tcc package¶

tcc.client¶

Class structures for communicating with the API server over HTTP

class tcc.client.Client(user=None, api_key=None, url='http://localhost:80', engine='TeraChem', verbose=False)[source]¶

Bases: object

Main class for communication with the TeraChem Cloud API server

compute(geom, options, sleep_seconds=1, max_poll=20)[source]¶

Convenience routine for synchronous use.

Check self.poll_for_results() for recommended way to check for job completion.

Parameters:	geom ((num_atom, 3) ndarray) – Geometry to consider options (dict) – Job options to pass to TeraChem Cloud server sleep_seconds (int) – Number of seconds to wait between poll loops for self.poll_for_results() max_poll (int) – Number of poll loops for self.poll_for_results() **kwargs – TCC configuration passed to self.submit()
Returns:	Job results from TCC server
Return type:	dict

compute_bulk(geoms, options, sleep_seconds=1, max_poll=20)[source]¶

Convenience routine for multiple geometries.

Check self.poll_for_bulk_results() for recommended way to check for job completion.

Parameters:	geoms (list of (num_atom, 3) ndarray) – Geometries to consider options (dict) – Job options to pass to TeraChem Cloud server sleep_seconds (int) – Number of seconds to wait between poll loops for self.poll_for_bulk_results() max_poll (int) – Number of poll loops for self.poll_for_bulk_results() **kwargs – TCC configuration passed to self.submit()
Returns:	List of Job results from TCC server
Return type:	list

get_results(job_id)[source]¶

Query API for results of calculations.

Recommended way to check for job completion:

results = client.get_results(job_id)
finished = client.is_finished(results)

Parameters:	job_id (str) – Job id to check status of
Returns:	Result dictionary from TCC server with job_id added for posterity
Return type:	dict

is_finished(results)[source]¶

Helper function to test whether a job is finished.

Parameters:	results (dict) – Job results from self.get_results()
Returns:	True if job succeeded/failed, False if job is running/submitted/pending
Return type:	bool

poll_for_bulk_results(job_ids, sleep_seconds=1, max_poll=20)[source]¶

Send http request every sleep_seconds seconds until a finished job is returned or max_poll requests have been sent.

Recommended way to check for job completion:

results_list = client.poll_for_bulk_results(job_ids)
finished = [client.is_finished(r) for r in results_list]

Parameters:	job_ids (list) – Job ids to poll for sleep_seconds (int) – Number of seconds to wait between poll loops max_poll (int) – Number of poll loops
Returns:	List of results dicts as given by self.get_results()
Return type:	list

poll_for_results(job_id, sleep_seconds=1, max_poll=20)[source]¶

Send http request every sleep_seconds seconds until a finished job is returned or max_poll requests have been sent.

Recommended way to check for job completion:

results = client.poll_for_results(job_id)
finished = client.is_finished(results)

Parameters:	job_id (str) – Job id to poll for sleep_seconds (int) – Number of seconds to wait between poll loops max_poll (int) – Number of poll loops
Returns:	Results dict as given by self.get_results()
Return type:	dict

submit(geom, options)[source]¶

Pack and send the current tc_config dict as a POST request to the Tornado API server This function returns a job_id and a message

Parameters:	geom (np.ndarray or list) – Cartesian geometry at which to perform the calculation options (dict) – Job options to pass to TeraChem Cloud server
Returns:	Job id dict: Results
Return type:	str