[Python API] robyn_refresh endpoint not yet develop?

[rupatrawut]

Project Robyn

Describe issue

I would like to do robyn_refresh to builds on top of an selected initial model with new coming data. However, I can't find the Python API endpoint for this function.

Is there any plan to develop this function? Or is it possible to run R function for model_refresh from Python outcome?

Environment & Robyn version

• Check and share Robyn version: 3.10.7.9000

[rupatrawut]

FYI, I try to save the model.json which is an input for robyn_refresh in Python using following code:

output_folder = Path(OutputCollect['plot_folder'][0])
for i in OutputCollect['clusters']['models']:
    print(i['solID'])
    write_robynmodel(sol=i['solID'], path=str(output_folder), InputJson=InputCollect, OutputJson=OutputCollect, OutputModels=OutputModels)

But got the MemoryError issue as below:

---

MemoryError Traceback (most recent call last)
Cell In[34], line 6
4 for i in OutputCollect['clusters']['models']:
5 print(i['solID'])
----> 6 write_robynmodel(sol=i['solID'], path=str(output_folder), InputJson=InputCollect, OutputJson=OutputCollect, OutputModels=OutputModels)
8 # OutputCollect['allSolutions']

File ~\workspace\ml-marketing-mix-modeling\ml_mmm_robyn\robyn_api\python_helper.py:301, in write_robynmodel(sol, path, InputJson, OutputJson, OutputModels)
293 payload = {
294 'InputCollect' : json.dumps(InputJson),
295 'OutputCollect' : json.dumps(OutputJson),
296 'OutputModels' : json.dumps(OutputModels),
297 "jsonWriteArgs": json.dumps(writeArgs)
298 }
300 # Get response
--> 301 respJson = robyn_api('robyn_write',payload=payload)
302 print('File written to path: ',updatedPath)

File ~\workspace\ml-marketing-mix-modeling\ml_mmm_robyn\robyn_api\python_helper.py:96, in robyn_api(argument, payload, api)
94 return respJson
95 else:
---> 96 response = requests.post(api.format(argument),data=payload)
97 respJson = json.loads(response.content.decode('utf-8'))
98 return respJson

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\api.py:115, in post(url, data, json, **kwargs)
103 def post(url, data=None, json=None, **kwargs):
104 r"""Sends a POST request.
105
106 :param url: URL for the new :class:Request object.
(...)
112 :rtype: requests.Response
113 """
--> 115 return request("post", url, data=data, json=json, **kwargs)

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\api.py:59, in request(method, url, **kwargs)
55 # By using the 'with' statement we are sure the session is closed, thus we
56 # avoid leaving sockets open which can trigger a ResourceWarning in some
57 # cases, and look like a memory leak in others.
58 with sessions.Session() as session:
---> 59 return session.request(method=method, url=url, **kwargs)

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\sessions.py:575, in Session.request(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)
562 # Create the Request.
563 req = Request(
564 method=method.upper(),
565 url=url,
(...)
573 hooks=hooks,
574 )
--> 575 prep = self.prepare_request(req)
577 proxies = proxies or {}
579 settings = self.merge_environment_settings(
580 prep.url, proxies, stream, verify, cert
581 )

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\sessions.py:486, in Session.prepare_request(self, request)
483 auth = get_netrc_auth(request.url)
485 p = PreparedRequest()
--> 486 p.prepare(
487 method=request.method.upper(),
488 url=request.url,
489 files=request.files,
490 data=request.data,
491 json=request.json,
492 headers=merge_setting(
493 request.headers, self.headers, dict_class=CaseInsensitiveDict
494 ),
495 params=merge_setting(request.params, self.params),
496 auth=merge_setting(auth, self.auth),
497 cookies=merged_cookies,
498 hooks=merge_hooks(request.hooks, self.hooks),
499 )
500 return p

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\models.py:371, in PreparedRequest.prepare(self, method, url, headers, files, data, params, auth, cookies, hooks, json)
369 self.prepare_headers(headers)
370 self.prepare_cookies(cookies)
--> 371 self.prepare_body(data, files, json)
372 self.prepare_auth(auth, url)
374 # Note that prepare_auth must be last to enable authentication schemes
375 # such as OAuth to work on a fully prepared request.
376
377 # This MUST go after prepare_auth. Authenticators could add a hook

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\models.py:559, in PreparedRequest.prepare_body(self, data, files, json)
557 else:
558 if data:
--> 559 body = self._encode_params(data)
560 if isinstance(data, basestring) or hasattr(data, "read"):
561 content_type = None

File c:\Users\rupatrawut\workspace\ml-marketing-mix-modeling\conda-env\Lib\site-packages\requests\models.py:129, in RequestEncodingMixin._encode_params(data)
124 for v in vs:
125 if v is not None:
126 result.append(
127 (
128 k.encode("utf-8") if isinstance(k, str) else k,
--> 129 v.encode("utf-8") if isinstance(v, str) else v,
130 )
131 )
132 return urlencode(result, doseq=True)
133 else:

MemoryError:

Which is wired because the model.json size when I ran the R code is just 20-30kb.

Environment:

• Windows 11
• RAM 32GB

[gufengzhou]

The refresh API endpoint is indeed not shipped yet. @yu-ya-tanaka do you want to have a look and talk to @ebyjose if you have time for this?

[rupatrawut]

Hi @gufengzhou, @yu-ya-tanaka thank you very much for all the support.
Is this feature in your developing plan?

By the way, the out-of-memory problem is solved, please ignore it.

[ebyjose]

Hi @rupatrawut : Currently we are still focusing on having the current python-api to be a safe and prioritizing support over development of new endpoints.
Thanks for confirming on the out-of memory solve. :)