masterscrat/chatistics

💬 Python scripts to parse Messenger, Hangouts, WhatsApp and Telegram chat logs into DataFrames.

chatistics's Issues

"Certifi" won't install

In requirements.txt, certifi won't install.

WhatsApp support?

Subscribe/comment to this issue if you would be interested in support of this platform.

charmap decode error - hangouts

while parsing hangouts json file

Traceback (most recent call last):
File "parsers/hangouts.py", line 24, in
archive = json.load(open(filePath))
File "C:\Users\San\Anaconda3\lib\json_init_.py", line 296, in load
return loads(fp.read(),
File "C:\Users\San\Anaconda3\lib\encodings\cp1252.py", line 23, in decode
return codecs.charmap_decode(input,self.errors,decoding_table)[0]
UnicodeDecodeError: 'charmap' codec can't decode byte 0x8d in position 103737: character maps to

ValueError: time data 'Tuesday, 21 November 2017 at 08:05 UTC+05:30' does not match format '%A, %B %d, %Y at %H:%M%p' (search)

While parsing messenger data , i got this

Unexpected date format. Falling back to infer_datetime_format, parsing will be slower.
Traceback (most recent call last):
File "C:\Users\San\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py", line 444, in _convert_listlike
values, tz = tslib.datetime_to_datetime64(arg)
File "pandas/_libs/tslib.pyx", line 1810, in pandas._libs.tslib.datetime_to_datetime64 (pandas_libs\tslib.c:33275)
TypeError: Unrecognized value type: <class 'str'>

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "parsers/messenger.py", line 87, in
timestamp = time.mktime(pd.to_datetime(content, format='%A, %B %d, %Y at %H:%M%p', exact=False).timetuple())
File "C:\Users\San\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py", line 518, in to_datetime
result = _convert_listlike(np.array([arg]), box, format)[0]
File "C:\Users\San\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py", line 447, in _convert_listlike
raise e
File "C:\Users\San\Anaconda3\lib\site-packages\pandas\core\tools\datetimes.py", line 414, in _convert_listlike
errors=errors)
File "pandas/_libs/tslib.pyx", line 3701, in pandas._libs.tslib.array_strptime (pandas_libs\tslib.c:63962)
ValueError: time data 'Tuesday, 21 November 2017 at 08:05 UTC+05:30' does not match format '%A, %B %d, %Y at %H:%M%p' (search)

Discord support?

Subscribe/comment to this issue if you would be interested in support of this platform.

PCDATA invalid Char value

Had this issue when running the parser and to solve is just add proper encoding on line 36 if anyone happens to face the same problem:

archive = etree.parse(filePath + "/" + filename,
                            etree.XMLParser(encoding='utf-8',
                                            ns_clean=True,
                                            recover=True))

should submit a PR?

charmap decode error - Whatsapp

after run "python parse.py whatsapp

Traceback (most recent call last):
File "parse.py", line 83, in
ArgParse()
File "parse.py", line 41, in init
getattr(self, args.command)()
File "parse.py", line 79, in whatsapp
main(args.own_name, args.file_path, args.max, args.infer_datetime)
File "C:\Users\rodrigo.teixeira\Documents\GitHub\Chatistics-master\parsers\whatsapp.py", line 62, in main
data = parse_messages(files, own_name, infer_datetime)
File "C:\Users\rodrigo.teixeira\Documents\GitHub\Chatistics-master\parsers\whatsapp.py", line 85, in parse_messages
regex_message = infer_datetime_regex(f_path)
File "C:\Users\rodrigo.teixeira\Documents\GitHub\Chatistics-master\parsers\whatsapp.py", line 24, in infer_datetime_regex
for c, line in enumerate(f):
File "C:\ProgramData\Anaconda3\envs\chatistics\lib\encodings\cp1252.py", line 23, in decode
return codecs.charmap_decode(input,self.errors,decoding_table)[0]
UnicodeDecodeError: 'charmap' codec can't decode byte 0x90 in position 4095: character maps to

Get most active users in a whatsapp group

Just thinking is there a way we can use this tool to get most active users in a Whatsapp group?

Telegram forwarded messages are not processed correctly

It seems that we need to support filtering messages that have been forwarded

Issue with parsing Messenger chat logs

After downloading the archive created by Facebook, we get chats in a directory named messages and messages.htm internally links to them. Even after adding the messages directory to the raw folder, no messages are being parsed. Need a little help here.

Language detection error when parsing messenger data

Am new to all this but am running into a problem when parsing my messenger data.
An exception gets thrown by lang detect when parsing the messages and leads to messages not being parsed.

Have included the log below to hopefully help -

2020-04-16 16:20:53,401 [INFO ] [parsers.mess]: Detecting languages...
Traceback (most recent call last):
File "parse.py", line 84, in
ArgParse()
File "parse.py", line 42, in init
getattr(self, args.command)()
File "parse.py", line 69, in messenger
main(args.own_name, args.file_path, args.max)
File "C:\Users\User\source\repos\Chatistics-master\parsers\messenger.py", line 31, in main
df = detect_language(df)
File "C:\Users\User\source\repos\Chatistics-master\parsers\utils.py", line 24, in detect_language
lang = langdetect.detect(text)
File "C:\Users\User\Anaconda3\lib\site-packages\langdetect\detector_factory.py", line 130, in detect
return detector.detect()
File "C:\Users\User\Anaconda3\lib\site-packages\langdetect\detector.py", line 136, in detect
probabilities = self.get_probabilities()
File "C:\Users\User\Anaconda3\lib\site-packages\langdetect\detector.py", line 143, in get_probabilities
self._detect_block()
File "C:\Users\User\Anaconda3\lib\site-packages\langdetect\detector.py", line 150, in _detect_block
raise LangDetectException(ErrorCode.CantDetectError, 'No features in text.')
langdetect.lang_detect_exception.LangDetectException: No features in text.

Pre-2014 Hangouts support

Google switched from "Google Talk" to "Google Hangouts" mid-2013.

Older messages could be retrieved using something like this: https://github.com/coandco/gtalk_export/

Messenger Parser outdated

I think facebook has recently changed the design for the archived messages recently; the parser for messenger is unable to detect any messages.

ValueError with datetime format

I am having the following issue with Pandas trying to read the datetime. Do you know why it's having trouble reading the full datetime?

Traceback

105 Person Name (group? False )
Traceback (most recent call last):
  File "parse_messenger.py", line 71, in <module>
    timestamp = pd.to_datetime(content[:-7], format='%A, %B %d, %Y at %H:%M%p').strftime("%s")
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\pandas\core\tools\datetimes.py", line 382, in to_datetime
    result = _convert_listlike(np.array([arg]), box, format)[0]
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\pandas\core\tools\datetimes.py", line 306, in _convert_listlike
    raise e
ValueError: time data 'Monday, January 22, 2018 at 6:4' does not match format '%A, %B %d, %Y at %H:%M%p' (match)

Text:

</style><title>Conversation with Friend Name</title></head><body><a href="html/messages.htm">Back</a><br /><br /><div class="thread"><h3>Conversation with Friend Name</h3>Participants: Friend Name<div class="message"><div class="message_header"><span class="user">Friend Name</span><span class="meta">Monday, January 22, 2018 at 6:46pm CST</span></div></div><p>Oh wow</p><div class="message"><div class="message_header"><span class="user">My Name</span><span class="meta">Monday, January 22, 2018 at 6:44pm CST</span></div></div><p>blah blah blah message was here</p><div class="message"><div class="message_header"><span class="user">Friend Name</span><span class="meta">Monday, January 22, 2018 at 6:40pm CST</span></div></div><p>

Signal support?

Subscribe/comment to this issue if you would be interested in support of this platform.

"ValueError: ordinal must be >= 1" when parsing Messenger chat logs

I still haven't had success yet with parsing the FB messages. One issues was that timestamp = pd.to_datetime(content, format='%A, %B %d, %Y at %H:%M%p', exact=False).strftime("%s") had an invalid format string. I changed it to "%S" since I assume that's what you meant. However, I now get the following issue when running analyse.py.

Traceback (most recent call last):
  File "analyse.py", line 90, in <module>
    print(plot)
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\ggplot\ggplot.py", line 116, in __repr__
    self.make()
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\ggplot\ggplot.py", line 641, in make
    self.apply_axis_labels()
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\ggplot\ggplot.py", line 231, in apply_axis_labels
    labels.append(self.xtick_formatter(label_text))
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\matplotlib\dates.py", line 542, in __call__
    dt = num2date(x, self.tz)
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\matplotlib\dates.py", line 445, in num2date
    return _from_ordinalf(x, tz)
  File "C:\Users\asros\Anaconda3\envs\py2\lib\site-packages\matplotlib\dates.py", line 260, in _from_ordinalf
    dt = datetime.datetime.fromordinal(ix).replace(tzinfo=UTC)
ValueError: ordinal must be >= 1

Module dependencies

Hey there,

So I've cloned the project a few times before, and even though I've gotten it to work, Chatistics doesn't work out of the box for me. My environment variables were already set, but LXML was giving me trouble, as where some dependencies that aren't listed in the requirements.txt. I noticed that there's a section for libxml in the README, but I just wanted to ask if you guys were taking any pull requests to add some fallbacks in case certain modules aren't already installed. Thanks!

Date parsing with Messenger chat logs is super slow

The dates we get in the Messenger chat logs do not use a consistent format (see #7).

I had to use infer_datetime_format=True to make date parsing usable for everyone. However, this is 10x slower than defining the date format explicitly.

Here are examples of dates found in the chat logs:

Sunday, 11 March 2012 at 15:11 UTC+01
Friday, November 10, 2017 at 1:19pm UTC+01
Monday, January 22, 2018 at 6:46pm CST

To speed this up, one solution would be to find out all the possible date formats used in the chat logs, write custom date parsers for them, and write regexps to recognize them.

WhatsApp parse error

I have exported the WhatsApp chat from a single conversation, and after running python parse.py whatsapp --own-name "firstname lastname" each row in the "senderName" column has a prefix of AM/PM (from the timestamp), e.g. "AM - firstname lastname" or "PM - firstname lastname". This leads to errors in visualization, etc.

Here are a few lines from the WhatsApp export for reproducibility:

9/18/18, 2:45 PM - firstname1 lastname1: I'm sorry! I was talking to people throughout the lunch break!
9/18/18, 4:24 PM - firstname2 lastname2: I'm glad it is going so well!
9/18/18, 4:25 PM - firstname2 lastname2: Here's a quick outtake from the filming
9/18/18, 4:25 PM - firstname2 lastname2: <Media omitted>

Troubles parsing messenger chats

Hello everyone!

I start saying I'm very inexperienced with... well... everything.

While parsing from WhatsApp works smoothly, I'm having some troubles with messenger chats.

This is the output I'm having:

2020-04-25 23:31:17,481 [INFO ] [parsers.mess]: Parsing Facebook messenger data...
Traceback (most recent call last):
  File "parse.py", line 84, in <module>
    ArgParse()
  File "parse.py", line 42, in __init__
    getattr(self, args.command)()
  File "parse.py", line 69, in messenger
    main(args.own_name, args.file_path, args.max)
  File "/Users/*****/Chatistics/parsers/messenger.py", line 22, in main
    data = parse_messages(file_path, own_name)
  File "/Users/*****/Chatistics/parsers/messenger.py", line 64, in parse_messages
    content = fix_text_encoding(content)
  File "/Users/*****/Chatistics/parsers/messenger.py", line 79, in fix_text_encoding
    return text.encode('latin1').decode('utf8')
UnicodeEncodeError: 'latin-1' codec can't encode character '\u201c' in position 17: ordinal not in range(256)

The output from export then prints:

...
2020-04-26 00:43:32,913 [INFO ] [utils       ]: Could not find any data for platform messenger

In /raw_data/messenger I have all the conversations folders that were previously contained in the inbox folder in the original archive downloaded from Facebook.
Like this: /raw_data/messenger/username/message_1.json.
(I think the documentation should be updated, it makes it look like that the inbox folder needs to be there too).

I'm working in Jupiter lab.

Thanks in advance!

Make beginner-friendly guide on how to use

Multiple people with little knowledge of Python have expressed interest in running the scripts.

There should be either a very simple guide or video, or a guided script that would let anyone render the final graphics.

Hangouts parser outdated

Google changed the format of the Hangouts.json data dump :-/

ValueError: time data 'Sunday, 11 March 2012 at 15:11 UTC+01' does not match format '%A, %B %d, %Y at %H:%M%p' (search)

I get a problem when trying to parse the facebook messenger files. I get
File "parse_messenger.py", line 76, in <module> timestamp = pd.to_datetime(content, format='%A, %B %d, %Y at %H:%M%p', exact=False).strftime("%s")

and

'ValueError: time data 'Sunday, 11 March 2012 at 15:11 UTC+01' does not match format '%A, %B %d, %Y at %H:%M%p' (search)

Is this perhaps due to different time stamps formats or that they are older?

Telegram support?

Subscribe/comment to this issue if you would be interested in support of this platform.

Improve command line filtering options

The command-line filtering options are limited for now, for example it is not possible to specify multiple names when using the -filterConversation flag.

In general, each of the available flag should handle a list of names instead of a single name.

unrecognized arguments

python analyse.py -data data/* raises below

analyse.py: error: unrecognized arguments: data/*

shared catalogs?

Are there any shared archives of threaded chat conversations? I'm trying to build a bot based on purely machine learning stuff, but my own logs aren't extensive enough.

I looked through some of the https://parl.ai/ corpus but there isn't much in the way of threaded conversation material.

any recommended sources would be appreciated.

Average response time between contacts?

I am curious about who doesn't respond to me quickly (or at all?) or who I am bad about responding to.

[Hangouts] KeyError: 'conversations'

Hi,

I tried:

$ python parse.py hangouts -f path/to/my/hangouts.json
2020-07-29 04:48:07,372 [INFO ] [parsers.hang]: Parsing Google Hangouts data...
2020-07-29 04:48:07,372 [INFO ] [parsers.hang]: Reading archive file path/to/my/hangouts.json...
2020-07-29 04:48:09,348 [INFO ] [parsers.hang]: Trying to infer own_name from data...
Traceback (most recent call last):
  File "parse.py", line 84, in <module>
    ArgParse()
  File "parse.py", line 42, in __init__
    getattr(self, args.command)()
  File "parse.py", line 60, in hangouts
    main(args.own_name, args.file_path, args.max)
  File "/path/to/Chatistics/parsers/hangouts.py", line 21, in main
    own_name = infer_own_name(archive)
  File "/path/to/Chatistics/parsers/hangouts.py", line 167, in infer_own_name
    for conversation in archive['conversations']:
KeyError: 'conversations'

Then, I tried:

$ python parse.py hangouts --own-name MARCO -f path/to/my/hangouts.json
2020-07-29 04:49:05,521 [INFO ] [parsers.hang]: Parsing Google Hangouts data...
2020-07-29 04:49:05,521 [INFO ] [parsers.hang]: Reading archive file path/to/my/hangouts.json...
2020-07-29 04:49:07,308 [INFO ] [parsers.hang]: Extracting messages...
Traceback (most recent call last):
  File "parse.py", line 84, in <module>
    ArgParse()
  File "parse.py", line 42, in __init__
    getattr(self, args.command)()
  File "parse.py", line 60, in hangouts
    main(args.own_name, args.file_path, args.max)
  File "/path/to/Chatistics/parsers/hangouts.py", line 22, in main
    data = parse_messages(archive, own_name)
  File "/path/to/Chatistics/parsers/hangouts.py", line 102, in parse_messages
    for conversation in archive['conversations']:
KeyError: 'conversations'

Then I tried to copy the file in ./raw_data/hangouts/, call python parse.py hangouts --own-name MARCO and got the same error.

WeChat support?

I would like to make some contribution if we can find a way to support wechat

Wrong distribution on breakdown plot

I have only exported 1 Whatsapp chat, and progressed to plot its breakdown.

First messages are in December 2017 (total of 30), but their column is also not reserved in plot.

Even though, total number of 783 messages is exchanged in
January 2020, the plot doesn't reflect these numbers, and shows like there are messages exchanged in upcoming months of 2020.

My steps:

1. Export Whatsapp chat from iOS 13 with email.

2. Put x_chat.txt to /raw_data

3. Run commands
   virtualenv chatistics
source chatistics/bin/activate
pip install -r requirements.txt
python parse.py whatsapp --own-name "My Name"
python visualize.py breakdown

4. Result
   https://i.imgur.com/TVgHaDX.png

Python 3.7.4

What can be the reason? How can I investigate it more and solve it?
Thanks.

When you parse logs from a group chat

2020-09-26 19:38:57,867 [INFO ] [matplotlib.f]: Generating new fontManager, this may take some time...
2020-09-26 19:39:01,947 [INFO ] [parsers.mess]: Parsing Facebook messenger data...
2020-09-26 19:39:01,948 [INFO ] [parsers.mess]: Trying to infer own_name from data...
2020-09-26 19:39:02,037 [INFO ] [parsers.mess]: Successfully inferred own-name to be User1
2020-09-26 19:39:02,061 [INFO ] [parsers.mess]: Group chats are not supported yet.
2020-09-26 19:39:02,079 [INFO ] [parsers.mess]: Group chats are not supported yet.
2020-09-26 19:39:02,100 [INFO ] [parsers.mess]: Group chats are not supported yet.
2020-09-26 19:39:02,109 [INFO ] [parsers.mess]: 0 messages parsed.
2020-09-26 19:39:02,109 [INFO ] [parsers.mess]: Nothing to save.

Can I kindly ask if there will be group chat support in the near future ?

@teamfridge not sure I understand. When you parse logs from a group chat, each message should have its own senderName, which allows you to perform analysis on a per-individual basis. You just have to group/filter the dataframes using this field.

Originally posted by @MasterScrat in #44 (comment)

Messenger badly encoded json

It appears that Facebook Messenger uses wrong encoding in the downloaded archive json files. I found a stackoverflow post which discussed it:
python - Facebook JSON badly encoded - Stack Overflow

I can indeed confirm that the Facebook download data is incorrectly encoded; a Mojibake. The original data is UTF-8 encoded but was decoded as Latin -1 instead.

I tried some weird unicode strings in my own json files:

>>> data = "\u00f0\u009f\u0098\u00a0"
>>> data
'ð\x9f\x98\xa0'
>>> data.encode('latin1').decode('utf8')
'😠'
>>>
>>> data = "\u00e6\u00b2\u00a1\u00e6\u009c\u0089\u00e5\u0095\u008a"
>>> data
'没æ\x9c\x89å\x95\x8a'
>>> data.encode('latin1').decode('utf8')
'没有啊'
>>>

It worked out just fine and seems like Facebook hasn't fixed this bug since then...

I applied this to all my data, then Chinese, Japanese and emojis are all correctly decoded. So at least Chinese, Japanese and emojis can be restored this way, and I guess it's the same for other languages (still need further tests tho).

So I think maybe you should add this .encode('latin1').decode('utf8') decode method for all the texts when parsing Messenger json files?

GroupMe support

Would you be open to a PR that added GroupMe support?

Group chat individualization

This may be functionality I haven't figured out yet, but would be great to parse group chats and get stats per individual user of that single group chat. Am exploring with Whatsapp exports but can't find a way as yet.

ModuleNotFoundError

Having ModuleNotFoundError while running the command

python parsers/messenger.py --own-name "John Doe"

There's __ init __.py in parsers subdirectory, still facing this issue.

Traceback (most recent call last): File "parsers/messenger.py", line 11, in <module> from parsers import log ModuleNotFoundError: No module named 'parsers'

Slack support?

Value Error in analyse.py with telegram.pkl

Hi there, thanks for these scripts.

when adding telegram data I get following error:

python analyse.py --data data/*

Traceback (most recent call last):
  File "Chatistics/lib/python3.6/site-packages/matplotlib/colors.py", line 132, in to_rgba
    rgba = _colors_full_map.cache[c, alpha]
KeyError: (736755, None)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "analyse.py", line 97, in <module>
    main()
  File "analyse.py", line 93, in main
    render(data, bin_width=args.bin_width, plot_density=args.density)
  File "analyse.py", line 81, in render
    print(plot)
  File "Chatistics/lib/python3.6/site-packages/ggplot/ggplot.py", line 116, in __repr__
    self.make()
  File "Chatistics/lib/python3.6/site-packages/ggplot/ggplot.py", line 636, in make
    layer.plot(ax, facetgroup, self._aes, **kwargs)
  File "Chatistics/lib/python3.6/site-packages/ggplot/geoms/geom_histogram.py", line 51, in plot
    ax.hist(x, **params)
  File "Chatistics/lib/python3.6/site-packages/matplotlib/__init__.py", line 1717, in inner
    return func(ax, *args, **kwargs)
  File "Chatistics/lib/python3.6/site-packages/matplotlib/axes/_axes.py", line 6137, in hist
    color = mcolors.to_rgba_array(color)
  File "/Chatistics/lib/python3.6/site-packages/matplotlib/colors.py", line 233, in to_rgba_array
    result[i] = to_rgba(cc, alpha)
  File "Chatistics/lib/python3.6/site-packages/matplotlib/colors.py", line 134, in to_rgba
    rgba = _to_rgba_no_colorcycle(c, alpha)
  File "Chatistics/lib/python3.6/site-packages/matplotlib/colors.py", line 185, in _to_rgba_no_colorcycle
    raise ValueError("Invalid RGBA argument: {!r}".format(orig_c))
ValueError: Invalid RGBA argument: 736755

using python analyse.py --plot-density --data data/* I get following:

Traceback (most recent call last):
  File "analyse.py", line 97, in <module>
    main()
  File "analyse.py", line 93, in main
    render(data, bin_width=args.bin_width, plot_density=args.density)
  File "analyse.py", line 81, in render
    print(plot)
  File "Chatistics/lib/python3.6/site-packages/ggplot/ggplot.py", line 116, in __repr__
    self.make()
  File "Chatistics/lib/python3.6/site-packages/ggplot/ggplot.py", line 641, in make
    self.apply_axis_labels()
  File "Chatistics/lib/python3.6/site-packages/ggplot/ggplot.py", line 231, in apply_axis_labels
    labels.append(self.xtick_formatter(label_text))
  File "Chatistics/lib/python3.6/site-packages/matplotlib/dates.py", line 578, in __call__
    dt = num2date(x, self.tz)
  File "Chatistics/lib/python3.6/site-packages/matplotlib/dates.py", line 466, in num2date
    return _from_ordinalf(x, tz)
  File "Chatistics/lib/python3.6/site-packages/matplotlib/dates.py", line 279, in _from_ordinalf
    dt = datetime.datetime.fromordinal(ix).replace(tzinfo=UTC)
ValueError: year 3997345 is out of range

Everything works normal using only facebook messages.
Any idea how to fix that?

No input files found under raw_data/messenger

I followed the steps outlined in the readme.

Downloaded messages from facebook in JSON format. Unzipped and put the content from the messages folder in my raw_data/messenger directory.

When I run python parse.py messenger I get this error:

2020-01-23 04:27:34,164 [INFO ] [parsers.mess]: Parsing Facebook messenger data...
2020-01-23 04:27:34,165 [ERROR] [parsers.mess]: No input files found under raw_data/messenger

Problems with ggplot: Histograms don't stack, density plots don't fill

As reported by @edvinli in #7 :

• I don't get the histogram to stack, as in the example in the readme. I get that the histogram when plotting my top 10 are added behind and in front of each other instead of stacking. Any idea as to why?

• The density plot does not fill, I haven't figured out why that is.

I guess there are breaking changes in ggplot between versions 0.6.8 and 0.11.5 (dependency updated in 3c78a53).

[ERROR] [parsers.hang]: Parsing error. Is your own_name ****** correct?

I have done the command !python parse.py hangouts --own-name "**********", and then I get the above error. How can I fix this?

Favorite emojis?

I'd be curious what my favorite emojis are and how that compares to my friend's favorite emojis

Problem activating Chatistics shell

I had it working on the 17th of January but I updated the master branch today and now conda activate chatistics fails with:

CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
To initialize your shell, run

    $ conda init <SHELL_NAME>

Currently supported shells are:
  - bash
  - fish
  - tcsh
  - xonsh
  - zsh
  - powershell

See 'conda init --help' for more information and options.

IMPORTANT: You may need to close and restart your shell after running 'conda init'.

Before trying this I deleted the existing environment out of ~/anaconda/envs/ and ran conda env create -f environment.yml

I also updated to Catalina in the meantime if that makes any difference.

Any ideas as to how I can get this working?

Edited to remove env from conda activate chatistics.