{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":198316102,"defaultBranch":"master","name":"fv3net","ownerLogin":"ai2cm","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2019-07-22T23:27:40.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/55798839?v=4","public":true,"private":false,"isOrgOwned":true},"refInfo":{"name":"","listCacheKey":"v0:1720550706.0","currentOid":""},"activityList":{"items":[{"before":"5a71d498b48d2d5c5d316958869c872eafb9c556","after":null,"ref":"refs/heads/allow-tapering-with-gpu","pushedAt":"2024-07-09T18:45:06.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"elynnwu","name":"Elynn Wu","path":"/elynnwu","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6588606?s=80&v=4"}},{"before":"b97bc25ab1fe16a82cae32122e1fe286eb6c887c","after":"79edf0dc6f7a53fba3e2d42835def4975b33662b","ref":"refs/heads/master","pushedAt":"2024-07-09T18:45:05.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"elynnwu","name":"Elynn Wu","path":"/elynnwu","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6588606?s=80&v=4"},"commit":{"message":"Tapering compatibility with cupy GPU arrays (#2403)\n\nThis PR adds compatibility for vertical tapering with cupy GPU arrays.","shortMessageHtmlLink":"Tapering compatibility with cupy GPU arrays (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2396835911\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ai2cm/fv3net/issues/2403\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ai2cm/fv3net/pull/2403/hovercard\" href=\"https://github.com/ai2cm/fv3net/pull/2403\">#2403</a>)"}},{"before":"76b5428be26aed5d550315d0bd56b0deb36c9814","after":null,"ref":"refs/heads/feature/derived-nir-vis-dif-dir-fracs","pushedAt":"2024-07-09T18:05:13.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"}},{"before":"af0ded8500c87110adcf63eddd1c45f82b1ad443","after":"b97bc25ab1fe16a82cae32122e1fe286eb6c887c","ref":"refs/heads/master","pushedAt":"2024-07-09T18:05:11.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Novelty detection compatibility with cupy GPU arrays (#2399)\n\nThis PR adds some basic infrastructure for handling Cupy arrays when\r\nusing novelty detection. Unfortunately due to some of the underlying\r\ncomponents, the novelty detection model cannot easily be moved to\r\nGPU-only calculations. So for now, we are transferring to CPU where\r\nnecessary to run the calculation.","shortMessageHtmlLink":"Novelty detection compatibility with cupy GPU arrays (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2389493730\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ai2cm/fv3net/issues/2399\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ai2cm/fv3net/pull/2399/hovercard\" href=\"https://github.com/ai2cm/fv3net/pull/2399\">#2399</a>)"}},{"before":"2a1ff7b021a6c58ffbb4eca42151dbe718f5da58","after":"5a71d498b48d2d5c5d316958869c872eafb9c556","ref":"refs/heads/allow-tapering-with-gpu","pushedAt":"2024-07-09T17:52:38.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"elynnwu","name":"Elynn Wu","path":"/elynnwu","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6588606?s=80&v=4"},"commit":{"message":"move np/cp check to the top","shortMessageHtmlLink":"move np/cp check to the top"}},{"before":"279b0783d3cb16b25b1872d8ad26a434350e93d8","after":"76b5428be26aed5d550315d0bd56b0deb36c9814","ref":"refs/heads/feature/derived-nir-vis-dif-dir-fracs","pushedAt":"2024-07-09T14:50:17.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Make num workers default congruent between the CLI and main","shortMessageHtmlLink":"Make num workers default congruent between the CLI and main"}},{"before":"3da92358260f11a3dcd9984ad5c349577ba96ef8","after":"2a1ff7b021a6c58ffbb4eca42151dbe718f5da58","ref":"refs/heads/allow-tapering-with-gpu","pushedAt":"2024-07-09T00:43:19.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"elynnwu","name":"Elynn Wu","path":"/elynnwu","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6588606?s=80&v=4"},"commit":{"message":"update cpu only cupy behavior","shortMessageHtmlLink":"update cpu only cupy behavior"}},{"before":null,"after":"3da92358260f11a3dcd9984ad5c349577ba96ef8","ref":"refs/heads/allow-tapering-with-gpu","pushedAt":"2024-07-09T00:31:13.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"elynnwu","name":"Elynn Wu","path":"/elynnwu","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6588606?s=80&v=4"},"commit":{"message":"tapering uses cupy when available","shortMessageHtmlLink":"tapering uses cupy when available"}},{"before":"48a85a7fadcf74baf48efcd92a781129e827fc1f","after":null,"ref":"refs/heads/fix/scream-ml-pr-diag-adjust","pushedAt":"2024-07-08T23:04:22.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"}},{"before":"92c797ef3952adc878221fe5afb541770d90cb71","after":"af0ded8500c87110adcf63eddd1c45f82b1ad443","ref":"refs/heads/master","pushedAt":"2024-07-08T23:04:21.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Fix bug to remove net ML moistening from SCREAM surface precip for diags (#2402)\n\nThe net moistening removal from SCREAM run output for diagnostics\r\nincorrectly subtracted the ML quantity (essentially doubling the\r\nprecip). This PR switches that removal to the correct sign.","shortMessageHtmlLink":"Fix bug to remove net ML moistening from SCREAM surface precip for di…"}},{"before":null,"after":"48a85a7fadcf74baf48efcd92a781129e827fc1f","ref":"refs/heads/fix/scream-ml-pr-diag-adjust","pushedAt":"2024-07-08T22:31:22.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Fix bug to remove net moistening","shortMessageHtmlLink":"Fix bug to remove net moistening"}},{"before":"4db5328a5640e79f6b8bab50a7cf221f04bd34cf","after":null,"ref":"refs/heads/feature/scream-rad-diags","pushedAt":"2024-07-08T21:45:55.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"}},{"before":"c37ecea26e5773f9c1c7daef421dc051dd4e2c86","after":"92c797ef3952adc878221fe5afb541770d90cb71","ref":"refs/heads/master","pushedAt":"2024-07-08T21:45:53.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Add SCREAM prescribed radiation to online diagnostic variables (#2401)\n\nAdds the prescribed/predicted radiation variables net sfc SW and net LW\r\ndown to the online diagnostics to compare to the physics rather than\r\nadding more confusion to the renaming matrix.","shortMessageHtmlLink":"Add SCREAM prescribed radiation to online diagnostic variables (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2396363338\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ai2cm/fv3net/issues/2401\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ai2cm/fv3net/pull/2401/hovercard\" href=\"https://github.com/ai2cm/fv3net/pull/2401\">#2401</a>)"}},{"before":"9430ce645dacefb5daa644825e3c54f3b86ac4ce","after":null,"ref":"refs/heads/feature/sw-fractions","pushedAt":"2024-07-08T21:45:48.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"}},{"before":"0a1c5dbeb48e413f94b67cb71e38dde840e7ad06","after":"c37ecea26e5773f9c1c7daef421dc051dd4e2c86","ref":"refs/heads/master","pushedAt":"2024-07-08T21:45:46.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Add NIR/VIS diffuse/direct fractions to derived variables (#2400)\n\nSCREAM requires some additional inputs to the surface scheme for\r\nradiation. This PR provides additional derived variables so we can train\r\nmodels to output fractions of the downwelling shortwave to allocate to\r\nNIR/VIS direct/diffuse components.","shortMessageHtmlLink":"Add NIR/VIS diffuse/direct fractions to derived variables (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2389503451\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ai2cm/fv3net/issues/2400\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ai2cm/fv3net/pull/2400/hovercard\" href=\"https://github.com/ai2cm/fv3net/pull/2400\">#2400</a>)"}},{"before":null,"after":"4db5328a5640e79f6b8bab50a7cf221f04bd34cf","ref":"refs/heads/feature/scream-rad-diags","pushedAt":"2024-07-08T19:08:58.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Add SCREAM prescribed radiation to online diagnostic variables","shortMessageHtmlLink":"Add SCREAM prescribed radiation to online diagnostic variables"}},{"before":"bbe520d864b09fae929fbdfeee79b4f7ba0fdf14","after":"279b0783d3cb16b25b1872d8ad26a434350e93d8","ref":"refs/heads/feature/derived-nir-vis-dif-dir-fracs","pushedAt":"2024-07-03T22:49:14.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Add the numpy array retreive fix for ocsvm","shortMessageHtmlLink":"Add the numpy array retreive fix for ocsvm"}},{"before":"cc5bd4c0e82a74d746fcbc4210603620fe77a7ad","after":"bbe520d864b09fae929fbdfeee79b4f7ba0fdf14","ref":"refs/heads/feature/derived-nir-vis-dif-dir-fracs","pushedAt":"2024-07-03T21:21:26.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Linting","shortMessageHtmlLink":"Linting"}},{"before":null,"after":"9430ce645dacefb5daa644825e3c54f3b86ac4ce","ref":"refs/heads/feature/sw-fractions","pushedAt":"2024-07-03T21:17:06.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Add derived variables","shortMessageHtmlLink":"Add derived variables"}},{"before":null,"after":"cc5bd4c0e82a74d746fcbc4210603620fe77a7ad","ref":"refs/heads/feature/derived-nir-vis-dif-dir-fracs","pushedAt":"2024-07-03T21:10:41.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Add batches and logging","shortMessageHtmlLink":"Add batches and logging"}},{"before":"dd3ee842ef8903c5c0630c40a5be0a6b6ab288f0","after":"85742d61021755aad54f7bf9f18370d301ca4bb7","ref":"refs/heads/feature/nir-vis-dir-dif-fractions","pushedAt":"2024-07-03T20:43:47.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Fix loader batch save call arguments","shortMessageHtmlLink":"Fix loader batch save call arguments"}},{"before":"77a44c25e6283a4a03ee36bbfeb915f6e4aebf11","after":null,"ref":"refs/heads/fix/_roundtrip_filter_dataset-attrs","pushedAt":"2024-07-03T20:36:31.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"spencerkclark","name":"Spencer Clark","path":"/spencerkclark","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6628425?s=80&v=4"}},{"before":"45954d91bf8a24e2a89a1affd2fe5aa09490e04e","after":"0a1c5dbeb48e413f94b67cb71e38dde840e7ad06","ref":"refs/heads/master","pushedAt":"2024-07-03T20:36:30.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"spencerkclark","name":"Spencer Clark","path":"/spencerkclark","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6628425?s=80&v=4"},"commit":{"message":"xtorch_harmonics: work around `xarray.merge` attribute promotion issue (#2398)\n\nWhen merging DataArrays, a known issue (pydata/xarray#5436) is that\r\nxarray will promote DataArray-level attributes to Dataset-level ones:\r\n```\r\n>>> a = xr.DataArray(0, name=\"a\", attrs={\"x\": \"y\"})\r\n>>> b = xr.DataArray(0, name=\"b\", attrs={\"m\": \"n\"})\r\n>>> xr.merge([a, b])\r\n<xarray.Dataset>\r\nDimensions:  ()\r\nData variables:\r\n    a        int64 0\r\n    b        int64 0\r\nAttributes:\r\n    x:        y\r\n```\r\nThis is generally not correct, and leads to distracting Dataset-level\r\nattributes on the results of `xtorch_harmonics.roundtrip_filter`. A\r\nsuggested workaround is to convert DataArrays to Datasets prior to\r\npassing to `merge`:\r\nhttps://github.com/pydata/xarray/issues/5436#issuecomment-1255120923.\r\nThis PR implements this suggestion and tests that DataArray-level\r\nattributes are no longer added at the Dataset level.\r\n\r\n- [x] Tests added\r\n\r\nCoverage reports (updated automatically):","shortMessageHtmlLink":"xtorch_harmonics: work around <code>xarray.merge</code> attribute promotion issue ("}},{"before":"95f9ec453d3a8d85689ccf3037d3148770be27c2","after":"dd3ee842ef8903c5c0630c40a5be0a6b6ab288f0","ref":"refs/heads/feature/nir-vis-dir-dif-fractions","pushedAt":"2024-07-03T20:28:33.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Fix loader batch save call arguments","shortMessageHtmlLink":"Fix loader batch save call arguments"}},{"before":"46a50b805b361cde19de98d6113c2e20c210c81f","after":null,"ref":"refs/heads/scream-for-merge","pushedAt":"2024-07-03T20:18:32.000Z","pushType":"branch_deletion","commitsCount":0,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"}},{"before":"fd7e6c895086b6ad2e9c65ef39317b19722c234a","after":"45954d91bf8a24e2a89a1affd2fe5aa09490e04e","ref":"refs/heads/master","pushedAt":"2024-07-03T20:18:30.000Z","pushType":"pr_merge","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Optimize diagnostics for larger HPC nodes (#2396)\n\nThis PR optimizes the online diagnostic calculation and report\r\ngeneration for use on HPC where there are typically many more cores. To\r\nachieve this I standardized some of the common aggregation functions for\r\ntaking means and switched to use of dask distributed, which has more\r\nfine-grained control over workers and memory usage. Additionally, the\r\ndiagnostic functions are batched to occur concurrently using `joblib`\r\nbut the individual dask tasks are elevated to the top-level of the\r\nscheduler to properly track individual task operations and memory usage.\r\n\r\nRefactored public API:\r\n- new flags for controlling the compute parallelism to the online\r\nprognostic run `compute.py`\r\n- `--scheduler-file` provide a dask scheduler file for an external dask\r\ndistributed cluster when doing multi-node processing\r\n- `--num-concurrent-2d-calcs` number of concurrent 2D diagnostics to\r\nexecute (i.e., diagnostic functions), defaults to 8\r\n- `--num-concurrent-3d-calcs` number of concurrent 3D diagnostics to\r\nexecute (i.e., diagnostic functions), defaults to 2\r\n\r\nSignificant internal changes:\r\n- Online diagnostics are now calculated using a dask distributed cluster\r\nwith progress visible via `localhost:8787` and parallelized using\r\nbatches of function calls with `joblib`\r\n- Generated diagnostic plots has been parallelized using `joblib`\r\n- Diurnal cycle now uses `flox` to speed up the dask groupby operation\r\n- `load_scream_data` now separates to physics-only tendencies for\r\n`due_to_scream_physics` variables, whereas previously, it included both\r\nthe nudging/ML update\r\n- `load_scream_data` now removes the ML precipitation from `PRATEsfc`\r\nsince the diagnostics expects this to be a physics only quantity\r\n\r\n\r\nCoverage reports (updated automatically):","shortMessageHtmlLink":"Optimize diagnostics for larger HPC nodes (<a class=\"issue-link js-issue-link\" data-error-text=\"Failed to load title\" data-id=\"2367298056\" data-permission-text=\"Title is private\" data-url=\"https://github.com/ai2cm/fv3net/issues/2396\" data-hovercard-type=\"pull_request\" data-hovercard-url=\"/ai2cm/fv3net/pull/2396/hovercard\" href=\"https://github.com/ai2cm/fv3net/pull/2396\">#2396</a>)"}},{"before":"4842d00ffea76d8f220412df0839387c481cc1b7","after":"46a50b805b361cde19de98d6113c2e20c210c81f","ref":"refs/heads/scream-for-merge","pushedAt":"2024-07-03T18:54:59.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Restore NaN ignore in RMS function","shortMessageHtmlLink":"Restore NaN ignore in RMS function"}},{"before":"7c2498cd5b76a2ae269a0ac28921cec6a9dbeba6","after":"95f9ec453d3a8d85689ccf3037d3148770be27c2","ref":"refs/heads/feature/nir-vis-dir-dif-fractions","pushedAt":"2024-07-03T14:37:09.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Merge branch 'scream-for-merge' into feature/nir-vis-dir-dif-fractions","shortMessageHtmlLink":"Merge branch 'scream-for-merge' into feature/nir-vis-dir-dif-fractions"}},{"before":"c04ea9d2bd6e932e2f4f8346d26cd6cfbc8fe687","after":"77a44c25e6283a4a03ee36bbfeb915f6e4aebf11","ref":"refs/heads/fix/_roundtrip_filter_dataset-attrs","pushedAt":"2024-07-03T14:21:51.000Z","pushType":"force_push","commitsCount":0,"pusher":{"login":"spencerkclark","name":"Spencer Clark","path":"/spencerkclark","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/6628425?s=80&v=4"},"commit":{"message":"Work around xarray.merge attribute promotion issue","shortMessageHtmlLink":"Work around xarray.merge attribute promotion issue"}},{"before":"8933dbf5ab146b5aa4d9917dc3f5095907d3a4a4","after":"7c2498cd5b76a2ae269a0ac28921cec6a9dbeba6","ref":"refs/heads/feature/nir-vis-dir-dif-fractions","pushedAt":"2024-07-03T14:06:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"frodre","name":"W. Andre Perkins","path":"/frodre","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/1440875?s=80&v=4"},"commit":{"message":"Adjustments for novelty detector tests","shortMessageHtmlLink":"Adjustments for novelty detector tests"}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEev-oQQA","startCursor":null,"endCursor":null}},"title":"Activity · ai2cm/fv3net"}