 Epoch 481/500 19963/19963 [==============================] - 4s - loss: 492.5939 - acc: 0.9733 - val_loss: 523.2088 - val_acc: 0.9695 Epoch 482/500 19963/19963 [==============================] - 4s - loss: 492.6055 - acc: 0.9726 - val_loss: 522.8233 - val_acc: 0.9685 Epoch 483/500 19963/19963 [==============================] - 4s - loss: 492.6285 - acc: 0.9717 - val_loss: 522.0761 - val_acc: 0.9728 Epoch 484/500 19963/19963 [==============================] - 4s - loss: 492.6015 - acc: 0.9723 - val_loss: 522.0767 - val_acc: 0.9705 Epoch 485/500 19963/19963 [==============================] - 4s - loss: 492.5926 - acc: 0.9735 - val_loss: 522.7379 - val_acc: 0.9728 Epoch 486/500 19963/19963 [==============================] - 4s - loss: 492.6344 - acc: 0.9724 - val_loss: 523.4349 - val_acc: 0.9701 Epoch 487/500 19963/19963 [==============================] - 4s - loss: 492.6152 - acc: 0.9713 - val_loss: 522.9709 - val_acc: 0.9717 Epoch 488/500 19963/19963 [==============================] - 4s - loss: 492.5943 - acc: 0.9722 - val_loss: 522.2342 - val_acc: 0.9709 Epoch 489/500
19963/19963 [==============================] - 4s - loss: 492.5984 - acc: 0.9722 - val_loss: 523.0990 - val_acc: 0.9693 Epoch 490/500 19963/19963 [==============================] - 4s - loss: 492.5692 - acc: 0.9723 - val_loss: 523.1680 - val_acc: 0.9687 Epoch 491/500 19963/19963 [==============================] - 4s - loss: 492.5955 - acc: 0.9722 - val_loss: 523.1605 - val_acc: 0.9711 Epoch 492/500 19963/19963 [==============================] - 4s - loss: 492.5674 - acc: 0.9730 - val_loss: 523.5230 - val_acc: 0.9709 Epoch 493/500 19963/19963 [==============================] - 4s - loss: 492.5607 - acc: 0.9719 - val_loss: 522.4805 - val_acc: 0.9705 Epoch 494/500 19963/19963 [==============================] - 4s - loss: 492.5855 - acc: 0.9726 - val_loss: 524.0435 - val_acc: 0.9701 Epoch 495/500 19963/19963 [==============================] - 4s - loss: 492.5620 - acc: 0.9731 - val_loss: 522.9716 - val_acc: 0.9657 Epoch 496/500 19963/19963 [==============================] - 4s - loss: 492.5732 - acc: 0.9718 - val_loss: 523.4193 - val_acc: 0.9697 Epoch 497/500 19963/19963 [==============================] - 4s - loss: 492.5679 - acc: 0.9719 - val_loss: 523.7644 - val_acc: 0.9717 Epoch 498/500 19963/19963 [==============================] - 4s - loss: 492.5677 - acc: 0.9723 - val_loss: 522.9556 - val_acc: 0.9719 Epoch 499/500 19963/19963 [==============================] - 4s - loss: 492.5498 - acc: 0.9729 - val_loss: 522.9463 - val_acc: 0.9738 Epoch 500/500 19963/19963 [==============================] - 4s - loss: 492.5578 - acc: 0.9728 - val_loss: 523.4870 - val_acc: 0.9669
In [600]:
pd.DataFrame(history.history)['acc'].plot()Out [600]:
<matplotlib.axes._subplots.AxesSubplot at 0x7febf067b710>
In [601]:
metrics = model.evaluate(X_test,y_test)
metrics = dict(zip(model.metrics_names, metrics))
metricsOut [601]:
2592/2773 [===========================>..] - ETA: 0s
{'acc': 0.96393797331410025, 'loss': 522.92700998696273}In [766]:
prices = df['sold_price_usd']
prices = prices[prices<500]
plt.hist(prices, bins=50)
plt.xlabel('$')
plt.xlim(0, 500)
plt.show()
plt.hist(np.log(df['sold_price_usd']), bins=50)
plt.xlabel('log price $')
plt.show()
In [ ]:
In [ ]:
In [843]:
# For predicting price lets uses parent generation, genes, and birth time
# We will normalize them by constants to ~0 to ~1
sire_genes = np.array([df['sire_genes']])[0]
sire_generation = df['sire_gen']/30
matron_genes = np.array([df['matron_genes']])[0]
matron_generation = df['matron_gen']/30
birth_time = df['birth_time']
birth_time = (birth_time - 1511466911)/(233588*3)
hour=df['birth_time'].apply(lambda x:datetime.datetime.fromtimestamp(x).hour)/24
weekday=df['birth_time'].apply(lambda x:datetime.datetime.fromtimestamp(x).weekday())/7
X = np.concatenate([
sire_genes,
sire_generation[:, np.newaxis],
matron_genes,
matron_generation[:, np.newaxis],
birth_time[:, np.newaxis],
hour[:, np.newaxis],
weekday[:, np.newaxis],
], 1)
# child genes
use_log_y = True
if use_log_y:
Y = np.log(np.stack(df['sold_price_usd'].values))
else:
Y = np.stack(df['sold_price_usd'].values)
X.shape, Y.shapeOut [843]:
((27727, 517), (27727,))
In [ ]:
In [844]:
# split into test and train, val (& shuffle)
import sklearn.model_selection
X_train, X_test, y_train, y_test = sklearn.model_selection.train_test_split(X,Y, random_state=42, test_size=0.1)
X_train.shape, y_train.shapeOut [844]:
((24954, 517), (24954,))
In [853]:
from sklearn.dummy import DummyRegressor
import sklearn.metrics
for strategy in ['mean', 'median']:
clf = DummyRegressor(strategy=strategy)
clf.fit(X_train, y_train)
y_pred = clf.predict(X_test)
if use_log_y:
mae = sklearn.metrics.mean_absolute_error(np.exp(y_test), np.exp(y_pred))
else:
mae = sklearn.metrics.mean_absolute_error(y_test, y_pred)
print(strategy,'mean absolute error ($)', mae)mean mean absolute error ($) 28.2072223243 median mean absolute error ($) 28.1958325447
In [ ]:
In [857]:
# Simple model with two layers
model = keras.models.Sequential()
model.add(keras.layers.InputLayer((517,)))
model.add(keras.layers.Dense(128, activation='elu'))
model.add(keras.layers.Dense(64, activation='elu'))
model.add(keras.layers.Dense(32, activation='elu'))
model.add(keras.layers.Dense(16, activation='elu'))
model.add(keras.layers.Dense(1))
model.compile(loss='mae',
optimizer=keras.optimizers.Adam(lr=1e-4),
metrics=['accuracy'])
model.summary()_________________________________________________________________ Layer (type) Output Shape Param # ================================================================= input_74 (InputLayer) (None, 517) 0 _________________________________________________________________ dense_126 (Dense) (None, 128) 66304 _________________________________________________________________ dense_127 (Dense) (None, 64) 8256 _________________________________________________________________ dense_128 (Dense) (None, 32) 2080 _________________________________________________________________ dense_129 (Dense) (None, 16) 528 _________________________________________________________________ dense_130 (Dense) (None, 1) 17 ================================================================= Total params: 77,185 Trainable params: 77,185 Non-trainable params: 0 _________________________________________________________________
In [858]:
history = model.fit(X_train, y_train, validation_split=0.2, epochs=100)Train on 19963 samples, validate on 4991 samples Epoch 1/100 19963/19963 [==============================] - 3s - loss: 1.0874 - acc: 0.0000e+00 - val_loss: 1.0069 - val_acc: 0.0000e+00 Epoch 2/100 19963/19963 [==============================] - 2s - loss: 1.0223 - acc: 0.0000e+00 - val_loss: 0.9994 - val_acc: 0.0000e+00 Epoch 3/100 19963/19963 [==============================] - 2s - loss: 1.0012 - acc: 0.0000e+00 - val_loss: 0.9899 - val_acc: 0.0000e+00 Epoch 4/100 19963/19963 [==============================] - 2s - loss: 0.9885 - acc: 0.0000e+00 - val_loss: 0.9825 - val_acc: 0.0000e+00 Epoch 5/100 19963/19963 [==============================] - 2s - loss: 0.9736 - acc: 0.0000e+00 - val_loss: 0.9748 - val_acc: 0.0000e+00 Epoch 6/100 19963/19963 [==============================] - 2s - loss: 0.9646 - acc: 0.0000e+00 - val_loss: 0.9937 - val_acc: 0.0000e+00 Epoch 7/100 4352/19963 [=====>........................] - ETA: 2s - loss: 0.9492 - acc: 0.0000e+00
[0;31m---------------------------------------------------------------------------[0m [0;31mKeyboardInterrupt[0m Traceback (most recent call last) [0;32m<ipython-input-858-f05a8e800ab9>[0m in [0;36m<module>[0;34m()[0m [0;32m----> 1[0;31m [0mhistory[0m [0;34m=[0m [0mmodel[0m[0;34m.[0m[0mfit[0m[0;34m([0m[0mX_train[0m[0;34m,[0m [0my_train[0m[0;34m,[0m [0mvalidation_split[0m[0;34m=[0m[0;36m0.2[0m[0;34m,[0m [0mepochs[0m[0;34m=[0m[0;36m100[0m[0;34m)[0m[0;34m[0m[0m [0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/keras/models.py[0m in [0;36mfit[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, **kwargs)[0m [1;32m 868[0m [0mclass_weight[0m[0;34m=[0m[0mclass_weight[0m[0;34m,[0m[0;34m[0m[0m [1;32m 869[0m [0msample_weight[0m[0;34m=[0m[0msample_weight[0m[0;34m,[0m[0;34m[0m[0m [0;32m--> 870[0;31m initial_epoch=initial_epoch) [0m[1;32m 871[0m [0;34m[0m[0m [1;32m 872[0m def evaluate(self, x, y, batch_size=32, verbose=1, [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/keras/engine/training.py[0m in [0;36mfit[0;34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, **kwargs)[0m [1;32m 1505[0m [0mval_f[0m[0;34m=[0m[0mval_f[0m[0;34m,[0m [0mval_ins[0m[0;34m=[0m[0mval_ins[0m[0;34m,[0m [0mshuffle[0m[0;34m=[0m[0mshuffle[0m[0;34m,[0m[0;34m[0m[0m [1;32m 1506[0m [0mcallback_metrics[0m[0;34m=[0m[0mcallback_metrics[0m[0;34m,[0m[0;34m[0m[0m [0;32m-> 1507[0;31m initial_epoch=initial_epoch) [0m[1;32m 1508[0m [0;34m[0m[0m [1;32m 1509[0m [0;32mdef[0m [0mevaluate[0m[0;34m([0m[0mself[0m[0;34m,[0m [0mx[0m[0;34m,[0m [0my[0m[0;34m,[0m [0mbatch_size[0m[0;34m=[0m[0;36m32[0m[0;34m,[0m [0mverbose[0m[0;34m=[0m[0;36m1[0m[0;34m,[0m [0msample_weight[0m[0;34m=[0m[0;32mNone[0m[0;34m)[0m[0;34m:[0m[0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/keras/engine/training.py[0m in [0;36m_fit_loop[0;34m(self, f, ins, out_labels, batch_size, epochs, verbose, callbacks, val_f, val_ins, shuffle, callback_metrics, initial_epoch)[0m [1;32m 1154[0m [0mbatch_logs[0m[0;34m[[0m[0;34m'size'[0m[0;34m][0m [0;34m=[0m [0mlen[0m[0;34m([0m[0mbatch_ids[0m[0;34m)[0m[0;34m[0m[0m [1;32m 1155[0m [0mcallbacks[0m[0;34m.[0m[0mon_batch_begin[0m[0;34m([0m[0mbatch_index[0m[0;34m,[0m [0mbatch_logs[0m[0;34m)[0m[0;34m[0m[0m [0;32m-> 1156[0;31m [0mouts[0m [0;34m=[0m [0mf[0m[0;34m([0m[0mins_batch[0m[0;34m)[0m[0;34m[0m[0m [0m[1;32m 1157[0m [0;32mif[0m [0;32mnot[0m [0misinstance[0m[0;34m([0m[0mouts[0m[0;34m,[0m [0mlist[0m[0;34m)[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1158[0m [0mouts[0m [0;34m=[0m [0;34m[[0m[0mouts[0m[0;34m][0m[0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/keras/backend/tensorflow_backend.py[0m in [0;36m__call__[0;34m(self, inputs)[0m [1;32m 2267[0m updated = session.run(self.outputs + [self.updates_op], [1;32m 2268[0m [0mfeed_dict[0m[0;34m=[0m[0mfeed_dict[0m[0;34m,[0m[0;34m[0m[0m [0;32m-> 2269[0;31m **self.session_kwargs) [0m[1;32m 2270[0m [0;32mreturn[0m [0mupdated[0m[0;34m[[0m[0;34m:[0m[0mlen[0m[0;34m([0m[0mself[0m[0;34m.[0m[0moutputs[0m[0;34m)[0m[0;34m][0m[0;34m[0m[0m [1;32m 2271[0m [0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/tensorflow/python/client/session.py[0m in [0;36mrun[0;34m(self, fetches, feed_dict, options, run_metadata)[0m [1;32m 893[0m [0;32mtry[0m[0;34m:[0m[0;34m[0m[0m [1;32m 894[0m result = self._run(None, fetches, feed_dict, options_ptr, [0;32m--> 895[0;31m run_metadata_ptr) [0m[1;32m 896[0m [0;32mif[0m [0mrun_metadata[0m[0;34m:[0m[0;34m[0m[0m [1;32m 897[0m [0mproto_data[0m [0;34m=[0m [0mtf_session[0m[0;34m.[0m[0mTF_GetBuffer[0m[0;34m([0m[0mrun_metadata_ptr[0m[0;34m)[0m[0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/tensorflow/python/client/session.py[0m in [0;36m_run[0;34m(self, handle, fetches, feed_dict, options, run_metadata)[0m [1;32m 1122[0m [0;32mif[0m [0mfinal_fetches[0m [0;32mor[0m [0mfinal_targets[0m [0;32mor[0m [0;34m([0m[0mhandle[0m [0;32mand[0m [0mfeed_dict_tensor[0m[0;34m)[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1123[0m results = self._do_run(handle, final_targets, final_fetches, [0;32m-> 1124[0;31m feed_dict_tensor, options, run_metadata) [0m[1;32m 1125[0m [0;32melse[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1126[0m [0mresults[0m [0;34m=[0m [0;34m[[0m[0;34m][0m[0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/tensorflow/python/client/session.py[0m in [0;36m_do_run[0;34m(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)[0m [1;32m 1319[0m [0;32mif[0m [0mhandle[0m [0;32mis[0m [0;32mNone[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1320[0m return self._do_call(_run_fn, self._session, feeds, fetches, targets, [0;32m-> 1321[0;31m options, run_metadata) [0m[1;32m 1322[0m [0;32melse[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1323[0m [0;32mreturn[0m [0mself[0m[0;34m.[0m[0m_do_call[0m[0;34m([0m[0m_prun_fn[0m[0;34m,[0m [0mself[0m[0;34m.[0m[0m_session[0m[0;34m,[0m [0mhandle[0m[0;34m,[0m [0mfeeds[0m[0;34m,[0m [0mfetches[0m[0;34m)[0m[0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/tensorflow/python/client/session.py[0m in [0;36m_do_call[0;34m(self, fn, *args)[0m [1;32m 1325[0m [0;32mdef[0m [0m_do_call[0m[0;34m([0m[0mself[0m[0;34m,[0m [0mfn[0m[0;34m,[0m [0;34m*[0m[0margs[0m[0;34m)[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1326[0m [0;32mtry[0m[0;34m:[0m[0;34m[0m[0m [0;32m-> 1327[0;31m [0;32mreturn[0m [0mfn[0m[0;34m([0m[0;34m*[0m[0margs[0m[0;34m)[0m[0;34m[0m[0m [0m[1;32m 1328[0m [0;32mexcept[0m [0merrors[0m[0;34m.[0m[0mOpError[0m [0;32mas[0m [0me[0m[0;34m:[0m[0;34m[0m[0m [1;32m 1329[0m [0mmessage[0m [0;34m=[0m [0mcompat[0m[0;34m.[0m[0mas_text[0m[0;34m([0m[0me[0m[0;34m.[0m[0mmessage[0m[0;34m)[0m[0;34m[0m[0m [0;32m~/.pyenv/versions/3.5.3/envs/jupyter3/lib/python3.5/site-packages/tensorflow/python/client/session.py[0m in [0;36m_run_fn[0;34m(session, feed_dict, fetch_list, target_list, options, run_metadata)[0m [1;32m 1304[0m return tf_session.TF_Run(session, options, [1;32m 1305[0m [0mfeed_dict[0m[0;34m,[0m [0mfetch_list[0m[0;34m,[0m [0mtarget_list[0m[0;34m,[0m[0;34m[0m[0m [0;32m-> 1306[0;31m status, run_metadata) [0m[1;32m 1307[0m [0;34m[0m[0m [1;32m 1308[0m [0;32mdef[0m [0m_prun_fn[0m[0;34m([0m[0msession[0m[0;34m,[0m [0mhandle[0m[0;34m,[0m [0mfeed_dict[0m[0;34m,[0m [0mfetch_list[0m[0;34m)[0m[0;34m:[0m[0;34m[0m[0m [0;31mKeyboardInterrupt[0m:
In [ ]:
pd.DataFrame(history.history)['acc'].plot()In [ ]:
In [859]:
metrics = model.evaluate(X_test,y_test)
metrics = dict(zip(model.metrics_names, metrics))
metricsOut [859]:
2304/2773 [=======================>......] - ETA: 0s
{'acc': 0.0, 'loss': 0.95822394375675557}In [860]:
# I got ~26 which is not great
y_pred = model.predict(X_test)
mae = sklearn.metrics.mean_absolute_error(np.exp(y_test), np.exp(y_pred))
maeOut [860]:
26.219251685788045
In [ ]: