Tools for Generation of Hybrid recordings #2436
Conversation
for more information, see https://pre-commit.ci
…rface into hybrid_raw_clustering
for more information, see https://pre-commit.ci
for more information, see https://pre-commit.ci
|
This PR depends on #2410 |
…e into hybrid_raw_clustering
…hybrid_raw_clustering
|
Added additional functionalities for hybrid generations:
Also extended the |
…e into hybrid_raw_clustering
|
@samuelgarcia @yger since I was working on it, I also added a Widen/Narrow option and scalebars for Screencast.from.14-05-2024.12.39.14.webmThis is ready to merge on my side! |
| else: | ||
| template_info = fetch_templates_info().iloc[template_df_or_indices] | ||
| requested_datasets = np.unique(template_info["dataset"]).tolist() | ||
| print(f"Fetching templates from {len(requested_datasets)} datasets") |
There was a problem hiding this comment.
a dreaded verbose option here?
| s3_path = f"s3://spikeinterface-template-database/{dataset}/" | ||
| zarr_group = zarr.open_consolidated(s3_path, storage_options={"anon": True}) | ||
|
|
||
| templates_object = Templates.from_zarr_group(zarr_group) | ||
|
|
||
| return templates_object | ||
|
|
||
|
|
||
| def fetch_templates_info() -> "pandas.DataFrame": |
There was a problem hiding this comment.
| def fetch_templates_info() -> "pandas.DataFrame": | |
| import functools | |
| @functools.cache | |
| def fetch_templates_info() -> "pandas.DataFrame": |
https://docs.python.org/3/library/functools.html#functools.cache
There was a problem hiding this comment.
Ah, this will fail if people still use python 3.8 : /
We can use functools.lru_cache(maxsize=1) instead if you want to support that but I would rather move forward to the bright future of python 3. 9 as minimal.
There was a problem hiding this comment.
yes let's do that for the sorting analyzer release
|
|
||
|
|
||
| def fetch_templates_from_database(dataset="test_templates.zarr") -> Templates: | ||
| def fetch_template_dataset(dataset="test_templates.zarr") -> Templates: |
There was a problem hiding this comment.
| def fetch_template_dataset(dataset="test_templates.zarr") -> Templates: | |
| @functools.cache | |
| def fetch_template_dataset(dataset="test_templates.zarr") -> Templates: |
| import zarr | ||
| import numpy as np | ||
|
|
There was a problem hiding this comment.
| import functools |
| channel_locations = sorting_analyzer.get_channel_locations()[ | ||
| sorting_analyzer.channel_ids_to_indices(channel_ids) | ||
| ] | ||
| unit_colors = get_unit_colors(unit_ids) |
There was a problem hiding this comment.
Use get_some_colors and revert signature
For the hybrid project, make a function that can get Templates from any recording. To be discussed, as some features to play/interact with Templates are currently only written for WaveformExtactor and might need to be duplicated