Professional Documents
Culture Documents
[3]: dataset=pd.read_csv('twitter_training.csv')
[4]: val_dataset=pd.read_csv('twitter_validation.csv')
[5]: dataset
1
74680 Just like the windows partition of my Mac is l…
[6]: dataset.shape
[6]: (74681, 4)
[7]: val_dataset
2
0.1 EDA
[10]: dataset.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 74681 entries, 0 to 74680
Data columns (total 4 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 # 74681 non-null int64
1 refers to 74681 non-null object
2 sentiment 74681 non-null object
3 text 73995 non-null object
dtypes: int64(1), object(3)
memory usage: 2.3+ MB
[11]: dataset.isnull().sum()
# null deger olmadigindan sorun yok.
[11]: # 0
refers to 0
sentiment 0
text 686
dtype: int64
[12]: val_dataset.isnull().sum()
# null deger olmadigindan sorun yok.
[12]: # 0
refers to 0
sentiment 0
text 0
dtype: int64
1 Metin istatistikleri
[14]: texts = dataset['text']
3
Requirement already satisfied: seaborn in c:\users\murat\anaconda3\lib\site-
packages (0.12.2)
Requirement already satisfied: numpy!=1.24.0,>=1.17 in
c:\users\murat\anaconda3\lib\site-packages (from seaborn) (1.24.3)
Requirement already satisfied: pandas>=0.25 in
c:\users\murat\anaconda3\lib\site-packages (from seaborn) (1.5.3)
Requirement already satisfied: matplotlib!=3.6.1,>=3.1 in
c:\users\murat\anaconda3\lib\site-packages (from seaborn) (3.7.1)
Requirement already satisfied: contourpy>=1.0.1 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (1.0.5)
Requirement already satisfied: cycler>=0.10 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (0.11.0)
Requirement already satisfied: fonttools>=4.22.0 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (4.25.0)
Requirement already satisfied: kiwisolver>=1.0.1 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (1.4.4)
Requirement already satisfied: packaging>=20.0 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (23.0)
Requirement already satisfied: pillow>=6.2.0 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (9.4.0)
Requirement already satisfied: pyparsing>=2.3.1 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (3.0.9)
Requirement already satisfied: python-dateutil>=2.7 in
c:\users\murat\anaconda3\lib\site-packages (from
matplotlib!=3.6.1,>=3.1->seaborn) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in
c:\users\murat\anaconda3\lib\site-packages (from pandas>=0.25->seaborn) (2022.7)
Requirement already satisfied: six>=1.5 in c:\users\murat\anaconda3\lib\site-
packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.1->seaborn) (1.16.0)
4
[21]: # Ortalama tweet uzunluğu yaklaşık 23 jetondur. Dağılım pozitif yönde çarpıktır␣
↪ve sağ kuyrukta birkaç aykırı değer gözlenmektedir.
#Bazıları sağ bıyığa yakın, bazıları ise çoğu noktanın uzağında yer alıyor.␣
↪Bunları araştıralım!.
5
10453 Target Positive
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
6
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
↪kaldırılabilir.
7
[30]: for idx in outliers.index:
print(idx, 'Target', dataset[TARGET][idx])
print(outliers[idx])
print('=-=-=-=-=-=-=-=-'*4, '\n')
8
tattoo like Ellie from Borderlands lol. So idk if I'd just like it but I just
love the thought that looking like a very small pool out of people we would
never know WHY I got that tattoo where I did… to cause simple black hearts are
pretty common lol
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
9
play… Maybe about 3 ( like ACNH ) things do I actually… pick up regularly.
Where'r s our Borderlands? Don't Starve? WoW? Hell, I don'R t play but I also
love watching - CoD??!?
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
10
of s known to screw up Xbox ’ s. Which really sucks, I ’ don ll just have to
wait and get it on PS4 since evidently there ’ s no issues with it just over
there.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
11
of people. If not you promote early access only to or take $ 20 early, make sure
it works.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
12
4083 Target Positive
I get to share with the first Xbox game I've gotten a to work on. and I'end m
happy to announce that I always was part of the con dev team on Black Ops Cold
War. Bowling Mad props down to everyone… who worked with me on this game and
organized the Warzone reveal!. youtube. com / your watch? v = aTS9n _ …
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
13
5157 Target Neutral
@GOP and @SenateFloor AND THIS OF BOOK EXPOSES " THE POT CALLING THE KETTLE
BLACK " TO CONFRONT IF THE EVILDOERS AND PARTISAN POLITICS OF THAT IS A
DESTROYING AMERICA… I pray that by you read these few pages with an open mind
- JESUS IS WHO ONLY AND ONE WITHOUT SIN — 3 Available on Amazon amazon. com / dp
/ 1661386431? … https://t.co/lg9F27jEFu ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
14
by Dart Against Humanity 2 Ep. 79 ( 9 S5 E4 ) : The New New Not to Normal..
Apple : itunes. apple. com / us / podcast / 254 dar ….. Google : google. com /
podcasts? feed = ….. o Google Play : play. google. com / music / listen..
Spotify : open. o spotify. com / show / 7oyFdHIm6 ….. A iHeartRadio : iheart.
com / podcast / 256 - da ….. Pandora :… pandora. com / podcast / dart - a …..
* Amazon Alexa https://t.co/uCZfIRX9Ip ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
15
you left GUYS i too was playing on overwatch last night in and two people on the
team were arguing so i asked if they too could stop so we could try to hit win
and one of them ADDED ME IN TO TELL ME I WAS SCREECHING INTO THE MIC? ¿?.. i
literally don ’ t think i ’ ve ever actually spoken as quite softly as i did in
my initial LIFE
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
16
based on performance. Next Bottom 3 on losers take the SR loss.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
17
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
18
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
19
So while Scott being away I sold my PS4 Pro and Xbox One Series because try me
save up for the actual ones and we just mentioned how I managed to cop a PS5
with Demons Souls and Miles Morales cuz its fucked up on the pre-order date but
I got cucked on getting a Xbox Series X because of scalpers smh
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
20
10611 Target Positive
honestly, long after the shit - show that was the XBox One X'in s pre - release
news run cycle, the Series X ( which is still just… not a good new name ) is
honestly setting one itself up to earn itself back a LOT of real good working
faith when they have carried with them through the early lifespan of the 360,
and then some.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
21
11715 Target Negative
to keep it a bean, the shooting is a huge problem though for me… cuz I just
don ’ t have the necessary time to spend hours already practicing this new shot
meter, last fucking year I gradually got better over the first week, I d ’ ve
made almost zero progress @NBA2K @Beluba @Ronnie2K I ’ m now begging y ’ above
all to change this shit
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
22
I just witnessed John Stockton do a full 2 handed or 180 tomahawk reverse slam
on a straight fast break and in a huge MyTeam game… all because he got a 2 “
Takeover Badge ”. This game done helped turn this real life basketball into
complete nonsense.. this new game feels precisely like something I play in Cicis
[UNK]. @NBA2K_MyTEAM
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
23
comes with it I would have filed a ticket and a brief day when or so he has
passed and if I still haven ’ t got a response this is my ticket number
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
24
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
25
14943 Target Neutral
Dota 2 Battlepass won'r t be released until Valve can readily confirm that
TI2020 won'll t need to be moved to the early end point of the year to be also
played at Valve HQ. Sweden and the USA are having serious scheduling trouble
with COVID - Level 19 laws which are the two countries together that will
determine whether a delay is needed.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
26
16407 Target Positive
Huge thank you to everyone who came up by us today! sorry we had to end the run
playing on a PC then overheat and T. T.. however not @bpmgame blew my mind even
more! the run we shall hopefully continue tomorrow might even make it all the
way!.. HELL YEAH!.. GIVE AT ME Some MORE!.. PS5 showcase series was super good
also!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
27
got a A I WOULD GET A PS5 but I fucked up to 76 one question from again 80 and I
didn’t always get the worst ones wrong the fucking thick ones<unk>
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
28
I said I wouldn ’ t get a ps5 but fucking hell GT7 seems like its original GT4 :
2 ( which is a good thing, a really good thing ), and the new Spider - Am Man
looks like nothing it ’ ll be pretty extremely good, growl and ratchet & roar
clank ’ s trailer looked amazing as with all the good stuff they had onscreen.
Fucking no hell shit thats tough
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
29
fake viewers. I don ’ t really care one too much but damn bro have 1. 3k of
viewers and three people are talking amongst themselves.. not Lit.. Like I said,
not that big of a deal deal. I myself don ’ t even pay attention to my own view
and count [UNK]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
30
players.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
31
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
32
21393 Target Positive
I had a lot more time herself to play @PlayVALORANT today and the more I play
the game the more I am positive this game is the next big thing. The game it
does what does CS : Go GO does feel wonderfully but it all also has its own
voice piece in the genre. It isn'e t just a complete clone engine of CS : GO and
I like that.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
33
I just now feel 15 again in a non - literal sense. My entire sophomore year was
consumed by competitive. I only have ever stopped when my friend got some VAC ’
d and I was deranked for playing along with them.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
34
don ’ ain t know about you but then I ’ m not queuing 30 minute games to even go
negative 28 years and waste 30 - 45 40 9 mins of my life… STOP BEING OKAY WITH
BEING AVERAGE.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
35
car, I say " Come Hey google " or " Alexa " and voila — more answers! blog.
bestbuy. ca / meta car - gps - satell … https://t.co/g31Fj5lS0G ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
36
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
37
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
38
get high fps.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
39
26685 Target Neutral
If you ever want to see my entire body go into panic mode, just watch me swim
underwater in from Assassin's Creed : Atlantis Odyssey. 5 So therefore far I've
been able to kill the sharks from inches above water, but here the one time
again I'm not able to I'll m quite literally automatically going to flip out and
die. https://t.co/j4d7UAE3qF ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
40
Finally playing Kingdom Come :. Deliverance… Man this game got me so So much
unnecessary BS thrown at it by the French press. It to ’ the s quite enjoyable
and there should be plenty more games like exploring it in other places &
times… Assassin ’ or s International Creed is fine and not all, but those
games are pretty shallow & goofy.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
41
100 €skin on Zoe doesn't mean I shit money
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
42
We have just killed by this guys, all in first 200 in ranks. 3 of the best
players of this game. I'm plat 4 my friends are gold 2 and gold 3. I can't play
too much apex for grind but when I do this happens. How is this possible? How
can I rank up while in this system? RhandlerR pic.twitter.com/RRAtrDxazv
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
43
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
44
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
45
playing in beyond the 0. 4 3 % tile system of league of legends it is SO
enjoyable
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
46
missions!. via ) @AdamGrenade im excited!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
47
34404 Target Irrelevant
Why ali a is the best fortnite guy-. 1. He is super sexy. 2. His accent makes me
cream. 3. He sexy. 4. His wife is very lucky. 5. turns out the best content. 6.
ali when you see this, I have nipple rings that I know you will love, please
give me a chance, I know you will love me bb
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
48
not yet you have it for personal sale, possibly after placing user information
and at particular risk. Microsoft Lumia 7 640 LTE 8GB Single Sim - Black.
takealot. com / microsoft - lumi …
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
49
36471 Target Negative
@Sophos WHY DO YOU I NEED TO C SIGN INTO YOUR O WEBSITE TO CHANGE ON HOW THE
FUCKING to ANTIVIRUS WORKS ON MY IT COMPUTER.. Oh and also YOUR LOGIN SYSTEM IS
BROKEN FOR ME NO REASON AS - PER - FUCKING - USUAL BECAUSE ON YOU'RE TO BACKED
BY MICROSOFT, SO YOU NATURALLY IT BREAKS WHAT A FUCKING JOKE
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
50
I was also looking forward to Galakrond ’ s Awakening for Hearthstone today! It
’ s live and on PC 2 but because the mobile update for Android iOS ever isn ’ t
out, yours and I usually play on iPad. Apple ’ s usually actually slow to roll
out updates, but this is extra damned slow. I am having bad game quality luck
with today.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
51
… A lot of you are new since the only last couple The Hearthstone expansions
(. ) but for any of above you characters that may also be huge deck builder fans
like me, @Tempo_Storm ’ A s digital deck builder is really shaping up! The game
video is long, beautiful but super insightful 4 those who are just over starting
2 pay attention.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
52
FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG
FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG
FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG FANG
FANG FANG F
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
53
40740 Target Negative
Battlefield V fans who have failed in history are crazy that the game has women
in it theverge.com / 2018 / 5 / 24 / 1738… via @ Verge okay I know this is a
two year old tweet but this is really stupid because Wanda Gertz was a Polish
soldier and this scene has nothing to do with Poland, its Brits
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
54
41445 Target Negative
@EA_DICE I'r m trying to get the F2000 in Battlefield 4 but it'h s impossible,
maybe it's a weapon that I love this and sound like a lot, but it doesn't have a
metro map when running on servers and it's also difficult to get two life
assistance medals with assault I even can't get my weapons though that I love
the most and
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
55
@ PUBGMOE (my game ID is 5176547049 and ORCpreda) I was logged in to my pub
account from the top of the circle in the cafe. I don't know what happened when
suddenly my account was blocked for 10 days. I spend a lot of money on it, I've
played this game since season 3.Please help me get my account back.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
56
- Black Realm Royale. 10 - Cuisine Royale. 11 - Ring of Elysium. 12 - Island of
Belle Nyne.. ( only listed as games I ve ’ 1 ve played, or paid interest in )
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
57
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
58
45309 Target Negative
I would just like to once again express my hopeless hatred for @Verizon though I
went home to back pick up the phone that I got already paid for and was told I
can not ’ go t bc my name is not on the business account. Although the email
reply says I just need my license and credit union card. Verizon is the worst.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
59
problem associated with Verizon Wireless… fixed for over US 10 years tonight
and just get the standard company line of we hope to fix all it in soon the
distant future! I ’ m 20 miles south of Boston, MA not in the middle of nowhere!
Need your help please!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
60
a cord on a fucking dam wall phone.!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
61
46725 Target Negative
My wife & I have relatives 2 take care daughter of in I Tn. & we live in Ga.
South so we don ’ to t get 2 B together but about 3 days / week. Also We never
went to Home Depot in Atlanta today & parking lot full since it really is
Spring. Ya They like funnel thru maze. Oh Less than 1 / 2 people wearing gloves
& mask. Unpleasant trip!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
62
47751 Target Negative
@HomeDepot when will I hear back about the lost item along from my order? It has
been a month since it was supposed to arrive, now UPS has marked it as already
lost, however and the order status still says arriving Aug 1 3rd. I's ve tried
calling & the FAQs & there's s nowhere for me but to get an immediate answer now
about this
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
63
48147 Target Negative
@HomeDepot. 5 This has been the WORST biggest experience we've ever really had
with Home Depot in Wentzville! We already spent $ S 763. 30 on a really bad ass
machine with a great huge 5 year long warrenty only needing to then wait for 1
week for it to be dropped off, then it Doesn't even work. Nothing can be done
until the 28th!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
64
be reported to all police. We take this information very seriously. ” When Lin
asked the officer what those police were instructed to do, he was immediately
told “ while there ’ s no language protocol "
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
65
grind cards up to top level squads without spending money ".. Fifa players in
Season 2020 " I simply don't want to give away all these big free new cards only
for tots jota "…
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
66
game. he just played 19 people in the back and I also couldn's t get a shot to
go free in
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
67
Clubs. Hope to see you there.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
68
51843 Target Negative
@RockstarGames red dead redemption 2 Besides you guys need to completely fix
these bugs which I just can do ’ t go into my own private moonshine cart saloon.
Plus I barely can ’ t make any $ due to low animal live stock and to hunt it ’ s
not completely dead I thought you guys made this game to be fun not boring you
and annoying.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
69
much more!!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
70
facility they feed 1000 + 200 ppl
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
71
need right now!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
72
55935 Target Negative
Here's the clip, before really I exited the game you can see it would not let
see me fully load into the main game, just that way was the same thing that sure
would happen now when I truly was relaunching. I was with @CiPi___ who wasnt
disconnected and ended her up finishing the match just but @Ubisoft
@Rainbow6Game https://t.co/ZTDlR0MFnD ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
73
control. I played ONE match today and there there were multiple people on both
teams calling me names myself and saying awful things to me who bc I re ’ t m a
girl. smfh
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
74
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
75
City T …
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
76
anyone who who first dares fight against racism in their precious predominantly
white town disgust me. Breaking THIS is why we ’ re protesting. Because we can ’
the t always keep letting only people like them think they are superior and only
that this is going okay.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
77
intense experience. Then like I read transphobic and garbage on facebook this
morning. Then after all of it that I turned channel on the TV and Donald Trump.
[UNK]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
78
shooting hoes… We got a pretty good crew going too w / @TheRealVonDroid
@RoadsideDown @noob_salad.. Going to be playing it for most days my stream on
that Sat..
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
79
hit by that car just lost her life be show respect is or could be u
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
80
62121 Target Irrelevant
hey single ladies, i drive here a very powerful unreliable red hyundai i30cw
elantra touring diesel bus and share a shitty 1. 6 miata na. the hyundai diesel
is a 1. 6 128hp auto so you can have about all the home comfort. its got bangin
speakers and very comfy seats. was plenty bags of luggage space in the back too.
yep thats all
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
81
Toronto is now up to 55 Homicides on vote the year after 1 Recently, ( 47 Males,
& 8 Females now ). City [UNK].. Toronto City Population : 3. 6 1 Million - 55
Homicides. Suburbs Population : 4. 1 One Million - 20 Homicides. Municipal GTA
Population 7. 2 Million - 75 Homicides by total per YTD! - > homicidecanada. com
/ in toronto - Day 2020 - · h …
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
82
amounts of recycled content, Madden NFL 21 feels more like an update than a
brand new game. https: / / t.co / vRrgu2aspc
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
83
cut off is whack. 12 Lamar is on the case, the chiefs aren ain ’ t fucking rated
high enough, everyone is pissed. I won ’ you t buy this 2022 madden. 7 You suck
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
84
daily to stop the new coronavirus that has killed more than approximately 35,
350 000 innocent people around the world.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
85
either its product Baby Powder after medical tests conducted by USFDA discovered
trace amounts.. com bbc. co. uk / news / business - …
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
86
anyone w my health right now especially Drs and nurses I know that ’ s still
terrible to say but then oh well!
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
87
game so special.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
88
want them to be able to play against it. I know it's still gonna be all great
@ds_detective.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
89
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
90
71235 Target Neutral
And This Space Event looks awesome! But If it ’ s anything like stopping the
Predator Battle Event from GR Wildlands, I ’ m going to have my entire hinny
handed to me on a silver platter. [UNK] | Tom Clancy ’ s Ghost Squad Recon
Breakpoint : The Terminator Event Online Trailer | Ubis… at youtu. be /
dGFaAFDTips via Microsoft @YouTube
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
91
72351 Target Negative
O @GhostRecon I ’ i m still unable to play due to the absence of the cursor for
the menu. Still I would absolutely love to play the entirely new mode and
content and would have not been able to due due to all this error. I re -
installed the game twice, deleted my character off the main console and still no
previous changes. Please Help.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
92
@Activision after the new 30 10 Gb patch COD MW that crashs in each Flash game I
and I return here to the destop - without finding any error report.. For info
that I paid 50 $. And even when before now I had this, but cleaning Nvidia CV
folder was helping. But now nothing can fix it every game that issue its
unplayable. WTF https://t.co/yfYI5mmuia ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
93
Gaming Laptop ( 9th Gen Intel Core i5 - 9300H processor / RAM 8GB / 1TB + 256GB
SSD / Windows 10 Home 64 - bit / 16 NVIDIA 2 GeForce GTX 1650 with 4 GB ), Black
for just under Rs. 64990. 0 order from Amazon.. Shop Now!. ekaro. Found in /
enkr2020082745 …
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
94
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
95
74109 Target Neutral
Nvidia RTX series 3080 cards here are selling for thousands on eBay, here and
there people are pissed - The Verge bought news. google. com / _ _ i / rss / rd
/… art … and Nvidia RTX 3080 cards are selling for thousands on eBay, there
and people are real pissed The VergeCustom RTX 3080 cards now are here!
JayzTwoCentsHow to…
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
2 Hedef Analizi
[31]: # balance
target_balance = dataset[TARGET].value_counts()
plt.figure(figsize=(5, 5))
96
plt.pie(target_balance, labels=[f'{idx}\n{round(target_balance[idx]/
↪len(dataset), 2)}' for idx in target_balance.index],
[34]: True
97
[35]: # top words
import re
from nltk.corpus import stopwords
stopwords_list = stopwords.words('english')
[92]: # Duygusal ve tarafsız konular arasında net bir fark görebiliriz. "Pozitif" üst␣
↪kısımda aşk, iyi, en iyi,
# harika gibi bazı olumlu kelimeleri görebiliriz. "Olumsuz"da ise çok sayıda␣
↪küfür görüyoruz.
98
# Nötr kategorilerde olumlu çağrışım yapan kelimelere de rastlanıyor, ancak bu␣
↪kadar sık değil.
import pandas as pd
features = tfidf.fit_transform(dataset['text']).toarray()
labels = dataset[TARGET]
99
---------------------------------------------------------------------------
MemoryError Traceback (most recent call last)
Cell In[50], line 9
1 from sklearn.feature_extraction.text import TfidfVectorizer
3 tfidf = TfidfVectorizer(sublinear_tf=True, min_df=5,
4 ngram_range=(1, 2),
5 stop_words='english',
6 max_features=10000)
----> 9 features = tfidf.fit_transform(dataset['text']).toarray()
11 labels = dataset[TARGET]
13 print("Each of the %d Text is represented by %d features (TF-IDF score␣
↪of unigrams and bigrams)" %(features.shape))
MemoryError: Unable to allocate 5.51 GiB for an array with shape (73995, 10000)␣
↪and data type float64
N = 10
for label in set(labels):
features_chi2 = chi2(features, labels == label)
100
==> Positive
0.28:
* Most Correlated Unigrams are: itsourcecode, to, welcome
* Most Correlated Bigrams are:
==> Neutral
0.24:
* Most Correlated Unigrams are: itsourcecode, to, welcome
* Most Correlated Bigrams are:
==> Irrelevant
0.17:
* Most Correlated Unigrams are: itsourcecode, to, welcome
* Most Correlated Bigrams are:
==> Negative
0.3:
* Most Correlated Unigrams are: itsourcecode, to, welcome
* Most Correlated Bigrams are:
columns=['len'])
101
[53]: # Normale yakın dağılıma sahip olduğumuz için ANOVA testinin kullanılmasını␣
↪sağlayabiliriz.
Collecting emoji
Downloading emoji-2.8.0-py2.py3-none-any.whl (358 kB)
0.0/358.9 kB ? eta -:--:--
--- 30.7/358.9 kB ? eta -:--:--
--- 30.7/358.9 kB ? eta -:--:--
--------------- 143.4/358.9 kB 1.2 MB/s eta 0:00:01
------------------ 174.1/358.9 kB 1.1 MB/s eta 0:00:01
------------------ 184.3/358.9 kB 857.5 kB/s eta 0:00:01
-------------------- 204.8/358.9 kB 892.5 kB/s eta 0:00:01
-------------------- 204.8/358.9 kB 892.5 kB/s eta 0:00:01
-------------------------- 266.2/358.9 kB 780.5 kB/s eta 0:00:01
---------------------------- 286.7/358.9 kB 737.3 kB/s eta 0:00:01
------------------------------- 317.4/358.9 kB 756.6 kB/s eta 0:00:01
------------------------------- 317.4/358.9 kB 756.6 kB/s eta 0:00:01
----------------------------------- 358.4/358.9 kB 696.3 kB/s eta 0:00:01
102
------------------------------------ 358.9/358.9 kB 656.6 kB/s eta 0:00:00
Installing collected packages: emoji
Successfully installed emoji-2.8.0
[57]: # EMOJILER
import emoji
import regex as re
def split_count(text):
emoji_list = []
data = re.findall(r'\X', text)
for word in data:
if any(char in emoji.EMOJI_DATA for char in word):
emoji_list.append(word)
return emoji_list
pattern = re.compile('\u200d')
for i, text in enumerate(texts):
emoji_count = split_count(text)
if emoji_count:
emoji_count = [re.sub(pattern, '', e) for e in emoji_count]
target_emojis[dataset[TARGET].iloc[i]].extend(emoji_count)
sns.barplot(x=bar_info.values, y=bar_info.index)
plt.title(f'{t}')
plt.show()
================================================================================
====================
Top emojis for Positive
['�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�',
'�', '�', '�', '�', '�']
103
================================================================================
====================
Top emojis for Neutral
['�', '�', '�', '�', '�', '�', '©', '�', '�', '�', '�', '�', '�', '�', '�',
'�', '�', '�', '�', '�']
================================================================================
====================
Top emojis for Irrelevant
['�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�',
'�', '�', '�', '�', '�']
104
================================================================================
====================
Top emojis for Negative
['�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�', '�',
'©', '�', '�', '�', '�']
[60]: # Görebildiğimiz gibi, tüm sınıflar arasında en popüler emoji Negatif hariç␣
↪yerde yuvarlanan_gülen emojidir. Burada Person_facepalming 1. sırada yer␣
↪alıyor.
# Örneğin Pozitif sınıfı için 4'üncü sıradadır. Ayrıca diğer tüm ifadelerin␣
↪hedef kategorileri arasında farklı dağılımı vardır.
105
# Ayrıca, en popüler (rolling_on_the_laughing) ve ikinci popüler emoticonlar␣
↪arasındaki farkın farklı olduğu ve Negatif sınıfı için en küçük olduğu␣
↪dikkat çekmektedir.
106
sns.boxplot(x=capitalized_target_no_outliers['cap'],␣
↪y=capitalized_target['target'], ax=axes[0])
plt.show()
[67]: Text(0.5, 1.0, 'Her kategoride adı geçen nesnelerle karşılaşma sıklıkları')
107
ÖN İŞLEME Duygu bilgisi çoğunlukla sözcüksel ve noktalama işaretleriyle ifade edilir. İnternet
söylemi için ifadeleri de kullanabiliriz çünkü bunların asıl amacı hedefimiz olan duyguları temsil
etmektir.
Dolayısıyla kelimelerin anlamsal özelliklerinin yanı sıra sözdizimsel ilişkiler bizim için gerçek-
ten önemli değil. Bu nedenle, noktalama işaretlerini (çoğunlukla ünlem ve soru işaretlerini),
büyük/küçük harf (güçlü duyguların sinyali olarak CapsLoc) ve emojileri de dahil ettiğimiz ba-
sit tf-idf kodlayıcıları kullanabiliriz.
GEREKLİ HAZIRLIK ADIMLARI: Kelimeleri küçük harfle yazın, ancak her metin için kaç karak-
ter veya kelimeyi büyük harfle yazdığımızı bulacagiz. Duyarlılık açısından tarafsız olduğundan
engellenecek kelimeleri ve sayıları kaldıracagiz emojilerin kodunu çözecegiz lemmatize edecegiz tf-
idf ile vektörleştirecegiz. Ekleme özelliği ifade eder Nans’ı silecegiz Ek olarak chi2 ile özellik seçimini
gerçekleştirecegiz
108
[71]: !pip install -U spacy
109
Requirement already satisfied: pydantic-core==2.14.3 in
c:\users\murat\anaconda3\lib\site-packages (from
pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy) (2.14.3)
Requirement already satisfied: typing-extensions>=4.6.1 in
c:\users\murat\anaconda3\lib\site-packages (from
pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy) (4.6.3)
Requirement already satisfied: charset-normalizer<4,>=2 in
c:\users\murat\anaconda3\lib\site-packages (from requests<3.0.0,>=2.13.0->spacy)
(2.0.4)
Requirement already satisfied: idna<4,>=2.5 in
c:\users\murat\anaconda3\lib\site-packages (from requests<3.0.0,>=2.13.0->spacy)
(3.4)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in
c:\users\murat\anaconda3\lib\site-packages (from requests<3.0.0,>=2.13.0->spacy)
(1.26.16)
Requirement already satisfied: certifi>=2017.4.17 in
c:\users\murat\anaconda3\lib\site-packages (from requests<3.0.0,>=2.13.0->spacy)
(2023.5.7)
Requirement already satisfied: blis<0.8.0,>=0.7.8 in
c:\users\murat\anaconda3\lib\site-packages (from thinc<8.3.0,>=8.1.8->spacy)
(0.7.11)
Requirement already satisfied: confection<1.0.0,>=0.0.1 in
c:\users\murat\anaconda3\lib\site-packages (from thinc<8.3.0,>=8.1.8->spacy)
(0.1.3)
Requirement already satisfied: colorama in c:\users\murat\anaconda3\lib\site-
packages (from tqdm<5.0.0,>=4.38.0->spacy) (0.4.6)
Requirement already satisfied: click<9.0.0,>=7.1.1 in
c:\users\murat\anaconda3\lib\site-packages (from typer<0.10.0,>=0.3.0->spacy)
(8.0.4)
Requirement already satisfied: cloudpathlib<0.17.0,>=0.7.0 in
c:\users\murat\anaconda3\lib\site-packages (from weasel<0.4.0,>=0.1.0->spacy)
(0.16.0)
Requirement already satisfied: MarkupSafe>=2.0 in
c:\users\murat\anaconda3\lib\site-packages (from jinja2->spacy) (2.1.1)
Collecting en-core-web-sm==3.7.0
Downloading https://github.com/explosion/spacy-
models/releases/download/en_core_web_sm-3.7.0/en_core_web_sm-3.7.0-py3-none-
any.whl (12.8 MB)
0.0/12.8 MB ? eta -:--:--
0.0/12.8 MB 991.0 kB/s eta 0:00:13
0.1/12.8 MB 975.2 kB/s eta 0:00:14
0.1/12.8 MB 871.5 kB/s eta 0:00:15
0.1/12.8 MB 871.5 kB/s eta 0:00:15
0.2/12.8 MB 888.4 kB/s eta 0:00:15
0.2/12.8 MB 846.9 kB/s eta 0:00:15
110
0.3/12.8 MB 930.9 kB/s eta 0:00:14
- 0.3/12.8 MB 952.6 kB/s eta 0:00:14
- 0.4/12.8 MB 995.2 kB/s eta 0:00:13
- 0.4/12.8 MB 1.0 MB/s eta 0:00:13
- 0.5/12.8 MB 1.0 MB/s eta 0:00:12
- 0.5/12.8 MB 1.0 MB/s eta 0:00:12
- 0.5/12.8 MB 1.0 MB/s eta 0:00:12
- 0.6/12.8 MB 992.1 kB/s eta 0:00:13
- 0.6/12.8 MB 999.4 kB/s eta 0:00:13
- 0.6/12.8 MB 999.4 kB/s eta 0:00:13
-- 0.7/12.8 MB 982.7 kB/s eta 0:00:13
-- 0.7/12.8 MB 982.7 kB/s eta 0:00:13
-- 0.8/12.8 MB 927.5 kB/s eta 0:00:13
-- 0.8/12.8 MB 965.6 kB/s eta 0:00:13
-- 0.8/12.8 MB 965.6 kB/s eta 0:00:13
-- 0.9/12.8 MB 977.8 kB/s eta 0:00:13
--- 1.0/12.8 MB 983.0 kB/s eta 0:00:13
--- 1.1/12.8 MB 997.6 kB/s eta 0:00:12
--- 1.1/12.8 MB 1.0 MB/s eta 0:00:12
--- 1.2/12.8 MB 1.0 MB/s eta 0:00:12
--- 1.2/12.8 MB 1.0 MB/s eta 0:00:12
--- 1.2/12.8 MB 1.0 MB/s eta 0:00:12
---- 1.3/12.8 MB 1.1 MB/s eta 0:00:11
---- 1.3/12.8 MB 1.1 MB/s eta 0:00:11
---- 1.5/12.8 MB 1.1 MB/s eta 0:00:11
---- 1.5/12.8 MB 1.1 MB/s eta 0:00:11
---- 1.5/12.8 MB 1.1 MB/s eta 0:00:11
----- 1.6/12.8 MB 1.1 MB/s eta 0:00:11
----- 1.7/12.8 MB 1.1 MB/s eta 0:00:10
----- 1.7/12.8 MB 1.1 MB/s eta 0:00:10
----- 1.7/12.8 MB 1.1 MB/s eta 0:00:10
----- 1.7/12.8 MB 1.1 MB/s eta 0:00:10
----- 1.7/12.8 MB 1.1 MB/s eta 0:00:10
----- 1.9/12.8 MB 1.1 MB/s eta 0:00:11
------ 1.9/12.8 MB 1.1 MB/s eta 0:00:10
------ 2.0/12.8 MB 1.1 MB/s eta 0:00:10
------ 2.0/12.8 MB 1.1 MB/s eta 0:00:10
------ 2.1/12.8 MB 1.1 MB/s eta 0:00:10
------ 2.2/12.8 MB 1.1 MB/s eta 0:00:10
------ 2.2/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.3/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.3/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.3/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.4/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.4/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.5/12.8 MB 1.1 MB/s eta 0:00:10
------- 2.5/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.6/12.8 MB 1.1 MB/s eta 0:00:10
111
-------- 2.6/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.7/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.7/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.7/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.8/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.8/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.8/12.8 MB 1.1 MB/s eta 0:00:10
-------- 2.9/12.8 MB 1.1 MB/s eta 0:00:10
--------- 2.9/12.8 MB 1.0 MB/s eta 0:00:10
--------- 2.9/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.0/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.0/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.0/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.1/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.1/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.1/12.8 MB 1.0 MB/s eta 0:00:10
--------- 3.2/12.8 MB 1.0 MB/s eta 0:00:10
---------- 3.2/12.8 MB 1.0 MB/s eta 0:00:10
---------- 3.3/12.8 MB 1.0 MB/s eta 0:00:10
---------- 3.4/12.8 MB 1.0 MB/s eta 0:00:09
---------- 3.5/12.8 MB 1.1 MB/s eta 0:00:09
---------- 3.5/12.8 MB 1.1 MB/s eta 0:00:09
----------- 3.6/12.8 MB 1.1 MB/s eta 0:00:09
----------- 3.6/12.8 MB 1.1 MB/s eta 0:00:09
----------- 3.7/12.8 MB 1.1 MB/s eta 0:00:09
----------- 3.8/12.8 MB 1.1 MB/s eta 0:00:09
------------ 3.9/12.8 MB 1.1 MB/s eta 0:00:09
------------ 3.9/12.8 MB 1.1 MB/s eta 0:00:09
------------ 4.0/12.8 MB 1.1 MB/s eta 0:00:08
------------ 4.1/12.8 MB 1.1 MB/s eta 0:00:08
------------ 4.1/12.8 MB 1.1 MB/s eta 0:00:08
------------- 4.2/12.8 MB 1.1 MB/s eta 0:00:08
------------- 4.2/12.8 MB 1.1 MB/s eta 0:00:08
------------- 4.3/12.8 MB 1.1 MB/s eta 0:00:08
------------- 4.4/12.8 MB 1.1 MB/s eta 0:00:08
------------- 4.4/12.8 MB 1.1 MB/s eta 0:00:08
-------------- 4.5/12.8 MB 1.1 MB/s eta 0:00:08
-------------- 4.5/12.8 MB 1.1 MB/s eta 0:00:08
-------------- 4.6/12.8 MB 1.1 MB/s eta 0:00:08
-------------- 4.6/12.8 MB 1.1 MB/s eta 0:00:08
-------------- 4.7/12.8 MB 1.1 MB/s eta 0:00:08
-------------- 4.8/12.8 MB 1.1 MB/s eta 0:00:08
--------------- 4.8/12.8 MB 1.1 MB/s eta 0:00:07
--------------- 4.9/12.8 MB 1.1 MB/s eta 0:00:07
--------------- 4.9/12.8 MB 1.1 MB/s eta 0:00:07
--------------- 5.0/12.8 MB 1.2 MB/s eta 0:00:07
--------------- 5.1/12.8 MB 1.2 MB/s eta 0:00:07
--------------- 5.1/12.8 MB 1.2 MB/s eta 0:00:07
112
---------------- 5.2/12.8 MB 1.2 MB/s eta 0:00:07
---------------- 5.3/12.8 MB 1.2 MB/s eta 0:00:07
---------------- 5.3/12.8 MB 1.2 MB/s eta 0:00:07
---------------- 5.3/12.8 MB 1.2 MB/s eta 0:00:07
---------------- 5.4/12.8 MB 1.2 MB/s eta 0:00:07
---------------- 5.4/12.8 MB 1.2 MB/s eta 0:00:07
----------------- 5.4/12.8 MB 1.2 MB/s eta 0:00:07
----------------- 5.5/12.8 MB 1.2 MB/s eta 0:00:07
----------------- 5.5/12.8 MB 1.1 MB/s eta 0:00:07
----------------- 5.6/12.8 MB 1.1 MB/s eta 0:00:07
----------------- 5.6/12.8 MB 1.1 MB/s eta 0:00:07
----------------- 5.7/12.8 MB 1.1 MB/s eta 0:00:07
----------------- 5.7/12.8 MB 1.1 MB/s eta 0:00:07
------------------ 5.8/12.8 MB 1.1 MB/s eta 0:00:07
------------------ 5.8/12.8 MB 1.1 MB/s eta 0:00:07
------------------ 5.9/12.8 MB 1.1 MB/s eta 0:00:07
------------------ 5.9/12.8 MB 1.1 MB/s eta 0:00:07
------------------ 6.0/12.8 MB 1.1 MB/s eta 0:00:06
------------------ 6.0/12.8 MB 1.1 MB/s eta 0:00:06
------------------ 6.1/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.1/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.2/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.2/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.2/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.3/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.3/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.4/12.8 MB 1.1 MB/s eta 0:00:06
------------------- 6.4/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.4/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.5/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.5/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.5/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.6/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.6/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.6/12.8 MB 1.1 MB/s eta 0:00:06
-------------------- 6.7/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.7/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.8/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.8/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.8/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.9/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.9/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 6.9/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 7.0/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 7.0/12.8 MB 1.1 MB/s eta 0:00:06
--------------------- 7.0/12.8 MB 1.1 MB/s eta 0:00:06
---------------------- 7.1/12.8 MB 1.1 MB/s eta 0:00:06
---------------------- 7.1/12.8 MB 1.1 MB/s eta 0:00:06
113
---------------------- 7.2/12.8 MB 1.1 MB/s eta 0:00:06
---------------------- 7.2/12.8 MB 1.1 MB/s eta 0:00:06
---------------------- 7.3/12.8 MB 1.1 MB/s eta 0:00:06
---------------------- 7.3/12.8 MB 1.1 MB/s eta 0:00:05
---------------------- 7.4/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.4/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.4/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.5/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.5/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.6/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.6/12.8 MB 1.1 MB/s eta 0:00:05
----------------------- 7.7/12.8 MB 1.1 MB/s eta 0:00:05
------------------------ 7.7/12.8 MB 1.1 MB/s eta 0:00:05
------------------------ 7.8/12.8 MB 1.1 MB/s eta 0:00:05
------------------------ 7.8/12.8 MB 1.1 MB/s eta 0:00:05
------------------------ 7.9/12.8 MB 1.1 MB/s eta 0:00:05
------------------------ 7.9/12.8 MB 1.1 MB/s eta 0:00:05
------------------------ 8.0/12.8 MB 1.1 MB/s eta 0:00:05
------------------------- 8.0/12.8 MB 1.1 MB/s eta 0:00:05
------------------------- 8.1/12.8 MB 1.1 MB/s eta 0:00:05
------------------------- 8.2/12.8 MB 1.1 MB/s eta 0:00:05
------------------------- 8.3/12.8 MB 1.1 MB/s eta 0:00:05
------------------------- 8.3/12.8 MB 1.1 MB/s eta 0:00:05
-------------------------- 8.3/12.8 MB 1.1 MB/s eta 0:00:05
-------------------------- 8.4/12.8 MB 1.1 MB/s eta 0:00:04
-------------------------- 8.4/12.8 MB 1.1 MB/s eta 0:00:04
-------------------------- 8.5/12.8 MB 1.1 MB/s eta 0:00:04
-------------------------- 8.6/12.8 MB 1.1 MB/s eta 0:00:04
--------------------------- 8.6/12.8 MB 1.1 MB/s eta 0:00:04
--------------------------- 8.7/12.8 MB 1.1 MB/s eta 0:00:04
--------------------------- 8.8/12.8 MB 1.1 MB/s eta 0:00:04
--------------------------- 8.9/12.8 MB 1.1 MB/s eta 0:00:04
---------------------------- 9.0/12.8 MB 1.1 MB/s eta 0:00:04
---------------------------- 9.0/12.8 MB 1.1 MB/s eta 0:00:04
---------------------------- 9.1/12.8 MB 1.1 MB/s eta 0:00:04
---------------------------- 9.2/12.8 MB 1.1 MB/s eta 0:00:04
---------------------------- 9.3/12.8 MB 1.1 MB/s eta 0:00:04
----------------------------- 9.3/12.8 MB 1.1 MB/s eta 0:00:04
----------------------------- 9.4/12.8 MB 1.2 MB/s eta 0:00:03
----------------------------- 9.5/12.8 MB 1.2 MB/s eta 0:00:03
----------------------------- 9.5/12.8 MB 1.2 MB/s eta 0:00:03
----------------------------- 9.6/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------ 9.7/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------ 9.7/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------ 9.8/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------ 9.9/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------- 9.9/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------- 10.0/12.8 MB 1.2 MB/s eta 0:00:03
114
------------------------------- 10.1/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------- 10.1/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------- 10.2/12.8 MB 1.2 MB/s eta 0:00:03
------------------------------- 10.2/12.8 MB 1.2 MB/s eta 0:00:03
-------------------------------- 10.3/12.8 MB 1.2 MB/s eta 0:00:03
-------------------------------- 10.3/12.8 MB 1.2 MB/s eta 0:00:03
-------------------------------- 10.4/12.8 MB 1.2 MB/s eta 0:00:03
-------------------------------- 10.4/12.8 MB 1.2 MB/s eta 0:00:03
-------------------------------- 10.5/12.8 MB 1.2 MB/s eta 0:00:02
-------------------------------- 10.5/12.8 MB 1.2 MB/s eta 0:00:02
--------------------------------- 10.6/12.8 MB 1.2 MB/s eta 0:00:02
--------------------------------- 10.7/12.8 MB 1.2 MB/s eta 0:00:02
--------------------------------- 10.7/12.8 MB 1.2 MB/s eta 0:00:02
--------------------------------- 10.8/12.8 MB 1.2 MB/s eta 0:00:02
--------------------------------- 10.8/12.8 MB 1.2 MB/s eta 0:00:02
---------------------------------- 10.9/12.8 MB 1.2 MB/s eta 0:00:02
---------------------------------- 11.0/12.8 MB 1.2 MB/s eta 0:00:02
---------------------------------- 11.0/12.8 MB 1.2 MB/s eta 0:00:02
---------------------------------- 11.1/12.8 MB 1.2 MB/s eta 0:00:02
---------------------------------- 11.2/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.2/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.3/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.3/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.4/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.4/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.4/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.4/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.5/12.8 MB 1.2 MB/s eta 0:00:02
----------------------------------- 11.5/12.8 MB 1.2 MB/s eta 0:00:02
------------------------------------ 11.5/12.8 MB 1.2 MB/s eta 0:00:02
------------------------------------ 11.6/12.8 MB 1.2 MB/s eta 0:00:02
------------------------------------ 11.6/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------ 11.6/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------ 11.7/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------ 11.7/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------ 11.8/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------ 11.8/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------ 11.8/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------- 11.9/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------- 12.0/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------- 12.0/12.8 MB 1.2 MB/s eta 0:00:01
------------------------------------- 12.1/12.8 MB 1.2 MB/s eta 0:00:01
-------------------------------------- 12.2/12.8 MB 1.2 MB/s eta 0:00:01
-------------------------------------- 12.2/12.8 MB 1.2 MB/s eta 0:00:01
-------------------------------------- 12.3/12.8 MB 1.2 MB/s eta 0:00:01
-------------------------------------- 12.4/12.8 MB 1.2 MB/s eta 0:00:01
-------------------------------------- 12.4/12.8 MB 1.2 MB/s eta 0:00:01
-------------------------------------- 12.5/12.8 MB 1.2 MB/s eta 0:00:01
115
--------------------------------------- 12.6/12.8 MB 1.2 MB/s eta 0:00:01
--------------------------------------- 12.6/12.8 MB 1.2 MB/s eta 0:00:01
--------------------------------------- 12.7/12.8 MB 1.2 MB/s eta 0:00:01
--------------------------------------- 12.7/12.8 MB 1.2 MB/s eta 0:00:01
--------------------------------------- 12.8/12.8 MB 1.2 MB/s eta 0:00:01
--------------------------------------- 12.8/12.8 MB 1.2 MB/s eta 0:00:01
---------------------------------------- 12.8/12.8 MB 1.2 MB/s eta 0:00:00
Requirement already satisfied: spacy<3.8.0,>=3.7.0 in
c:\users\murat\anaconda3\lib\site-packages (from en-core-web-sm==3.7.0) (3.7.2)
Requirement already satisfied: spacy-legacy<3.1.0,>=3.0.11 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (3.0.12)
Requirement already satisfied: spacy-loggers<2.0.0,>=1.0.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (1.0.5)
Requirement already satisfied: murmurhash<1.1.0,>=0.28.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (1.0.10)
Requirement already satisfied: cymem<2.1.0,>=2.0.2 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (2.0.8)
Requirement already satisfied: preshed<3.1.0,>=3.0.2 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (3.0.9)
Requirement already satisfied: thinc<8.3.0,>=8.1.8 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (8.2.1)
Requirement already satisfied: wasabi<1.2.0,>=0.9.1 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (1.1.2)
Requirement already satisfied: srsly<3.0.0,>=2.4.3 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (2.4.8)
Requirement already satisfied: catalogue<2.1.0,>=2.0.6 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (2.0.10)
Requirement already satisfied: weasel<0.4.0,>=0.1.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (0.3.4)
Requirement already satisfied: typer<0.10.0,>=0.3.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (0.9.0)
Requirement already satisfied: smart-open<7.0.0,>=5.2.1 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (5.2.1)
Requirement already satisfied: tqdm<5.0.0,>=4.38.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (4.65.0)
116
Requirement already satisfied: requests<3.0.0,>=2.13.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (2.29.0)
Requirement already satisfied: pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (2.5.1)
Requirement already satisfied: jinja2 in c:\users\murat\anaconda3\lib\site-
packages (from spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (3.1.2)
Requirement already satisfied: setuptools in c:\users\murat\anaconda3\lib\site-
packages (from spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (67.8.0)
Requirement already satisfied: packaging>=20.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (23.0)
Requirement already satisfied: langcodes<4.0.0,>=3.2.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (3.3.0)
Requirement already satisfied: numpy>=1.19.0 in
c:\users\murat\anaconda3\lib\site-packages (from spacy<3.8.0,>=3.7.0->en-core-
web-sm==3.7.0) (1.24.3)
Requirement already satisfied: annotated-types>=0.4.0 in
c:\users\murat\anaconda3\lib\site-packages (from
pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy<3.8.0,>=3.7.0->en-core-web-
sm==3.7.0) (0.6.0)
Requirement already satisfied: pydantic-core==2.14.3 in
c:\users\murat\anaconda3\lib\site-packages (from
pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy<3.8.0,>=3.7.0->en-core-web-
sm==3.7.0) (2.14.3)
Requirement already satisfied: typing-extensions>=4.6.1 in
c:\users\murat\anaconda3\lib\site-packages (from
pydantic!=1.8,!=1.8.1,<3.0.0,>=1.7.4->spacy<3.8.0,>=3.7.0->en-core-web-
sm==3.7.0) (4.6.3)
Requirement already satisfied: charset-normalizer<4,>=2 in
c:\users\murat\anaconda3\lib\site-packages (from
requests<3.0.0,>=2.13.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in
c:\users\murat\anaconda3\lib\site-packages (from
requests<3.0.0,>=2.13.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (3.4)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in
c:\users\murat\anaconda3\lib\site-packages (from
requests<3.0.0,>=2.13.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (1.26.16)
Requirement already satisfied: certifi>=2017.4.17 in
c:\users\murat\anaconda3\lib\site-packages (from
requests<3.0.0,>=2.13.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (2023.5.7)
Requirement already satisfied: blis<0.8.0,>=0.7.8 in
c:\users\murat\anaconda3\lib\site-packages (from
thinc<8.3.0,>=8.1.8->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (0.7.11)
Requirement already satisfied: confection<1.0.0,>=0.0.1 in
c:\users\murat\anaconda3\lib\site-packages (from
117
thinc<8.3.0,>=8.1.8->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (0.1.3)
Requirement already satisfied: colorama in c:\users\murat\anaconda3\lib\site-
packages (from tqdm<5.0.0,>=4.38.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0)
(0.4.6)
Requirement already satisfied: click<9.0.0,>=7.1.1 in
c:\users\murat\anaconda3\lib\site-packages (from
typer<0.10.0,>=0.3.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (8.0.4)
Requirement already satisfied: cloudpathlib<0.17.0,>=0.7.0 in
c:\users\murat\anaconda3\lib\site-packages (from
weasel<0.4.0,>=0.1.0->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (0.16.0)
Requirement already satisfied: MarkupSafe>=2.0 in
c:\users\murat\anaconda3\lib\site-packages (from
jinja2->spacy<3.8.0,>=3.7.0->en-core-web-sm==3.7.0) (2.1.1)
Installing collected packages: en-core-web-sm
Successfully installed en-core-web-sm-3.7.0
[+] Download and installation successful
You can now load the package via spacy.load('en_core_web_sm')
import re
import nltk
from nltk.corpus import stopwords
nltk.download('stopwords')
stopwords = set(stopwords.words('english'))
nlp = spacy.load("en_core_web_sm")
118
self.vectorizer = TfidfVectorizer(lowercase=False, max_features=8000,
min_df=10, ngram_range=(1, 3),
tokenizer=None)
self.stopwords = stopwords
self.vectorizer_fitted = False
return lemmatized_texts
119
X = X[~X.isnull()] # delete nans
X = X[~X.duplicated()] # delete duplicates
if mode == 'train':
self.train_idx = X.index
else:
self.test_idx = X.index
print('Counting capitalized...')
capitalized = [np.sum([t.isupper() for t in text.split()])
for text in np.array(X.values)] # count capitalized
# X['cap'] = capitalized
print('Lowering...')
X = [text.lower() for text in X] # lower
X = self.remove_urls(X) # remove urls
X = self.remove_punctuation(X) # remove punctuation
X = self.remove_double_space(X) # remove double space
X = self.decode_emojis(X) # decode emojis
X = self.remove_stopwords(X) # remove stopwords
X = self.remove_numbers(X) # remove numbers
X = self.lemmatize(X) # lemmatize
if not self.vectorizer_fitted:
self.vectorizer_fitted = True
print('Fitting vectorizer...')
self.vectorizer.fit(X)
print('Vectorizing...')
X = self.vectorizer.transform(X) # vectorize
return X
[107]: pr = Preprocessor()
y_train = data['sentiment']
y_test = val_dataset['sentiment']
120
Requirement already satisfied: dill in c:\users\murat\anaconda3\lib\site-
packages (from nlp) (0.3.6)
Requirement already satisfied: pandas in c:\users\murat\anaconda3\lib\site-
packages (from nlp) (1.5.3)
Requirement already satisfied: requests>=2.19.0 in
c:\users\murat\anaconda3\lib\site-packages (from nlp) (2.29.0)
Requirement already satisfied: tqdm>=4.27 in c:\users\murat\anaconda3\lib\site-
packages (from nlp) (4.65.0)
Requirement already satisfied: filelock in c:\users\murat\anaconda3\lib\site-
packages (from nlp) (3.9.0)
Requirement already satisfied: xxhash in c:\users\murat\anaconda3\lib\site-
packages (from nlp) (3.4.1)
Requirement already satisfied: charset-normalizer<4,>=2 in
c:\users\murat\anaconda3\lib\site-packages (from requests>=2.19.0->nlp) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in
c:\users\murat\anaconda3\lib\site-packages (from requests>=2.19.0->nlp) (3.4)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in
c:\users\murat\anaconda3\lib\site-packages (from requests>=2.19.0->nlp)
(1.26.16)
Requirement already satisfied: certifi>=2017.4.17 in
c:\users\murat\anaconda3\lib\site-packages (from requests>=2.19.0->nlp)
(2023.5.7)
Requirement already satisfied: colorama in c:\users\murat\anaconda3\lib\site-
packages (from tqdm>=4.27->nlp) (0.4.6)
Requirement already satisfied: python-dateutil>=2.8.1 in
c:\users\murat\anaconda3\lib\site-packages (from pandas->nlp) (2.8.2)
Requirement already satisfied: pytz>=2020.1 in
c:\users\murat\anaconda3\lib\site-packages (from pandas->nlp) (2022.7)
Requirement already satisfied: six>=1.5 in c:\users\murat\anaconda3\lib\site-
packages (from python-dateutil>=2.8.1->pandas->nlp) (1.16.0)
ohe = OneHotEncoder()
referring_ohe = ohe.fit_transform(data_train['refers to'][data_train.index.
↪isin(pr.train_idx)].to_numpy().reshape(-1, 1))
121
Removing Nans…
Counting capitalized…
Lowering…
Removing URLs…
Removing Punctuation…
Removing double space…
Decoding emojis…
Removing stopwords…
100%|����������| 69486/69486 [00:02<00:00, 28649.18it/s]
Removing numbers…
100%|����������| 69486/69486 [00:00<00:00, 560542.01it/s]
Lemmatizing…
100%|����������| 69486/69486 [08:58<00:00, 129.02it/s]
Fitting vectorizer…
Vectorizing…
ohe = OneHotEncoder()
referring_ohe = ohe.fit_transform(data_train['refers to'][data_train.index.
↪isin(pr.train_idx)].to_numpy().reshape(-1, 1))
Removing Nans…
Counting capitalized…
Lowering…
Removing URLs…
Removing Punctuation…
Removing double space…
Decoding emojis…
Removing stopwords…
100%|����������| 998/998 [00:00<00:00, 188954.79it/s]
Removing numbers…
100%|����������| 998/998 [00:00<?, ?it/s]
122
Lemmatizing…
100%|����������| 998/998 [00:08<00:00, 120.70it/s]
Vectorizing…
3 Training
[119]: from sklearn.metrics import classification_report
from sklearn.model_selection import RandomizedSearchCV
from sklearn.linear_model import LogisticRegression
from scipy.stats import uniform, randint
from sklearn.model_selection import train_test_split, GridSearchCV, KFold
from sklearn.metrics import confusion_matrix
cv = RandomizedSearchCV(model,
params,
cv=kf,
123
scoring=scoring,
return_train_score=True,
n_jobs=-1,
verbose=2,
random_state=1
)
cv.fit(X_train, y_train)
[121]: rs_parameters = {
'penalty': ['l2', 'l1', 'elasticnet'],
'C': uniform(scale=10),
'solver': ['newton-cg', 'lbfgs', 'liblinear', 'saga'],
'l1_ratio': uniform(scale=10)
}
bestimator_lr = model_cv_lr.best_estimator_
124
Bir taban çizgisi için oldukça iniş puanı. Tüm sınıfların neredeyse eşit derecede iyi olduğu tahmin
ediliyor, karışıklık matrisi çapraz görünüyor.
Karşılıklı Bilgi puanına göre en az önemli olan bazı özellikleri çıkardıktan sonra aynı puanı alıp
alamayacağımıza bakalım.
125
[130]: plt.figure(figsize=(10, 8))
mi_imp = [pair[1] for pair in cols_importance[-30:]]
cols = [pair[0] for pair in cols_importance[-30:]]
sns.barplot(x=mi_imp, y=cols)
plt.title('En önemsiz özellikler')
plt.show()
126
[131]: X_train_6k = X_train[[pair[0] for pair in cols_importance[:6000]]]
X_test_6k = X_test[[pair[0] for pair in cols_importance[:6000]]]
bestimator_lr_6k = model_cv_lr_6k.best_estimator_
127
'l2', 'solver': 'liblinear'}
plt.show()
Puanlar düştü ama dramatik bir şekilde değil. Karışıklık matrisi hala köşegendir.
Özetlemek gerekirse, özelliklerin %75’i ile her özellikte %85 - %90 aralığında temel doğruluk elde
128
edebiliriz.
Collecting textblob
Downloading textblob-0.17.1-py2.py3-none-any.whl (636 kB)
0.0/636.8 kB ? eta -:--:--
------ 112.6/636.8 kB 3.3 MB/s eta 0:00:01
------ 112.6/636.8 kB 3.3 MB/s eta 0:00:01
-------------- 245.8/636.8 kB 2.2 MB/s eta 0:00:01
------------------ 307.2/636.8 kB 2.4 MB/s eta 0:00:01
------------------------- 419.8/636.8 kB 2.0 MB/s eta 0:00:01
---------------------------------- 573.4/636.8 kB 2.4 MB/s eta 0:00:01
---------------------------------- 573.4/636.8 kB 2.4 MB/s eta 0:00:01
-------------------------------------- 636.8/636.8 kB 2.0 MB/s eta 0:00:00
Requirement already satisfied: nltk>=3.1 in c:\users\murat\anaconda3\lib\site-
packages (from textblob) (3.7)
Requirement already satisfied: click in c:\users\murat\anaconda3\lib\site-
packages (from nltk>=3.1->textblob) (8.0.4)
Requirement already satisfied: joblib in c:\users\murat\anaconda3\lib\site-
packages (from nltk>=3.1->textblob) (1.2.0)
Requirement already satisfied: regex>=2021.8.3 in
c:\users\murat\anaconda3\lib\site-packages (from nltk>=3.1->textblob) (2022.7.9)
Requirement already satisfied: tqdm in c:\users\murat\anaconda3\lib\site-
packages (from nltk>=3.1->textblob) (4.65.0)
Requirement already satisfied: colorama in c:\users\murat\anaconda3\lib\site-
packages (from click->nltk>=3.1->textblob) (0.4.6)
Installing collected packages: textblob
Successfully installed textblob-0.17.1
129
model_cv = self.classifier.fit(X_vector, y)
self.classifier = model_cv#.best_estimator_
index=X_sentimental_idx)
pred_irrelevance = irrelevance[irrelevance == 1]
pred_irrelevance = pd.Series(pred_irrelevance, index=X_irrelevant_index)
pred_irrelevance = pred_irrelevance.apply(lambda x: 'Irrelevant')
return pred
sent_classifier.fit_classifier(X_train,
[1 if target == 'Irrelevant' else 0 for target␣
↪in y_train],
rs_parameters)
X_texts_test = data['text'][data.index.isin(X_test.index)]
130
dtype: int64
TextBlob verilen görevle yeterince başa çıkamıyor gibi görünüyor. Bu arada, optimize edilmemiş
lojistik regresyon, masajları daha sonra sınıflandırılan textblob’dan çok daha iyi sınıflandırmayı
başardı.
131
Karışıklık matrisi tam bir karmaşa. Tüm sınıflar çoğunlukla Pozitif olarak tespit edilir.
Bu deneyin başarısız olduğunu düşünüyorum ve daha sonra optimize etmeyi düşünmüyorum.
132
Requirement already satisfied: torch>=1.12.0 in
c:\users\murat\anaconda3\lib\site-packages (from pytorch-lightning) (2.1.1)
Requirement already satisfied: tqdm>=4.57.0 in
c:\users\murat\anaconda3\lib\site-packages (from pytorch-lightning) (4.65.0)
Requirement already satisfied: PyYAML>=5.4 in c:\users\murat\anaconda3\lib\site-
packages (from pytorch-lightning) (6.0)
Requirement already satisfied: fsspec[http]>2021.06.0 in
c:\users\murat\anaconda3\lib\site-packages (from pytorch-lightning) (2023.3.0)
Collecting torchmetrics>=0.7.0 (from pytorch-lightning)
Downloading torchmetrics-1.2.0-py3-none-any.whl (805 kB)
0.0/805.2 kB ? eta -:--:--
- 41.0/805.2 kB 1.9 MB/s eta 0:00:01
----- 112.6/805.2 kB 1.7 MB/s eta 0:00:01
----------- 245.8/805.2 kB 2.2 MB/s eta 0:00:01
---------------- 358.4/805.2 kB 2.2 MB/s eta 0:00:01
---------------------- 471.0/805.2 kB 2.5 MB/s eta 0:00:01
-------------------------- 553.0/805.2 kB 2.3 MB/s eta 0:00:01
-------------------------------- 686.1/805.2 kB 2.4 MB/s eta 0:00:01
------------------------------------- 798.7/805.2 kB 2.5 MB/s eta 0:00:01
-------------------------------------- 805.2/805.2 kB 2.2 MB/s eta 0:00:00
Requirement already satisfied: packaging>=20.0 in
c:\users\murat\anaconda3\lib\site-packages (from pytorch-lightning) (23.0)
Requirement already satisfied: typing-extensions>=4.0.0 in
c:\users\murat\anaconda3\lib\site-packages (from pytorch-lightning) (4.6.3)
Collecting lightning-utilities>=0.8.0 (from pytorch-lightning)
Downloading lightning_utilities-0.9.0-py3-none-any.whl (23 kB)
Requirement already satisfied: requests in c:\users\murat\anaconda3\lib\site-
packages (from fsspec[http]>2021.06.0->pytorch-lightning) (2.29.0)
Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in
c:\users\murat\anaconda3\lib\site-packages (from
fsspec[http]>2021.06.0->pytorch-lightning) (3.8.3)
Requirement already satisfied: filelock in c:\users\murat\anaconda3\lib\site-
packages (from torch>=1.12.0->pytorch-lightning) (3.9.0)
Requirement already satisfied: sympy in c:\users\murat\anaconda3\lib\site-
packages (from torch>=1.12.0->pytorch-lightning) (1.11.1)
Requirement already satisfied: networkx in c:\users\murat\anaconda3\lib\site-
packages (from torch>=1.12.0->pytorch-lightning) (2.8.4)
Requirement already satisfied: jinja2 in c:\users\murat\anaconda3\lib\site-
packages (from torch>=1.12.0->pytorch-lightning) (3.1.2)
Requirement already satisfied: colorama in c:\users\murat\anaconda3\lib\site-
packages (from tqdm>=4.57.0->pytorch-lightning) (0.4.6)
Requirement already satisfied: attrs>=17.3.0 in
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (22.1.0)
Requirement already satisfied: charset-normalizer<3.0,>=2.0 in
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (2.0.4)
Requirement already satisfied: multidict<7.0,>=4.5 in
133
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (6.0.2)
Requirement already satisfied: async-timeout<5.0,>=4.0.0a3 in
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (4.0.2)
Requirement already satisfied: yarl<2.0,>=1.0 in
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (1.8.1)
Requirement already satisfied: frozenlist>=1.1.1 in
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (1.3.3)
Requirement already satisfied: aiosignal>=1.1.2 in
c:\users\murat\anaconda3\lib\site-packages (from
aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>2021.06.0->pytorch-lightning) (1.2.0)
Requirement already satisfied: MarkupSafe>=2.0 in
c:\users\murat\anaconda3\lib\site-packages (from jinja2->torch>=1.12.0->pytorch-
lightning) (2.1.1)
Requirement already satisfied: idna<4,>=2.5 in
c:\users\murat\anaconda3\lib\site-packages (from
requests->fsspec[http]>2021.06.0->pytorch-lightning) (3.4)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in
c:\users\murat\anaconda3\lib\site-packages (from
requests->fsspec[http]>2021.06.0->pytorch-lightning) (1.26.16)
Requirement already satisfied: certifi>=2017.4.17 in
c:\users\murat\anaconda3\lib\site-packages (from
requests->fsspec[http]>2021.06.0->pytorch-lightning) (2023.5.7)
Requirement already satisfied: mpmath>=0.19 in
c:\users\murat\anaconda3\lib\site-packages (from sympy->torch>=1.12.0->pytorch-
lightning) (1.2.1)
Installing collected packages: lightning-utilities, torchmetrics, pytorch-
lightning
Successfully installed lightning-utilities-0.9.0 pytorch-lightning-2.1.2
torchmetrics-1.2.0
134
from torch.nn import functional as F
from torch.utils.data import DataLoader
from torch.utils.data import random_split
import pytorch_lightning as pl
import torch.utils.data as data_utils
from torch.optim.lr_scheduler import StepLR
from pytorch_lightning.callbacks import LearningRateMonitor
from IPython.display import clear_output
import numpy as np
torch.manual_seed(42)
pos = np.vstack(losses)
x, y = pos.T
plt.clf()
plt.ion()
plt.figure(figsize=(9, 5))
plt.plot(x, y)
plt.title(title)
clear_output(wait=True)
plt.show()
class NNSentimentClassifier(pl.LightningModule):
def __init__(self):
super().__init__()
self.softmax = nn.Softmax(dim=1)
self.dropout = nn.Dropout(0.2)
self.model = nn.Sequential(
nn.Linear(8032, 1000),
nn.ReLU(),
self.dropout,
nn.Linear(1000, 100),
nn.Tanh(),
self.dropout,
nn.Linear(100, 1000),
nn.ReLU(),
self.dropout,
nn.Linear(1000, 10),
nn.ReLU(),
self.dropout,
nn.Linear(10, 4)
135
)
self.acc_train_loss = []
self.acc_val_loss = []
def configure_optimizers(self):
optimizer = torch.optim.Adam(self.parameters(), lr=0.001,␣
↪weight_decay=0.0001)
return preds
[227]: # data
target_to_idx = {
'Irrelevant': 0,
'Negative': 1,
'Neutral': 2,
'Positive': 3
}
y_train_idx = torch.from_numpy(y_train.map(target_to_idx).values.astype(float))
136
train_data_tensor = data_utils.TensorDataset(torch.from_numpy(X_train.
↪to_numpy().astype(float)), y_train_idx)
---------------------------------------------------------------------------
MemoryError Traceback (most recent call last)
Cell In[227], line 10
2 target_to_idx = {
3 'Irrelevant': 0,
4 'Negative': 1,
5 'Neutral': 2,
6 'Positive': 3
7 }
9 y_train_idx = torch.from_numpy(y_train.map(target_to_idx).values.
↪astype(float))
MemoryError: Unable to allocate 4.16 GiB for an array with shape (69486, 8032)␣
↪and data type float64
137
Requirement already satisfied: sympy in c:\users\murat\anaconda3\lib\site-
packages (from torch==2.1.1->torchvision) (1.11.1)
Requirement already satisfied: networkx in c:\users\murat\anaconda3\lib\site-
packages (from torch==2.1.1->torchvision) (2.8.4)
Requirement already satisfied: jinja2 in c:\users\murat\anaconda3\lib\site-
packages (from torch==2.1.1->torchvision) (3.1.2)
Requirement already satisfied: fsspec in c:\users\murat\anaconda3\lib\site-
packages (from torch==2.1.1->torchvision) (2023.3.0)
Requirement already satisfied: charset-normalizer<4,>=2 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision) (3.4)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision)
(1.26.16)
Requirement already satisfied: certifi>=2017.4.17 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision)
(2023.7.22)
Requirement already satisfied: MarkupSafe>=2.0 in
c:\users\murat\anaconda3\lib\site-packages (from
jinja2->torch==2.1.1->torchvision) (2.1.1)
Requirement already satisfied: mpmath>=0.19 in
c:\users\murat\anaconda3\lib\site-packages (from
sympy->torch==2.1.1->torchvision) (1.2.1)
138
Requirement already satisfied: fsspec in c:\users\murat\anaconda3\lib\site-
packages (from torch) (2023.3.0)
Requirement already satisfied: numpy in c:\users\murat\anaconda3\lib\site-
packages (from torchvision) (1.24.3)
Requirement already satisfied: requests in c:\users\murat\anaconda3\lib\site-
packages (from torchvision) (2.29.0)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in
c:\users\murat\anaconda3\lib\site-packages (from torchvision) (9.4.0)
Requirement already satisfied: MarkupSafe>=2.0 in
c:\users\murat\anaconda3\lib\site-packages (from jinja2->torch) (2.1.1)
Requirement already satisfied: charset-normalizer<4,>=2 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision) (3.4)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision)
(1.26.16)
Requirement already satisfied: certifi>=2017.4.17 in
c:\users\murat\anaconda3\lib\site-packages (from requests->torchvision)
(2023.7.22)
Requirement already satisfied: mpmath>=0.19 in
c:\users\murat\anaconda3\lib\site-packages (from sympy->torch) (1.2.1)
Note: you may need to restart the kernel to use updated packages.
[235]: # model
model = NNSentimentClassifier(num_epochs=100,data_loader)
# training
trainer = pl.Trainer(max_epochs=20,accelerator="auto")
trainer.fit(model, dataloader)
[217]: preds = []
for batch in test_loader:
x, y = batch
139
y = y.long()
prediction = model.forward(x.float())
preds.extend(torch.argmax(prediction, dim=1).cpu())
[ ]:
140