[BUG] ConnectionResetError while processing big dataset

[mariannaparzych]

Describe the problem

I am trying to remove near-duplicates from big dataset (~91K images). After few hours of processing I got ConnectionResetError error.

Code to reproduce issue

Code I use:

def gen_approx_duplicate_groups_view(dataset, index):
    """This function is used to generate the approximate duplicate groups view.
    From: https://github.com/jacobmarks/image-deduplication-plugin/blob/main/approx_dups.py
    """

    dup_ids = index.duplicates_view().values("id")
    view = dataset.select(dup_ids)

    for rep_id, dups in index.neighbors_map.items():
        ids = [rep_id] + [d[0] for d in dups]
        subview = view.select(ids)
        for sample in subview:
            sample["approx_dup_group_id"] = rep_id
            sample.save()

    approx_dup_groups_view = view.group_by("approx_dup_group_id")
    dataset.save_view("approx_dup_groups_view", approx_dup_groups_view, overwrite=True)


def remove_approx_duplicates(dataset: fo.Dataset, fraction: float) -> fo.Dataset:
    similarity = fob.compute_similarity(dataset)
    similarity.find_duplicates(fraction=fraction)

    logging.debug("Preparing duplicates view")
    approx_dup_view = similarity.duplicates_view()
    dataset.save_view("approx_dup_view", approx_dup_view, overwrite=True)
    approx_dup_view = dataset.load_saved_view("approx_dup_view")
    logging.debug("Grouping duplicates view")
    gen_approx_duplicate_groups_view(dataset, similarity)

    approx_duplicates_list = []
    logging.debug("Adding near-duplicates to list.")
    for group_id in approx_dup_view.distinct("approx_dup_group_id"):
        group_view = approx_dup_view.match(F("approx_dup_group_id") == group_id)
        group_view = group_view.sort_by("filepath")
        approx_duplicates_list.append(group_view.values("id")[1:])
    approx_duplicates_list = list(set([item for sublist in approx_duplicates_list for item in sublist]))
    logging.info(f"Near-duplicates found. Removing {len(approx_duplicates_list)} of samples...")
    dataset.delete_samples(approx_duplicates_list)
    logging.info("Done")
    return dataset

System information

• OS Platform and Distribution: Linux Ubuntu 22.04
• Python version : 3.10.12
• FiftyOne version : FiftyOne v0.23.7, Voxel51, Inc.
• FiftyOne installed from : pip

Other info/logs

{"t":{"$date":"2024-04-23T05:21:33.443Z"},"s":"I",  "c":"CONTROL",  "id":20697,   "ctx":"-","msg":"Renamed existing log file","attr":{"oldLogPath":"/home/marianna.parzych/.fiftyone/var/lib/mongo/log/mong
o.log","newLogPath":"/home/marianna.parzych/.fiftyone/var/lib/mongo/log/mongo.log.2024-04-23T05-21-33"}}
Subprocess ['/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/db/bin/mongod', '--dbpath', '/home/marianna.parzych/.fiftyone/var/lib
/mongo', '--logpath', '/home/marianna.parzych/.fiftyone/var/lib/mongo/log/mongo.log', '--port', '0', '--nounixsocket'] exited with error -9:
Traceback (most recent call last):
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/pool.py", line 968, in command
    return command(
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/network.py", line 182, in command
    reply = receive_message(conn, request_id)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/network.py", line 257, in receive_message
    length, _, response_to, op_code = _UNPACK_HEADER(_receive_data_on_socket(conn, 16, deadline))
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/network.py", line 340, in _receive_data_on_socket
    chunk_length = conn.conn.recv_into(mv[bytes_read:])
ConnectionResetError: [Errno 104] Connection reset by peer

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 309, in <module>
    model_name: str,
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/click/core.py", line 1157, in __call__
    return self.main(*args, **kwargs)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/click/core.py", line 1078, in main
    rv = self.invoke(ctx)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/click/core.py", line 1434, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/click/core.py", line 783, in invoke
    return __callback(*args, **kwargs)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 294, in main
    type=list[str],
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 153, in deduplicate_image_dataset
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 133, in remove_approx_duplicates
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/collections.py", line 6514, in sort_by
    return self._add_view_stage(
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/view.py", line 1660, in _add_view_stage
    stage.validate(self)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/stages.py", line 6970, in validate
    sample_collection.create_index(index_spec)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/collections.py", line 8970, in create_index
    index_info = self.get_index_information()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/collections.py", line 8903, in get_index_information
    sample_info = self._dataset._sample_collection.index_information()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/collection.py", line 2347, in index_information
File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 294, in main
    type=list[str],
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 153, in deduplicate_image_dataset
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/unstructured_sg/scripts/sample_from_SEC_data_name-based_clusters.py", line 133, in remove_approx_duplicates
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/collections.py", line 6514, in sort_by
    return self._add_view_stage(
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/view.py", line 1660, in _add_view_stage
    stage.validate(self)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/stages.py", line 6970, in validate
    sample_collection.create_index(index_spec)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/collections.py", line 8970, in create_index
    index_info = self.get_index_information()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/fiftyone/core/collections.py", line 8903, in get_index_information
    sample_info = self._dataset._sample_collection.index_information()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/collection.py", line 2347, in index_information
    cursor = self.list_indexes(session=session, comment=comment)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/collection.py", line 2310, in list_indexes
    return self.__database.client._retryable_read(_cmd, read_pref, s)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/mongo_client.py", line 1492, in _retryable_read
    return self._retry_internal(
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/_csot.py", line 107, in csot_wrapper
    return func(self, *args, **kwargs)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/mongo_client.py", line 1462, in _retry_internal
    ).run()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/mongo_client.py", line 2320, in run
    self._check_last_error()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/mongo_client.py", line 2392, in _check_last_error
    raise self._last_error
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/mongo_client.py", line 2315, in run
    return self._read() if self._is_read else self._write()
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/mongo_client.py", line 2445, in _read
    return self._func(self._session, self._server, conn, read_pref)  # type: ignore
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/collection.py", line 2289, in _cmd
    cursor = self._command(conn, cmd, read_preference, codec_options, session=session)[
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/collection.py", line 308, in _command
    return conn.command(
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/helpers.py", line 322, in inner
    return func(*args, **kwargs)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/pool.py", line 996, in command
    self._raise_connection_failure(error)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/pool.py", line 1171, in _raise_connection_failure
    _raise_connection_failure(self.address, error, timeout_details=details)
  File "/mnt/ml-team/homes/marianna.parzych/Unstructured/od-modelling-super-gradients/venv/lib/python3.10/site-packages/pymongo/pool.py", line 411, in _raise_connection_failure
    raise AutoReconnect(msg) from error
pymongo.errors.AutoReconnect: localhost:37343: [Errno 104] Connection reset by peer (configured timeouts: connectTimeoutMS: 20000.0ms)
Skipping automatic non-persistent dataset cleanup. This action requires read access of the 'admin' database
[2024-04-23 12:54:42] WARNING - database.py - Skipping automatic non-persistent dataset cleanup. This action requires read access of the 'admin' database

Willingness to contribute

The FiftyOne Community encourages bug fix contributions. Would you or another member of your organization be willing to contribute a fix for this bug to the FiftyOne codebase?

• [ ] Yes. I can contribute a fix for this bug independently
• [ ] Yes. I would be willing to contribute a fix for this bug with guidance from the FiftyOne community
• [x] No. I cannot contribute a bug fix at this time