issue_comments
103 rows where issue = 253136694 sorted by updated_at descending
This data as json, CSV (advanced)
Suggested facets: reactions, created_at (date), updated_at (date)
issue 1
- WIP: Zarr backend · 103 ✖
id | html_url | issue_url | node_id | user | created_at | updated_at ▲ | author_association | body | reactions | performed_via_github_app | issue |
---|---|---|---|---|---|---|---|---|---|---|---|
365412033 | https://github.com/pydata/xarray/pull/1528#issuecomment-365412033 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NTQxMjAzMw== | martindurant 6042212 | 2018-02-13T21:35:03Z | 2018-02-13T21:35:03Z | CONTRIBUTOR | Yeah, ideally when adding a variable like
On the other hand, implementing
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364954680 | https://github.com/pydata/xarray/pull/1528#issuecomment-364954680 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDk1NDY4MA== | rabernat 1197350 | 2018-02-12T15:21:51Z | 2018-02-12T15:21:51Z | MEMBER | I'm enjoying this discussion. Zarr offers lots of new possibilities for appending / updating datasets that we should try to support. I personally would really like to be able to append / extend existing arrays from within xarray. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364817111 | https://github.com/pydata/xarray/pull/1528#issuecomment-364817111 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgxNzExMQ== | martindurant 6042212 | 2018-02-12T02:43:43Z | 2018-02-12T03:47:48Z | CONTRIBUTOR | OK, so the way to do this in pure-zarr appears to be to simply create the appropriate zarr array and set it's dimensions attribute:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364812486 | https://github.com/pydata/xarray/pull/1528#issuecomment-364812486 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgxMjQ4Ng== | jakirkham 3019665 | 2018-02-12T01:51:40Z | 2018-02-12T01:51:40Z | NONE | So Zarr supports storing structured arrays. Maybe that’s what you are looking for, @martindurant? Would suggest using the latest 2.2.0 RC though as it fixed a few issues in this regard (particularly with NumPy 1.14). |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364804697 | https://github.com/pydata/xarray/pull/1528#issuecomment-364804697 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwNDY5Nw== | martindurant 6042212 | 2018-02-12T00:19:55Z | 2018-02-12T00:19:55Z | CONTRIBUTOR | It might be enough, in this case, to provide some helper function in zarr to create and fetch arrays that will show up as variables in xarray - this need not be specific to being used via dask. I am assuming with the work done in this PR, that there is an unambiguous way to determine if a zarr group can be interpreted as an xarray dataset, and that zarr then knows how to add things that look like variables (which generally in the zarr case don't involve writing any actual data until the parts of the array are filled in). |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364804265 | https://github.com/pydata/xarray/pull/1528#issuecomment-364804265 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwNDI2NQ== | shoyer 1217238 | 2018-02-12T00:15:23Z | 2018-02-12T00:15:23Z | MEMBER | See https://github.com/dask/dask/issues/2000 for the dask issue. Once this works in dask it should be quite easy to implement in xarray, too. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364804162 | https://github.com/pydata/xarray/pull/1528#issuecomment-364804162 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwNDE2Mg== | shoyer 1217238 | 2018-02-12T00:14:22Z | 2018-02-12T00:14:22Z | MEMBER | @martindurant that could probably be addressed most cleanly by improving |
{ "total_count": 2, "+1": 2, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364803984 | https://github.com/pydata/xarray/pull/1528#issuecomment-364803984 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwMzk4NA== | martindurant 6042212 | 2018-02-12T00:12:36Z | 2018-02-12T00:12:36Z | CONTRIBUTOR | @jhamman , that partially solves what I mean, I can probably turn my data into dask arrays with some difficulty; but really I was hoping for something like the following:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364802374 | https://github.com/pydata/xarray/pull/1528#issuecomment-364802374 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwMjM3NA== | jhamman 2443309 | 2018-02-11T23:54:01Z | 2018-02-11T23:54:01Z | MEMBER | @martindurant - If I understand your question correctly, I think you should be able to follow a pretty standard xarray workflow: ```Python ds = xr.Dataset() ds['your_varname'] = xr.DataArray(some_dask_array, dims=['dimname0', 'dimname1', ...], coords=dict_of_preknown_coords) repeat for each variable you want in your datasetds.to_zarr(some_zarr_store) then to opends2 = xr.open_zarr(some_zarr_store) ``` Two things to note: 1) if you are looking for decent performance when writing to a remote store, make sure you're working off xarray@master as #1800 fixed a number of choke points in the to_zarr implementation
2) if you are pushing to GCS, |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364801395 | https://github.com/pydata/xarray/pull/1528#issuecomment-364801395 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwMTM5NQ== | mrocklin 306380 | 2018-02-11T23:40:18Z | 2018-02-11T23:40:18Z | MEMBER | Does the to_zarr method suffice: http://xarray.pydata.org/en/latest/generated/xarray.Dataset.to_zarr.html#xarray.Dataset.to_zarr ? On Sun, Feb 11, 2018 at 6:35 PM, Martin Durant notifications@github.com wrote:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
364801073 | https://github.com/pydata/xarray/pull/1528#issuecomment-364801073 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM2NDgwMTA3Mw== | martindurant 6042212 | 2018-02-11T23:35:34Z | 2018-02-11T23:35:34Z | CONTRIBUTOR | Question: how would one build a zarr-xarray dataset? With zarr you can open an array that contains no data, and use set-slice notation to fill in the values (which is what dask's store essentially does). If I have some pre-known coordinates and bigger-than-memory data arrays, how would I go about getting the values into the zarr structure? If this can't be done directly with the xarray interface, is there a way to call zarr's open/create/zeros such that the corresponding array will appear as a variable when the same dataset is opened with xarray? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
351588678 | https://github.com/pydata/xarray/pull/1528#issuecomment-351588678 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MTU4ODY3OA== | shoyer 1217238 | 2017-12-14T02:23:03Z | 2017-12-14T02:23:03Z | MEMBER | woohoo, thank you Ryan! |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
351401474 | https://github.com/pydata/xarray/pull/1528#issuecomment-351401474 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MTQwMTQ3NA== | rabernat 1197350 | 2017-12-13T14:09:12Z | 2017-12-13T14:09:12Z | MEMBER | Will merge later today if no further comments. |
{ "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350557153 | https://github.com/pydata/xarray/pull/1528#issuecomment-350557153 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDU1NzE1Mw== | fmaussion 10050469 | 2017-12-10T15:45:13Z | 2017-12-10T15:45:13Z | MEMBER | Thanks for the tremendous work @rabernat , looking forward to testing this! In the future it would be nice to shortly describe the advantages of zarr over netcdf for new users. A speed benchmark could help, too! This can be done once the backend has more maturity, and when we will refactor the I/O docs |
{ "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350504017 | https://github.com/pydata/xarray/pull/1528#issuecomment-350504017 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDUwNDAxNw== | jakirkham 3019665 | 2017-12-09T20:38:58Z | 2017-12-09T20:38:58Z | NONE |
As a minor point to complement what Matthew and Alistair have already said, one can pretty easily |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350375750 | https://github.com/pydata/xarray/pull/1528#issuecomment-350375750 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDM3NTc1MA== | alimanfoo 703554 | 2017-12-08T21:24:45Z | 2017-12-08T22:27:47Z | CONTRIBUTOR | Just to confirm, if writes are aligned with chunk boundaries in the destination array then no locking is required. Also if you're going to be moving large datasets into cloud storage and doing distributed computing then it may be worth investigating compressors and compressor options as good compression ratio may make a big difference where network bandwidth may be the limiting factor. I would suggest using the Blosc compressor with cname='zstd'. I would also suggest using shuffle, the Blosc codec in latest numcodecs has an AUTOSHUFFLE option so byte shuffle is used for arrays with >1 byte item size and bit shuffle is used for arrays with 1 byte item size . I would also experiment with compression level (clevel) to see how speed balances against compression ratio. E.g., Blosc(cname='zstd', clevel=5, shuffle=Blosc.AUTOSHUFFLE) may be a good starting point. The default compressor is Blosc(cname='lz4', ...) is more optimised for fast local storage, so speed is very good but compression ratio is moderate, this may not be best for distributed computing. |
{ "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350379064 | https://github.com/pydata/xarray/pull/1528#issuecomment-350379064 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDM3OTA2NA== | alimanfoo 703554 | 2017-12-08T21:40:40Z | 2017-12-08T22:27:35Z | CONTRIBUTOR | Some examples of compressor benchmarking here may be useful http://alimanfoo.github.io/2016/09/21/genotype-compression-benchmark.html The specific conclusions probably won't apply to your data but some of the code and ideas may be useful. Since writing that article I added Zstd and LZ4 compressors in numcodecs so those may also be worth trying in addition to Blosc with various configurations. (Blosc breaks up each chunk into blocks which enables multithreaded compression/decompression but can also reduce compression ratio over the same compressor library used without Blosc. I.e., Blosc(cname='zstd', clevel=1) will behave differently from Zstd(level=1) even though the same underlying compression library (Zstandard) is being used.) |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350365780 | https://github.com/pydata/xarray/pull/1528#issuecomment-350365780 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDM2NTc4MA== | rabernat 1197350 | 2017-12-08T20:36:26Z | 2017-12-08T20:36:26Z | MEMBER | Any more reviews? @fmaussion & @pwolfram: you have experience with backends. Your reviews would be valuable. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350352097 | https://github.com/pydata/xarray/pull/1528#issuecomment-350352097 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDM1MjA5Nw== | shoyer 1217238 | 2017-12-08T19:34:09Z | 2017-12-08T19:34:09Z | MEMBER |
Oops, this is my fault! Instead, try:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350343117 | https://github.com/pydata/xarray/pull/1528#issuecomment-350343117 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDM0MzExNw== | mrocklin 306380 | 2017-12-08T18:55:35Z | 2017-12-08T18:55:35Z | MEMBER | Not as far as I know. On Fri, Dec 8, 2017 at 1:53 PM, Ryan Abernathey notifications@github.com wrote:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
350336238 | https://github.com/pydata/xarray/pull/1528#issuecomment-350336238 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM1MDMzNjIzOA== | rabernat 1197350 | 2017-12-08T18:26:58Z | 2017-12-08T18:26:58Z | MEMBER | There is a silly lingering issue that I need help resolving. In a8b478543a978bd98c37711609c610432fdc7d07, @jhamman added a function
The |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349992006 | https://github.com/pydata/xarray/pull/1528#issuecomment-349992006 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTk5MjAwNg== | rabernat 1197350 | 2017-12-07T14:59:12Z | 2017-12-07T14:59:12Z | MEMBER | @jhamman, I can't reproduce your error. If you can give me a reproducible example, I will make a test for it. I think this is converging. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349766763 | https://github.com/pydata/xarray/pull/1528#issuecomment-349766763 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTc2Njc2Mw== | rabernat 1197350 | 2017-12-06T20:36:03Z | 2017-12-06T20:36:03Z | MEMBER | @jhamman - but the error being raised is wrong! There is a string formatting error raised in trying to generate a useful, informative error message. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349738624 | https://github.com/pydata/xarray/pull/1528#issuecomment-349738624 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTczODYyNA== | jhamman 2443309 | 2017-12-06T18:54:41Z | 2017-12-06T18:54:56Z | MEMBER | @rabernat - in trying out your branch, I've run into this error (mentioned by @mrocklin in pangeo-data/pangeo#19): ```Python-traceback ... ~/anaconda/envs/pangeo-dev/lib/python3.6/site-packages/xarray-0.10.0_79_g7b50320-py3.6.egg/xarray/backends/zarr.py in _extract_zarr_variable_encoding(variable, raise_on_invalid) 228 229 chunks = _determine_zarr_chunks(encoding.get('chunks'), variable.chunks, --> 230 variable.ndim) 231 encoding['chunks'] = chunks 232 return encoding ~/anaconda/envs/pangeo-dev/lib/python3.6/site-packages/xarray-0.10.0_79_g7b50320-py3.6.egg/xarray/backends/zarr.py in _determine_zarr_chunks(enc_chunks, var_chunks, ndim)
134 "Zarr requires uniform chunk sizes excpet for final chunk."
135 " Variable %r has incompatible chunks. Consider "
--> 136 "rechunking using TypeError: not all arguments converted during string formatting ``` As far as I can tell, reworking my chunk sizes to divide evenly into the dataset dimensions has corrected the problem. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349554730 | https://github.com/pydata/xarray/pull/1528#issuecomment-349554730 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTU1NDczMA== | shoyer 1217238 | 2017-12-06T07:10:37Z | 2017-12-06T07:10:37Z | MEMBER | I just pushed a commit adding a test for |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349540155 | https://github.com/pydata/xarray/pull/1528#issuecomment-349540155 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTU0MDE1NQ== | rabernat 1197350 | 2017-12-06T05:38:26Z | 2017-12-06T05:38:26Z | MEMBER | I believe that this is now complete enough to consider merging. I have addressed nearly all of @shoyer's suggestions. I have added a bunch more tests and am now quite satisfied with the test suite. I wrote some basic documentation, with the usual disclaimers about the experimental nature of this new feature. The zarr tests will not run if the zarr version is less than 2.2.0. This is not released yet. This means that only the py36-zarr-dev build actually runs the zarr tests. Once @alimanfoo releases the next version, the zarr tests should kick in on all the builds. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349495568 | https://github.com/pydata/xarray/pull/1528#issuecomment-349495568 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTQ5NTU2OA== | rabernat 1197350 | 2017-12-06T01:08:11Z | 2017-12-06T01:08:11Z | MEMBER | @jhamman - could you elaborate on the nature of the error you got with uneven dask chunks. We should be catching this and raising a useful error message. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
349488598 | https://github.com/pydata/xarray/pull/1528#issuecomment-349488598 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0OTQ4ODU5OA== | mrocklin 306380 | 2017-12-06T00:30:21Z | 2017-12-06T00:30:21Z | MEMBER | We tried this out on a cloud-deployed cluster on GCE and things worked pleasantly. Some conversation here: https://github.com/pangeo-data/pangeo/issues/19 |
{ "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 1, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
348839453 | https://github.com/pydata/xarray/pull/1528#issuecomment-348839453 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0ODgzOTQ1Mw== | alimanfoo 703554 | 2017-12-04T01:40:57Z | 2017-12-04T01:40:57Z | CONTRIBUTOR | I know you're not including string support in this PR, but for interest, there are a couple of changes coming into zarr via https://github.com/alimanfoo/zarr/pull/212 that may be relevant in future. It should now be impossible to generate a segfault via a badly configured object array. It is also now much harder to badly configure an object array. When creating an object array, an object codec should be provided via the |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
348569223 | https://github.com/pydata/xarray/pull/1528#issuecomment-348569223 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0ODU2OTIyMw== | shoyer 1217238 | 2017-12-01T18:20:32Z | 2017-12-01T18:20:32Z | MEMBER |
Variable length strings are stored with |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
348564159 | https://github.com/pydata/xarray/pull/1528#issuecomment-348564159 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0ODU2NDE1OQ== | rabernat 1197350 | 2017-12-01T17:58:59Z | 2017-12-01T17:59:06Z | MEMBER | Sorry this has become such a behemoth. I know it is hard to review. I couldn't see how to make a more atomic PR because a new backend has lots of interrelated parts that need each other in order to work. To finish it up, I propose to raise an error when attempting to encode variable-length string data. If someone can give me a quick one liner to help identify such datatypes, that would be helpful. We will revisit these encoding issues once Stephan's refactoring is merged. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
348560326 | https://github.com/pydata/xarray/pull/1528#issuecomment-348560326 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0ODU2MDMyNg== | shoyer 1217238 | 2017-12-01T17:43:03Z | 2017-12-01T17:43:03Z | MEMBER | I'll give this another look over the weekend. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
348414545 | https://github.com/pydata/xarray/pull/1528#issuecomment-348414545 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0ODQxNDU0NQ== | jhamman 2443309 | 2017-12-01T06:40:47Z | 2017-12-01T06:40:47Z | MEMBER | @rabernat - following @shoyer's thoughts here and in #1753, I'm not apposed to skipping the last few failing tests and live to fight strings another day. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347989858 | https://github.com/pydata/xarray/pull/1528#issuecomment-347989858 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0Nzk4OTg1OA== | rabernat 1197350 | 2017-11-29T20:42:34Z | 2017-11-29T20:42:34Z | MEMBER | Actually, I think I just realized how to do it without too much pain. Stand by. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347987097 | https://github.com/pydata/xarray/pull/1528#issuecomment-347987097 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0Nzk4NzA5Nw== | rabernat 1197350 | 2017-11-29T20:32:07Z | 2017-11-29T20:32:07Z | MEMBER |
Because of the way the backends are structured right now, it is hard to bypass the existing encoding and replace it with a new encoding scheme. #1087 will make this easy to do. But now it is complicated. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347984582 | https://github.com/pydata/xarray/pull/1528#issuecomment-347984582 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0Nzk4NDU4Mg== | shoyer 1217238 | 2017-11-29T20:22:33Z | 2017-11-29T20:22:33Z | MEMBER | I'm fine skipping strings entirely for now. They are indeed unneeded for most netCDF datasets. On Wed, Nov 29, 2017 at 8:18 PM Ryan Abernathey notifications@github.com wrote:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347983854 | https://github.com/pydata/xarray/pull/1528#issuecomment-347983854 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0Nzk4Mzg1NA== | mrocklin 306380 | 2017-11-29T20:19:37Z | 2017-11-29T20:19:37Z | MEMBER |
Is it possible to add one of these filters to XArray's default use of Zarr? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347983448 | https://github.com/pydata/xarray/pull/1528#issuecomment-347983448 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0Nzk4MzQ0OA== | rabernat 1197350 | 2017-11-29T20:18:08Z | 2017-11-29T20:18:08Z | MEMBER | Right now I am in a dilemma over how to move forward. Fixing this string encoding issue will require some serious hacks to cf encoding. If I do this before #1087 is finished, it will be a waste of time (and a pain). On the other hand #1087 could take a long time, since it is a major refactor itself. Is there some way to punt on the multi-length string encoding for now? We could just error if such variables are present. This would allow us to get the experimental zarr backend out into the wild. FWIW, none of the datasets I want to use this with actually have any string data variables at all. I believe 95% of netcdf datasets are just regular numbers. This is an edge case. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347981682 | https://github.com/pydata/xarray/pull/1528#issuecomment-347981682 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0Nzk4MTY4Mg== | mrocklin 306380 | 2017-11-29T20:11:25Z | 2017-11-29T20:11:25Z | MEMBER | FWIW my vote is for msgpack over pickle for both performance and cross-language reasons |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347351224 | https://github.com/pydata/xarray/pull/1528#issuecomment-347351224 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM1MTIyNA== | shoyer 1217238 | 2017-11-27T22:32:47Z | 2017-11-28T07:51:31Z | MEMBER |
Agreed! I wonder why zarr doesn't have a UTF-8 variable length string type (https://github.com/alimanfoo/zarr/issues/206) -- that would feel like the obvious first choice for encoding this data. That said, xarary should be able to use fixed-length bytes just fine, doing UTF-8 encoding/decoding on the fly. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347385269 | https://github.com/pydata/xarray/pull/1528#issuecomment-347385269 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM4NTI2OQ== | alimanfoo 703554 | 2017-11-28T01:36:29Z | 2017-11-28T01:49:24Z | CONTRIBUTOR | FWIW I think the best option at the moment is to make sure you add either Pickle or MsgPack filter for any zarr array with an object dtype. BTW I was thinking that zarr should automatically add one of these filters any time someone creates an array with an object dtype, to avoid them hitting the pointer issue. If you have any thoughts on best solution drop them here: https://github.com/alimanfoo/zarr/issues/208 |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347382612 | https://github.com/pydata/xarray/pull/1528#issuecomment-347382612 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM4MjYxMg== | rabernat 1197350 | 2017-11-28T01:21:34Z | 2017-11-28T01:21:34Z | MEMBER |
Do you think this persistence could affect xarray's tests? The way the tests work is via a context manager, like this
Do we need to add an extra step after |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347381865 | https://github.com/pydata/xarray/pull/1528#issuecomment-347381865 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM4MTg2NQ== | rabernat 1197350 | 2017-11-28T01:16:58Z | 2017-11-28T01:16:58Z | MEMBER |
Perhaps zarr should raise an error when assigning |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347381734 | https://github.com/pydata/xarray/pull/1528#issuecomment-347381734 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM4MTczNA== | alimanfoo 703554 | 2017-11-28T01:16:07Z | 2017-11-28T01:16:07Z | CONTRIBUTOR | When still in the original interpreter session, all the objects still exist in memory, so all the pointers stored in the array are still valid. Restart the session and the objects are gone and the pointers are invalid. On Tue, Nov 28, 2017 at 1:14 AM, Alistair Miles alimanfoo@googlemail.com wrote:
-- Alistair Miles Head of Epidemiological Informatics Centre for Genomics and Global Health http://cggh.org Big Data Institute Building Old Road Campus Roosevelt Drive Oxford OX3 7LF United Kingdom Phone: +44 (0)1865 743596 Email: alimanfoo@googlemail.com Web: http://a http://purl.org/net/alimanlimanfoo.github.io/ Twitter: https://twitter.com/alimanfoo |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347381500 | https://github.com/pydata/xarray/pull/1528#issuecomment-347381500 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM4MTUwMA== | alimanfoo 703554 | 2017-11-28T01:14:42Z | 2017-11-28T01:14:42Z | CONTRIBUTOR | Try exiting and restarting the interpreter, then running: zgs = zarr.open_group(store='zarr_directory') zgs.x[:] On Tue, Nov 28, 2017 at 1:10 AM, Ryan Abernathey notifications@github.com wrote:
-- Alistair Miles Head of Epidemiological Informatics Centre for Genomics and Global Health http://cggh.org Big Data Institute Building Old Road Campus Roosevelt Drive Oxford OX3 7LF United Kingdom Phone: +44 (0)1865 743596 Email: alimanfoo@googlemail.com Web: http://a http://purl.org/net/alimanlimanfoo.github.io/ Twitter: https://twitter.com/alimanfoo |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347380750 | https://github.com/pydata/xarray/pull/1528#issuecomment-347380750 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM4MDc1MA== | rabernat 1197350 | 2017-11-28T01:10:01Z | 2017-11-28T01:10:10Z | MEMBER |
@alimanfoo: the following also seems to works with directory store
This seems to contradict your statement above. What am I missing? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347363503 | https://github.com/pydata/xarray/pull/1528#issuecomment-347363503 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzM2MzUwMw== | alimanfoo 703554 | 2017-11-27T23:27:41Z | 2017-11-27T23:27:41Z | CONTRIBUTOR | For variable length strings (or any array with an object dtype) zarr needs a filter that can encode and pack the strings into a single buffer, except in the special case where the data are being stored in-memory (as in your first example). The filter has to be specified manually, some examples here: http://zarr.readthedocs.io/en/master/tutorial.html#string-arrays. There are two codecs currently in numcodecs that can do this, one is Pickle, the other is MsgPack. I haven't done any benchmarking of data size or encoding speed, but MsgPack may be preferable because it's more portable. There was some discussion a while back about creating a codec that handles variable-length strings by encoding via UTF8 then concatenating encoded bytes and lengths or offsets, IIRC similar to Arrow, and maybe even creating a special "text" dtype that inserts this filter automatically so you don't have to add it manually. But there hasn't been a strong motivation so far. On Mon, Nov 27, 2017 at 10:32 PM, Stephan Hoyer notifications@github.com wrote:
-- Alistair Miles Head of Epidemiological Informatics Centre for Genomics and Global Health http://cggh.org Big Data Institute Building Old Road Campus Roosevelt Drive Oxford OX3 7LF United Kingdom Phone: +44 (0)1865 743596 Email: alimanfoo@googlemail.com Web: http://a http://purl.org/net/alimanlimanfoo.github.io/ Twitter: https://twitter.com/alimanfoo |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
347323043 | https://github.com/pydata/xarray/pull/1528#issuecomment-347323043 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NzMyMzA0Mw== | rabernat 1197350 | 2017-11-27T20:48:35Z | 2017-11-27T20:53:28Z | MEMBER | After a few more tweaks, this is now quite close to passing all the The remaining issues are all related to the encoding of strings. Basically, zarr's handling of strings:
http://zarr.readthedocs.io/en/latest/tutorial.html?highlight=strings#string-arrays
is considerably different from netCDF's. Because Consider the following direct creation of a variable length string in zarr:
It seems we can encode variable-length strings into objects just fine. ( However, after passing through xarray's cf encoding, this no longer works:
Here is everything that happens in The challenge now is to figure out which parts of this we need to bypass for zarr and how to implement that bypassing. Overall, I find the At this point, I would appreciate some input from an encoding expert before I go refactoring stuff. edit: The actual tests that fail are |
{ "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345778844 | https://github.com/pydata/xarray/pull/1528#issuecomment-345778844 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTc3ODg0NA== | mrocklin 306380 | 2017-11-20T18:05:25Z | 2017-11-20T18:05:25Z | MEMBER |
It's so nice when well-designed things come together and just work as planned :) |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345770374 | https://github.com/pydata/xarray/pull/1528#issuecomment-345770374 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTc3MDM3NA== | martindurant 6042212 | 2017-11-20T17:37:01Z | 2017-11-20T17:37:01Z | CONTRIBUTOR | This is, of course, by design :) I imagine there is much that could be done to optimise performance, but for fewer, larger chunks, it should be pretty good. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345619509 | https://github.com/pydata/xarray/pull/1528#issuecomment-345619509 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTYxOTUwOQ== | alimanfoo 703554 | 2017-11-20T08:07:44Z | 2017-11-20T08:07:44Z | CONTRIBUTOR | Fantastic! On Monday, November 20, 2017, Matthew Rocklin notifications@github.com wrote:
-- Alistair Miles Head of Epidemiological Informatics Centre for Genomics and Global Health http://cggh.org Big Data Institute Building Old Road Campus Roosevelt Drive Oxford OX3 7LF United Kingdom Phone: +44 (0)1865 743596 Email: alimanfoo@googlemail.com Web: http://a http://purl.org/net/alimanlimanfoo.github.io/ Twitter: https://twitter.com/alimanfoo |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345575240 | https://github.com/pydata/xarray/pull/1528#issuecomment-345575240 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTU3NTI0MA== | mrocklin 306380 | 2017-11-20T02:28:07Z | 2017-11-20T02:28:07Z | MEMBER | That is, indeed, quite exciting. Also exciting is that I was able to look at and compute on your data easily. ```python In [1]: import zarr In [2]: import gcsfs In [3]: fs = gcsfs.GCSFileSystem(project='pangeo-181919') In [4]: gcsmap = gcsfs.mapping.GCSMap('zarr_store_test', gcs=fs, check=True, create=False) In [5]: import xarray as xr In [6]: ds_gcs = xr.open_zarr(gcsmap, mode='r') In [7]: ds_gcs Out[7]: <xarray.Dataset> Dimensions: (x: 200, y: 100) Coordinates: * x (x) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ... * y (y) int64 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 ... Data variables: bar (x) float64 dask.array<shape=(200,), chunksize=(40,)> foo (y, x) float32 dask.array<shape=(100, 200), chunksize=(50, 40)> Attributes: array_atr: [1, 2] some_attr: copana In [8]: ds_gcs.sum() Out[8]: <xarray.Dataset> Dimensions: () Data variables: bar float64 dask.array<shape=(), chunksize=()> foo float32 dask.array<shape=(), chunksize=()> In [9]: ds_gcs.sum().compute() Out[9]: <xarray.Dataset> Dimensions: () Data variables: bar float64 0.0 foo float32 20000.0 ``` |
{ "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 1, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345574445 | https://github.com/pydata/xarray/pull/1528#issuecomment-345574445 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTU3NDQ0NQ== | rabernat 1197350 | 2017-11-20T02:21:08Z | 2017-11-20T02:21:08Z | MEMBER | Those following this thread will probably be very excited to learn that the following code works with my zarr_backend branch:
I never doubted this would be possible, but seeing it in action is quite exciting! |
{ "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 1, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345128506 | https://github.com/pydata/xarray/pull/1528#issuecomment-345128506 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTEyODUwNg== | jhamman 2443309 | 2017-11-17T02:38:41Z | 2017-11-17T02:38:41Z | MEMBER | @rabernat - It might a little but we'll sort it out. See https://github.com/rabernat/xarray/pull/3. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345126452 | https://github.com/pydata/xarray/pull/1528#issuecomment-345126452 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTEyNjQ1Mg== | rabernat 1197350 | 2017-11-17T02:24:56Z | 2017-11-17T02:24:56Z | MEMBER | @jhamman would it screw you up if I pushed a few commits tonight? I won’t touch the ZarrArrayWrapper. But I figured out how to fix auto_chunk. Sent from my iPhone
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345104713 | https://github.com/pydata/xarray/pull/1528#issuecomment-345104713 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTEwNDcxMw== | mrocklin 306380 | 2017-11-17T00:12:01Z | 2017-11-17T00:12:01Z | MEMBER | Hooray for standard interfaces! |
{ "total_count": 1, "+1": 0, "-1": 0, "laugh": 1, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345104440 | https://github.com/pydata/xarray/pull/1528#issuecomment-345104440 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTEwNDQ0MA== | martindurant 6042212 | 2017-11-17T00:10:19Z | 2017-11-17T00:10:19Z | CONTRIBUTOR |
|
{ "total_count": 2, "+1": 2, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345101150 | https://github.com/pydata/xarray/pull/1528#issuecomment-345101150 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTEwMTE1MA== | mrocklin 306380 | 2017-11-16T23:52:48Z | 2017-11-16T23:52:48Z | MEMBER | The gcsfs library also provides a MutableMapping for Google Cloud Storage. The dask.distributed library now also provides a distributed lock for synchronization, if necessary though in practice we should just rechunk the dask.array before writing. |
{ "total_count": 1, "+1": 1, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345091139 | https://github.com/pydata/xarray/pull/1528#issuecomment-345091139 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTA5MTEzOQ== | shoyer 1217238 | 2017-11-16T23:02:14Z | 2017-11-16T23:02:14Z | MEMBER |
We will need to write new adapter code to map xarray's explicit indexer classes onto the appropriate zarr methods, e.g., ```python def getitem(self, key): array = self.get_arraay() if isinstance(key, BasicIndexer): return array[key.tuple] elif isinstance(key, VectorizedIndexer): return array.vindex[_replace_slices_with_arrays(key.tuple, self.shape)] else: assert isinstance(key, OuterIndexer) return array.oindex[key.tuple] untested, but I think this does the appropriate shape munging to make slicesappear as the last axes of the result arraydef _replace_slice_with_arrays(key, shape): num_slices = sum(1 for k in key if isinstance(k, slice)) num_arrays = len(shape) - num_slices new_key = [] slice_count = 0 for k, size in zip(key, shape): if isinstance(k, slice): array = np.arange(*k.indices(size)) sl = [np.newaxis] * len(shape) sl[num_arrays + slice_count] = np.newaxis k = array[sl] slice_count += 1 else: assert isinstance(k, numpy.ndarray) k = k[(slice(None),) * num_arrays + (np.newaxis,) * num_slices] new_key.append(k) return tuple(new_key) ``` |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345080945 | https://github.com/pydata/xarray/pull/1528#issuecomment-345080945 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTA4MDk0NQ== | alimanfoo 703554 | 2017-11-16T22:18:04Z | 2017-11-16T22:18:04Z | CONTRIBUTOR | Re different zarr storage backends, main options are plain dict, DirectoryStore, ZipStore, and there's a new DBMStore class just merged which enables storage in any DBM-style database (e.g., Berkeley DB). ZipStore has some constraints because of how zip files work, you can't really replace an entry in a zip file which means anything that writes the same array chunk more than once will generate warnings. Dask's S3Map should also work, I haven't tried it and obviously not ideal for unit tests but I'd be interested if you get any experience with it. Re different combinations of zarr and dask chunks, it can be thread safe even if chunks are not aligned, just need to pass a synchronizer when instantiating the array or group. Zarr has a ThreadSynchronizer class which can be used for thread-based parallelism. If a synchronizer is provided, it is used to lock each chunk individually during write operations. More info here. Re fill values, zarr has a native concept of fill value for each array, with the fill value stored as part of the array metadata. Array metadata are stored as JSON and I recently merged a fix so that a bytes fill values could be used (via base64 encoding). I believe the netcdf way is to store fill value separately as value of "_FillValue" attribute? You could do this with zarr but user attributes are also JSON and so you would need to do your own encoding/decoding. But if possible I'd suggest using the native zarr fill_value support as it handles bytes fill value encoding and also checks to ensure fill values are valid wrt the array dtype. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345034208 | https://github.com/pydata/xarray/pull/1528#issuecomment-345034208 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTAzNDIwOA== | rabernat 1197350 | 2017-11-16T19:22:01Z | 2017-11-16T19:22:01Z | MEMBER | Some things I would like to add to the zarr test suite:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345030848 | https://github.com/pydata/xarray/pull/1528#issuecomment-345030848 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTAzMDg0OA== | rabernat 1197350 | 2017-11-16T19:10:31Z | 2017-11-16T19:10:31Z | MEMBER |
Great! If you use the latest zarr master, you should get the same test results as this travis build: https://travis-ci.org/pydata/xarray/jobs/301606996 There are two outstanding failures related to encoding ( The biggest problem is that, for reasons I don't understand, my "auto-chunking" behavior does not work (this is covered by the only zarr-specific test method: |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
345026224 | https://github.com/pydata/xarray/pull/1528#issuecomment-345026224 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NTAyNjIyNA== | jhamman 2443309 | 2017-11-16T18:53:42Z | 2017-11-16T18:53:42Z | MEMBER | @rabernat - FYI: I'm playing with your branch a bit today. @shoyer and @rabernat, can we brainstorm what a ```Python class ZarrArrayWraper(BackendArray): def init(self, variable_name, datastore): self.datastore = datastore self.variable_name = variable_name array = self.get_array() self.shape = array.shape self.dtype = np.dtype(array.dtype.kind + str(array.dtype.itemsize))
``` |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
344040853 | https://github.com/pydata/xarray/pull/1528#issuecomment-344040853 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NDA0MDg1Mw== | rabernat 1197350 | 2017-11-13T20:04:12Z | 2017-11-13T20:04:12Z | MEMBER | 😬 that's my punishment for being slow! |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
344040250 | https://github.com/pydata/xarray/pull/1528#issuecomment-344040250 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDM0NDA0MDI1MA== | shoyer 1217238 | 2017-11-13T20:02:03Z | 2017-11-13T20:02:03Z | MEMBER | @rabernat sorry for the churn here, but you are also probably going to need to update after the explicit indexing changes in #1705. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
339897936 | https://github.com/pydata/xarray/pull/1528#issuecomment-339897936 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzOTg5NzkzNg== | alimanfoo 703554 | 2017-10-27T07:42:34Z | 2017-10-27T07:42:34Z | CONTRIBUTOR | Suggest testing against GitHub master, there are a few other issues I'd like to work through before next release. On Thu, 26 Oct 2017 at 23:07, Ryan Abernathey notifications@github.com wrote:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
339815147 | https://github.com/pydata/xarray/pull/1528#issuecomment-339815147 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzOTgxNTE0Nw== | rabernat 1197350 | 2017-10-26T22:07:10Z | 2017-10-26T22:07:10Z | MEMBER | Fantastic! Are you planning a release any time soon? If not we can set up to test against the github master. Sent from my iPhone
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
339800443 | https://github.com/pydata/xarray/pull/1528#issuecomment-339800443 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzOTgwMDQ0Mw== | alimanfoo 703554 | 2017-10-26T21:04:17Z | 2017-10-26T21:04:17Z | CONTRIBUTOR | Just to say, support for 0d arrays, and for arrays with one or more zero-length dimensions, is in zarr master. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
335186616 | https://github.com/pydata/xarray/pull/1528#issuecomment-335186616 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNTE4NjYxNg== | alimanfoo 703554 | 2017-10-09T15:07:29Z | 2017-10-09T17:23:21Z | CONTRIBUTOR | I'm on paternity leave for the next 2 weeks, then will be catching up for a couple of weeks I expect. May be able to merge straightforward PRs but will have limited bandwidth. |
{ "total_count": 3, "+1": 0, "-1": 0, "laugh": 0, "hooray": 3, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
335204883 | https://github.com/pydata/xarray/pull/1528#issuecomment-335204883 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNTIwNDg4Mw== | rabernat 1197350 | 2017-10-09T16:09:50Z | 2017-10-09T16:09:50Z | MEMBER |
Congratulations! If you could just merge alimanfoo/zarr#154, it would really help us move forward. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
335162205 | https://github.com/pydata/xarray/pull/1528#issuecomment-335162205 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNTE2MjIwNQ== | rabernat 1197350 | 2017-10-09T13:43:49Z | 2017-10-09T13:43:49Z | MEMBER |
Does this include merging PRs? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
335030993 | https://github.com/pydata/xarray/pull/1528#issuecomment-335030993 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNTAzMDk5Mw== | alimanfoo 703554 | 2017-10-08T19:17:27Z | 2017-10-08T23:37:47Z | CONTRIBUTOR | FWIW I think some JSON encoders for attributes would ultimately be a useful addition to zarr, but I won't be able to put any effort into zarr in the next month, so workarounds in xarray sounds like a good idea for now. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
335027491 | https://github.com/pydata/xarray/pull/1528#issuecomment-335027491 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNTAyNzQ5MQ== | rabernat 1197350 | 2017-10-08T18:23:50Z | 2017-10-08T18:23:50Z | MEMBER |
My impression is that zarr development is moving conservatively, so we would be better off finding workarounds in xarray. @shoyer: where in the code would you recommend putting this logic? It seems like part of encoding / decoding to me. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
334981929 | https://github.com/pydata/xarray/pull/1528#issuecomment-334981929 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNDk4MTkyOQ== | rabernat 1197350 | 2017-10-08T04:16:58Z | 2017-10-08T18:21:30Z | MEMBER | There are two zarr issues that are causing some tests to fail:
Most of the failures of tests inherited from |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
335015485 | https://github.com/pydata/xarray/pull/1528#issuecomment-335015485 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNTAxNTQ4NQ== | shoyer 1217238 | 2017-10-08T15:46:36Z | 2017-10-08T15:46:36Z | MEMBER | For serializing attributes, the easiest fix is to call |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
334982373 | https://github.com/pydata/xarray/pull/1528#issuecomment-334982373 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNDk4MjM3Mw== | rabernat 1197350 | 2017-10-08T04:31:02Z | 2017-10-08T04:31:09Z | MEMBER | I worked on this on the plane back from Seattle. Yay for having no internet access! Would appreciate feedback on the questions raised above from @shoyer, @jhamman, and anyone else with backend expertise. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
334633708 | https://github.com/pydata/xarray/pull/1528#issuecomment-334633708 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNDYzMzcwOA== | rabernat 1197350 | 2017-10-06T01:15:05Z | 2017-10-06T01:15:05Z | MEMBER | Here is where we are at with the Zarr backend tests
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
334633152 | https://github.com/pydata/xarray/pull/1528#issuecomment-334633152 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNDYzMzE1Mg== | rabernat 1197350 | 2017-10-06T01:10:29Z | 2017-10-06T01:10:29Z | MEMBER | With @jhamman's help, I just made a little progress on this. We now have a bare bones test suite for the zarr backend. This is very helpful for revealing where more work is needed: encoding. So the next step is to seriously confront that issue. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
334316122 | https://github.com/pydata/xarray/pull/1528#issuecomment-334316122 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzNDMxNjEyMg== | jhamman 2443309 | 2017-10-04T23:14:58Z | 2017-10-04T23:14:58Z | MEMBER | @rabernat - testing should be fully functional now. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
333579128 | https://github.com/pydata/xarray/pull/1528#issuecomment-333579128 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzMzU3OTEyOA== | jhamman 2443309 | 2017-10-02T15:58:05Z | 2017-10-02T15:58:05Z | MEMBER | @rabernat - re backends testing, #1557 is pretty close. I can wrap it up this week. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
333400272 | https://github.com/pydata/xarray/pull/1528#issuecomment-333400272 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzMzQwMDI3Mg== | martindurant 6042212 | 2017-10-01T19:26:22Z | 2017-10-01T19:26:22Z | CONTRIBUTOR | I have not done anything, I'm afraid, since posting my commit, the content of which is just an example of how you might pass parameters down to zarr, and a test-case which shows that the basic data is round-tripping properly, but actually the dataset does not come back with the same structure as it started off. We can loop back and decide where to go from here. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
333336320 | https://github.com/pydata/xarray/pull/1528#issuecomment-333336320 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMzMzMzNjMyMA== | rabernat 1197350 | 2017-09-30T21:13:48Z | 2017-09-30T21:13:48Z | MEMBER | @martindurant: I may have some time to get back to working on this next week. (Especially if @jhamman can help me sort out the backend testing.) What is the status of your branch? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
327901739 | https://github.com/pydata/xarray/pull/1528#issuecomment-327901739 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNzkwMTczOQ== | martindurant 6042212 | 2017-09-07T19:36:15Z | 2017-09-07T19:36:15Z | CONTRIBUTOR | @shoyer , is https://github.com/martindurant/xarray/commit/6c1fb6b76ebba862a1c5831210ce026160da0065 a reasonable start ? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
327900874 | https://github.com/pydata/xarray/pull/1528#issuecomment-327900874 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNzkwMDg3NA== | shoyer 1217238 | 2017-09-07T19:32:41Z | 2017-09-07T19:32:41Z | MEMBER | @rabernat indeed, the backend tests are not terribly well organized right now. Probably the place to start is to inherit from |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
327849640 | https://github.com/pydata/xarray/pull/1528#issuecomment-327849640 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNzg0OTY0MA== | rabernat 1197350 | 2017-09-07T16:17:13Z | 2017-09-07T16:17:13Z | MEMBER | I am stuck on figuring out how to develop a new test case for this. (It doesn't help that #1531 is messing up the backend tests.) If @shoyer can give us a few hints about how to best implement a test class (i.e. what to subclass, etc.), I think that could jumpstart testing and move the PR forward. I welcome contributions from others such as @martindurant on this. I won't have much time in the near future, since a new semester just dropped on me like a load of bricks. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
327833777 | https://github.com/pydata/xarray/pull/1528#issuecomment-327833777 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNzgzMzc3Nw== | martindurant 6042212 | 2017-09-07T15:23:31Z | 2017-09-07T15:23:31Z | CONTRIBUTOR | @rabernat , is there anything I can do to help push this along? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325813339 | https://github.com/pydata/xarray/pull/1528#issuecomment-325813339 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTgxMzMzOQ== | alimanfoo 703554 | 2017-08-29T21:43:48Z | 2017-08-29T21:43:48Z | CONTRIBUTOR | On Tuesday, August 29, 2017, Ryan Abernathey notifications@github.com wrote:
-- Alistair Miles Head of Epidemiological Informatics Centre for Genomics and Global Health http://cggh.org Big Data Institute Building Old Road Campus Roosevelt Drive Oxford OX3 7LF United Kingdom Phone: +44 (0)1865 743596 Email: alimanfoo@googlemail.com Web: http://a http://purl.org/net/alimanlimanfoo.github.io/ Twitter: https://twitter.com/alimanfoo |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325742232 | https://github.com/pydata/xarray/pull/1528#issuecomment-325742232 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTc0MjIzMg== | shoyer 1217238 | 2017-08-29T17:50:04Z | 2017-08-29T17:50:04Z | MEMBER |
The only advantage here would be for non-xarray users, who could use zarr to do this decoding/encoding automatically. For what it's worth, the implementation of scale offsets in xarray looks basically equivalent to what's done in zarr. I don't think there's a performance difference either way.
If you use chunks, I believe HDF5/NetCDF4 do the same thing, e.g., ``` In [10]: with h5py.File('one-chunk.h5') as f: f.create_dataset('foo', (100, 100), chunks=(100, 100)) In [11]: with h5py.File('many-chunk.h5') as f: f.create_dataset('foo', (100000, 100000), chunks=(100, 100)) In [12]: ls -l | grep chunk.h5 -rw-r--r-- 1 shoyer eng 1400 Aug 29 10:48 many-chunk.h5 -rw-r--r-- 1 shoyer eng 1400 Aug 29 10:48 one-chunk.h5 ``` (Note the same file-size) |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325738019 | https://github.com/pydata/xarray/pull/1528#issuecomment-325738019 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTczODAxOQ== | rabernat 1197350 | 2017-08-29T17:35:09Z | 2017-08-29T17:35:09Z | MEMBER | One path forward for now would be to ignore the filters like If we think there is an advantage to using the zarr native filters, that could be added via a future PR once we have the basic backend working. @alimanfoo: when do you anticipate the 2.2 zarr release to happen? Will the API change significantly? If so, I will wait for that to move forward here. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325729013 | https://github.com/pydata/xarray/pull/1528#issuecomment-325729013 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTcyOTAxMw== | alimanfoo 703554 | 2017-08-29T17:02:41Z | 2017-08-29T17:02:41Z | CONTRIBUTOR | FWIW all filter (codec) classes have been migrated from zarr to a separate packaged called numcodecs and will be imported from there in the next (2.2) zarr release. Here is FixedScaleOffset. Implementation is basic numpy, probably some room for optimization. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325728378 | https://github.com/pydata/xarray/pull/1528#issuecomment-325728378 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTcyODM3OA== | martindurant 6042212 | 2017-08-29T17:00:29Z | 2017-08-29T17:00:29Z | CONTRIBUTOR | A further rather big advantage in zarr that I'm not aware of in cdf/hdf (I may be wrong) is not just null values, but not having a given block be written to disc at all if it only contains null data. This probably meshes perfectly well with most user's understanding of missing data/fill value. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325727354 | https://github.com/pydata/xarray/pull/1528#issuecomment-325727354 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTcyNzM1NA== | martindurant 6042212 | 2017-08-29T16:57:10Z | 2017-08-29T16:57:10Z | CONTRIBUTOR | Worth pointing out here, that the zarr filter-set is extensible (I suppose hdf5 is too, but I don't think this is ever done in practice), but I don't think it makes any particular claims to performance. I think both of the options above are reasonable, and there is no particular reason to exclude either: a zarr variable could look to xarray like floats but actually be stored as ints (i.e., arguments are passed to zarr), or it could look like ints which xarray expects to inflate to floats (i.e., stored as an attribute). I mean, if a user stores a float variable, but includes kwargs to zarr for scale/filter (or any other filter arguments), we should make no attempt to interrupt that. The only question is, if the user wishes to apply scale/offset in xarray, which is their most likely intention? I would guess the latter, compute in xarray and use attributes, since xarray users probably don't know about zarr and its filters. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325727280 | https://github.com/pydata/xarray/pull/1528#issuecomment-325727280 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTcyNzI4MA== | alimanfoo 703554 | 2017-08-29T16:56:55Z | 2017-08-29T16:56:55Z | CONTRIBUTOR | Following this with interest. Regarding autoclose, just to confirm that zarr doesn't really have any notion of whether something is open or closed. When using the DirectoryStore storage class (most common use case I imagine), all files are automatically closed, nothing is kept open. There are some storage classes (e.g., ZipStore) that do require an explicit close call to finalise the file on disk if you have been writing data, but I think you can ignore this in xarray and leave it up to the user to manage this themselves. Out of interest, @shoyer do you still think there would be value in writing a wrapper for zarr analogous to h5netcdf? Or does this PR provide all the necessary functionality? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325723577 | https://github.com/pydata/xarray/pull/1528#issuecomment-325723577 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTcyMzU3Nw== | shoyer 1217238 | 2017-08-29T16:43:58Z | 2017-08-29T16:44:25Z | MEMBER |
Yes, exactly.
Typically, we store things in encoding that are attributes on the underlying NetCDF file, but no longer make sense to describe the decoded data. For example:
- On the file,
Currently, we assume that stores never do this, and always handle it ourselves. We might need a special exception for zarr and scale/offset encoding.
Maybe, though again it will probably need slightly customized conventions for writing data (if we let zarr handling scale/offset encoding).
We have two options:
1. Handle it all in xarray via the machinery in I think (2) would be the preferred way to do this. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325716892 | https://github.com/pydata/xarray/pull/1528#issuecomment-325716892 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTcxNjg5Mg== | shoyer 1217238 | 2017-08-29T16:19:57Z | 2017-08-29T16:19:57Z | MEMBER | @rabernat I think this is #1531 -- |
{ "total_count": 1, "+1": 0, "-1": 0, "laugh": 1, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325690352 | https://github.com/pydata/xarray/pull/1528#issuecomment-325690352 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTY5MDM1Mg== | rabernat 1197350 | 2017-08-29T14:54:53Z | 2017-08-29T14:54:53Z | MEMBER | I am now trying to understand the backend test suite structure. Can someone explain to me why so many tests are skipped? For example, if I run
I get ``` ================================================== test session starts ================================================== platform darwin -- Python 3.6.1, pytest-3.0.7, py-1.4.33, pluggy-0.4.0 -- /Users/rpa/anaconda/bin/python cachedir: .cache rootdir: /Users/rpa/RND/Public/xarray, inifile: setup.cfg plugins: cov-2.5.1 collected 683 items xarray/tests/test_backends.py::GenericNetCDFDataTest::test_coordinates_encoding SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_cross_engine_read_write_netcdf3 PASSED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_dataset_caching SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_dataset_compute SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_default_fill_value SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_encoding_kwarg SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_encoding_same_dtype SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_encoding_unlimited_dims PASSED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_engine PASSED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_invalid_dataarray_names_raise SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_load SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_orthogonal_indexing PASSED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_pickle SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_pickle_dataarray SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_None_variable SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_boolean_dtype SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_coordinates SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_datetime_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_endian SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_example_1_netcdf SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_float64_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_mask_and_scale SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_object_dtype SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_string_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_strings_with_fill_value SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_test_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_roundtrip_timedelta_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_write_store PASSED xarray/tests/test_backends.py::GenericNetCDFDataTest::test_zero_dimensional_variable SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_coordinates_encoding SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_cross_engine_read_write_netcdf3 PASSED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_dataset_caching SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_dataset_compute SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_default_fill_value SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_encoding_kwarg SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_encoding_same_dtype SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_encoding_unlimited_dims PASSED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_engine PASSED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_invalid_dataarray_names_raise SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_load SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_orthogonal_indexing PASSED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_pickle SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_pickle_dataarray SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_None_variable SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_boolean_dtype SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_coordinates SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_datetime_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_endian SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_example_1_netcdf SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_float64_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_mask_and_scale SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_object_dtype SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_string_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_strings_with_fill_value SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_test_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_roundtrip_timedelta_data SKIPPED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_write_store PASSED xarray/tests/test_backends.py::GenericNetCDFDataTestAutocloseTrue::test_zero_dimensional_variable SKIPPED ================================================ short test summary info ================================================ SKIP [2] xarray/tests/test_backends.py:382: requires pynio SKIP [2] xarray/tests/test_backends.py:214: requires pynio SKIP [2] xarray/tests/test_backends.py:178: requires pynio SKIP [2] xarray/tests/test_backends.py:468: requires pynio SKIP [2] xarray/tests/test_backends.py:439: requires pynio SKIP [2] xarray/tests/test_backends.py:490: requires pynio SKIP [2] xarray/tests/test_backends.py:428: requires pynio SKIP [2] xarray/tests/test_backends.py:145: requires pynio SKIP [2] xarray/tests/test_backends.py:197: requires pynio SKIP [2] xarray/tests/test_backends.py:207: requires pynio SKIP [2] xarray/tests/test_backends.py:230: requires pynio SKIP [2] xarray/tests/test_backends.py:311: requires pynio SKIP [2] xarray/tests/test_backends.py:300: requires pynio SKIP [2] xarray/tests/test_backends.py:271: requires pynio SKIP [2] xarray/tests/test_backends.py:409: requires pynio SKIP [2] xarray/tests/test_backends.py:291: requires pynio SKIP [2] xarray/tests/test_backends.py:286: requires pynio SKIP [2] xarray/tests/test_backends.py:362: requires pynio SKIP [2] xarray/tests/test_backends.py:235: requires pynio SKIP [2] xarray/tests/test_backends.py:264: requires pynio SKIP [2] xarray/tests/test_backends.py:334: requires pynio SKIP [2] xarray/tests/test_backends.py:139: requires pynio SKIP [2] xarray/tests/test_backends.py:280: requires pynio SKIP [2] xarray/tests/test_backends.py:109: requires pynio ``` Those line numbers refer to all of the skipped methods. Why should I need pynio to run those tests? It looks like the same thing is happening on travis: https://travis-ci.org/pydata/xarray/jobs/268805771#L1527 Maybe @pwolfram understands this stuff? |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325660754 | https://github.com/pydata/xarray/pull/1528#issuecomment-325660754 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTY2MDc1NA== | rabernat 1197350 | 2017-08-29T13:18:33Z | 2017-08-29T13:18:33Z | MEMBER |
Is the goal here to be able to round-trip the file, such that calling I don't understand how encoding interacts with attributes? When is something an attribute vs. an encoding (
Does this mean that my Regarding encoding, zarr has its own internal mechanism for encoding, which it calls "filters", that closely resemble some of the CF encoding options. For example the I don't yet understand how to make these elements work together properly, for example, do avoid applying the scale / offset function twice, as I mentioned above. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325525827 | https://github.com/pydata/xarray/pull/1528#issuecomment-325525827 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTUyNTgyNw== | shoyer 1217238 | 2017-08-29T01:14:05Z | 2017-08-29T01:14:05Z | MEMBER |
Yes, probably, if we want to handle netcdf conventions for times, fill values and scaling.
This would be nice! But it's also a bigger issue (will look for the number, I think it's already been opened).
Still need to think about this one.
I guess we can ignore them (maybe add a warning?) -- they're not part of the zarr data model.
I don't think we need any autoclose logic at all -- zarr doesn't leave open files hanging around already. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325390391 | https://github.com/pydata/xarray/pull/1528#issuecomment-325390391 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTM5MDM5MQ== | martindurant 6042212 | 2017-08-28T15:41:08Z | 2017-08-28T15:41:08Z | CONTRIBUTOR | @rabernat : on actually looking through your code :) Happy to see you doing exactly as I felt I was not knowledgeable to do and poke xarray's guts. If I can help in any way, please let me know, although I don't have a lot of spare hours right now. |
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 | |
325226656 | https://github.com/pydata/xarray/pull/1528#issuecomment-325226656 | https://api.github.com/repos/pydata/xarray/issues/1528 | MDEyOklzc3VlQ29tbWVudDMyNTIyNjY1Ng== | rabernat 1197350 | 2017-08-27T21:42:23Z | 2017-08-27T21:42:23Z | MEMBER |
This is also part of my goal. I think all the metadata can be stored internally to zarr via attributes. There just have to be some "special" attributes that xarray hides from the user. This is the same as h5netcdf. @alimanfoo suggested this should be possible in that earlier thread:
|
{ "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 } |
WIP: Zarr backend 253136694 |
Advanced export
JSON shape: default, array, newline-delimited, object
CREATE TABLE [issue_comments] ( [html_url] TEXT, [issue_url] TEXT, [id] INTEGER PRIMARY KEY, [node_id] TEXT, [user] INTEGER REFERENCES [users]([id]), [created_at] TEXT, [updated_at] TEXT, [author_association] TEXT, [body] TEXT, [reactions] TEXT, [performed_via_github_app] TEXT, [issue] INTEGER REFERENCES [issues]([id]) ); CREATE INDEX [idx_issue_comments_issue] ON [issue_comments] ([issue]); CREATE INDEX [idx_issue_comments_user] ON [issue_comments] ([user]);
user 8