Upload folder using huggingface_hub

README.md

@@ -19,7 +19,7 @@ task_categories:
 # NA-SAR
 
 NA-SAR is a North America synthetic aperture radar pretraining dataset built
-from the NASA OPERA archive. It is intended for self-supervised and masked-image
+from NASA OPERA products. It is intended for self-supervised and masked-image
 pretraining of SAR foundation models.
 
 This release is an unsplit pretraining corpus. There are no train/validation/test
@@ -30,6 +30,7 @@ appropriate splits.
 ## Dataset Contents
 
 - Samples: 1,099,604
+- Shards: 91 WebDataset tar shards
 - Patch size: 128 x 128 pixels
 - RTC views: two spatial views per sample
 - Temporal RTC acquisitions: prime and secondary
@@ -37,9 +38,8 @@ appropriate splits.
 - Incidence angle: one channel per spatial view
 - InSAR phase: stored as cos(phi), sin(phi)
 - Coherence: one channel per spatial view
-- DEM elevation: one global-land normalized terrain channel per sample
-- DEM relief: one patch-local robust elevation channel per sample
-- DEM slope: one normalized slope-degree channel per sample
+- DEM elevation: raw elevation in meters
+- DEM slope: raw terrain slope in degrees
 
 Each tensor sample contains:
 
@@ -55,14 +55,16 @@ Each tensor sample contains:
 | `coh_view_0` | `(1, 128, 128)` | Coherence for view 0 |
 | `ifg_view_1` | `(2, 128, 128)` | InSAR phase for view 1 as cos/sin |
 | `coh_view_1` | `(1, 128, 128)` | Coherence for view 1 |
-| `dem` | `(1, 128, 128)` | Elevation normalized to the global-land robust range `[-100 m, 4000 m]` and clipped to `[0, 1]` |
-| `dem_relief` | `(1, 128, 128)` | Patch-local robust elevation relief using the patch p5/p95 range, clipped to `[0, 1]` |
-| `slope` | `(1, 128, 128)` | Terrain slope in degrees normalized by `[0 deg, 45 deg]` and clipped to `[0, 1]` |
+| `dem` | `(1, 128, 128)` | Raw DEM elevation in meters |
+| `slope_deg` | `(1, 128, 128)` | Raw terrain slope angle in degrees |
+
+`dem_relief` and normalized `slope` are not stored in the WebDataset. They are
+derived at loading time from the raw `dem` and `slope_deg` arrays.
 
 ## Metadata
 
-`metadata_hf.parquet` is the publishable metadata table for the sharded release.
-It includes OPERA provenance, patch geometry, quality score, polarization
+`metadata.parquet` is the publishable metadata table for the sharded release. It
+includes OPERA provenance, patch geometry, quality score, polarization
 availability, and the following sharding columns:
 
 - `sample_key`: sample key inside the WebDataset shard
@@ -72,8 +74,8 @@ availability, and the following sharding columns:
 
 The metadata is filtered to samples with `quality_score > 0.53`.
 
-`webdataset_summary.json` records the shard count and marks this release with
-`dem_extended: true`.
+`webdataset_summary.json` records the shard count and DEM storage policy. For
+this release, `dem_extended` is `"raw"` and `dem_arrays` is `["dem", "slope_deg"]`.
 
 ## Loading With Hugging Face Datasets
 
@@ -99,19 +101,25 @@ print(arrays.files)
 
 ## Loading With PyTorch
 
-The repository includes a lightweight PyTorch helper:
+The release includes `nasar_dataset.py`, a lightweight PyTorch loader for the
+local shard layout:
 
 ```python
-from nasar_dataset import NASARRTCDataset, NASARInSARDataset
+from nasar_dataset import NASARWebRTCDataset, NASARWebInSARDataset
 
-rtc_ds = NASARRTCDataset("/path/to/NA-SAR")
-insar_ds = NASARInSARDataset("/path/to/NA-SAR")
+rtc_ds = NASARWebRTCDataset("/path/to/NA-SAR-HF")
+insar_ds = NASARWebInSARDataset("/path/to/NA-SAR-HF", require_dem=True)
 ```
 
-For streaming at scale, use the WebDataset tar shards under `data/`.
-Each tar member is a compressed `.npz` tensor file named `{sample_key}.npz`.
-`NASARWebInSARDataset` returns a 6-channel InSAR view when collated as
-`[cos(phi), sin(phi), coherence, dem, dem_relief, slope]`.
+For InSAR, `NASARWebInSARDataset` returns SAR arrays plus DEM-derived channels:
+
+| Loader output key | Source array | Default normalization |
+| --- | --- | --- |
+| `dem_view_0/1` | `dem` | global elevation, `(-100 m, 4000 m) -> [0, 1]` |
+| `dem_relief_view_0/1` | `dem` | patch-local p5/p95 relief -> `[0, 1]` |
+| `slope_view_0/1` | `slope_deg` | slope degrees, `(0 deg, 45 deg) -> [0, 1]` |
+
+These normalization ranges can be changed through the loader constructor.
 
 ## Preprocessing Notes
 
@@ -132,16 +140,17 @@ zero-padded after preprocessing.
 InSAR phase is stored as cosine and sine channels instead of wrapped phase
 radians to avoid phase discontinuities at the wrap boundary.
 
-DEM channels are static for the spatial patch and are shared across both orbit
-views. Elevation values are derived from aligned Copernicus DEM patches in
-meters, then stored as normalized float32 arrays. The `dem` channel preserves an
-absolute elevation signal using a fixed global-land range; `dem_relief` preserves
-local terrain contrast; `slope` stores terrain steepness normalized from degrees.
+DEM arrays are static for the spatial patch and shared across both orbit views.
+`dem` is raw elevation in meters. `slope_deg` is raw terrain slope in degrees.
+Derived terrain features, including global-normalized DEM, patch-local relief,
+and normalized slope, are intentionally computed during loading rather than
+stored as processed arrays.
 
 ## Source and Scope
 
-The source data comes from OPERA SAR products over North America. Users should
-follow the applicable terms and citation guidance for OPERA source products.
+The source SAR data comes from NASA OPERA products over North America. Terrain
+comes from aligned DEM patches. Users should follow the applicable terms and
+citation guidance for OPERA and DEM source products.
 
 ## Intended Use
 
@@ -156,3 +165,5 @@ self-supervised pretraining. It is not an evaluation benchmark by itself.
 - Missing RTC polarizations are represented by zero-padded channels. The
   metadata includes polarization availability flags so users can distinguish
   true zeros from missing channels.
+- DEM fields are aligned to the 128 x 128 patch grid and should not be treated as
+  a standalone high-resolution terrain product.

data/nasar-train-000000.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1bc2080f2f44052a545cc9bbc18b8cfeb904ac90d878e4aed8a63c374cb03f9
-size 12953507840
+oid sha256:a3738d9f62b775391bdedc686c69adf3c9e3db86dc4a43d36612e67dc200f076
+size 12258744320

data/nasar-train-000001.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:08f926d7ab4cafb29308d316d99d3596d31af5ca06fe754b0266f263f65b4a3b
-size 12707031040
+oid sha256:c66352a6b9617aa7b3e8c2aad616c5fd4867345e723aeb5c55d2c40a3ac32aea
+size 12094156800

data/nasar-train-000002.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f6b182f19c6e1c7f4f74736ab4d857191fd69191a5f82e19180d303b13d6bb3
-size 12639938560
+oid sha256:aa0335cdd0118fe173e95a09da3deae6fdc0108fe9d0a81ab50881d2a25c34f4
+size 12044544000

data/nasar-train-000003.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c240c2ea126a49ad0fc804c0f279a49a9561af1dd98fca87dcd28150796a072e
-size 12637491200
+oid sha256:54678967a2dfd87b0c9cf43682f3dbb3ee530232f4f27a357b334a707a85f55e
+size 12049213440

data/nasar-train-000004.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be8f83a06f94ebeb30fd2a8084b88396bf538623ff1d2641372f372d9c99111e
-size 12665016320
+oid sha256:bed9eb8018139e73c1a2d05ba42b51abc83f06cd66915971b18d3e8ca88a5f40
+size 12062945280

data/nasar-train-000005.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59c2a1a4ebdf31e4ce4290c2e923332f5423778ad533aff0d33e1755520392b8
-size 12664309760
+oid sha256:72416b5ddae96c4e77db41e1a22b19e09720be0dd8cc282b9f6f23475ea5c15b
+size 12060999680

data/nasar-train-000006.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a36f356399513ccc28d0570eda6d23fce10dcd8f9928c66b4aa5dc909892f48a
-size 12683786240
+oid sha256:2c8d1dfb15cfb58ebb6396752d85596ce3cbb091fad077ee61718fbf477ea444
+size 12071587840

data/nasar-train-000007.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbd9c576efb3629144e8178a59d2dad9974f83956718e7230b569952dbe03b57
-size 12684687360
+oid sha256:5bc6e5c40836e499f5c84156500b8b10476917990ad89a27acf0873fff0d5bd0
+size 12071577600

data/nasar-train-000008.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd6a07648869e034b41fefdb5be6bae18b97067e20b2ebec0c6d2f6b4484de92
-size 12673925120
+oid sha256:bf1a9652ebf44905eea992771e5f4079f94009e402d7bc2aeeeb844ea4597920
+size 12067461120

data/nasar-train-000009.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c5d4d9220767e3515c841181fb6cc1fb3a09b4031cdabb597abcc926f7f1861
-size 12706478080
+oid sha256:fee8b123a74f0a8a50812c27a9f6f97f5d304d0aba5ab2f518d0cc63c1632ddb
+size 12088494080

data/nasar-train-000010.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f985d76d67b5c2e76b93e5d0040fb05f41714f93fd7049206a0eb055e1583e8
-size 12759767040
+oid sha256:b87b0e04417cba0eb71a64fc94182ed7d5d7d64d249528221e3ecbb961ef1280
+size 12120647680

data/nasar-train-000011.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:491d221f752591816776f9bff7ad9278b1e9c634d163fcfe51ec944a1427e8e2
-size 12715509760
+oid sha256:3acb43e2a7fe21a2693724fec87444376d4c6e0f89511eae95ba06fa68eba15b
+size 12091361280

data/nasar-train-000012.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d4f68177c4b92387f4c62c2cc718f92f30e3a270c3e5f5d0746309998bf975ce
-size 12717363200
+oid sha256:8d172d54b6fcb990655a1b2643183ff41cb0bb23aa63abf36b20b9fc8099eaa2
+size 12093112320

data/nasar-train-000013.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d1a7b0fbc40cc58695212eeba5cb04916c3844c4b0d19c34ec742d7431d25eb
-size 12703191040
+oid sha256:4def9bcc71e3b20944a306e9c5f8374dd63cdd58cd25127349d9b5919b87fe8c
+size 12082176000

data/nasar-train-000014.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e6ec56e37e2b757f772e714e9bc47134357a72a3a41a2112d0d284b76c0d02b
-size 12688865280
+oid sha256:e4d5df4426aaeaf2baf5a721fe2b801be7b53617bc5fb4dd9db499894684ab5e
+size 12075264000

data/nasar-train-000015.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1798c1b65288ed238032fe3f6c6b651c2e6ae65ecfa959e3095d9efaa1efeaa1
-size 12693442560
+oid sha256:872107a3d85d86f93eb5589d00359b1e425b90546118696a80430985b9486e2d
+size 12079319040

data/nasar-train-000016.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48b3468d0ee4a5925b8b45877f6c08d44aa1ad714a542cb8dec70b08937aa2d2
-size 12717885440
+oid sha256:acebea5e6c8039bd15437b49cc370f2159942c4f150779f22c9439130337edac
+size 12092170240

data/nasar-train-000017.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:11170a75fe95369c210710e14f25622ea32a752ac8ccc8539abac09d4b187928
-size 12725463040
+oid sha256:72f92f828a00f11dad56badae91e473028e865ed831c06a686aabc07c9ba49cd
+size 12096993280

data/nasar-train-000018.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b496ecca1f45cb564f9f1b1b4145f360e1c9c0e609f76dc61f96754f66a110c
-size 12699269120
+oid sha256:1c33b017f05e63890b82591e455bba32e1dc4889c50d8aa160c0ed8a2984ea5e
+size 12083937280

data/nasar-train-000019.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:042c1e99aa1341be7ceaa3a821370216cfb4400a47d57843c1a26cccbcd6892d
-size 12702976000
+oid sha256:b9a82910c62904a04b6865897081aa1be2a733f989ef53aebc4a569e5e4ee0d4
+size 12083896320

data/nasar-train-000020.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:369c82ff206161e525a29fe534e150252bf3732da28f3dd871a14a5626f8ee51
-size 12745093120
+oid sha256:513b3908bbf0788b10d9e724925d82e51ad0000831e743ec96361da09d028448
+size 12107929600

data/nasar-train-000021.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a455e8f12132e88a2753ab2902ae6bf59cf9371bb1b6b0e7f5bc80f7a4a788d1
-size 12711843840
+oid sha256:eac53fb92cacc92b8dd66898468a270ef94c339c9874d8d2b44a0a11f509731a
+size 12089815040

data/nasar-train-000022.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:025c1120043570b4688878977c30764dd217216fa25cae4c4a0b0eb959c9a08c
-size 12727162880
+oid sha256:57b07a83e9c363c8f0e32dd58fb18dec6d9f8d1f2c667f609d8bdeda994b3140
+size 12102031360

data/nasar-train-000023.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01081f2a8fbb29785dbe9153368a1df4a3e4e1802461ac3ad94c6329d9078c97
-size 12738344960
+oid sha256:5c97a8f09dee037af8c1eb35377e562175a2b6c812a99c183e39bda93d00a934
+size 12103823360

data/nasar-train-000024.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2310f80d57f8a50b2841f547ee6f666ab7c3e1c6d317acb4e3ad814ee178bb87
-size 12689623040
+oid sha256:710cfaeef3e7097d3919d5264f1a5d6115493bf4b67fc9a7c1694a354847853f
+size 12074977280

data/nasar-train-000025.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:267939d4be3660392c5975469122f7dad2441ab0f211043a0891307a35ee553d
-size 12708669440
+oid sha256:8ca438fc1df28d2c114647d182d665a79d95b45bbd3259598f1d155a2afa14e7
+size 12089436160

data/nasar-train-000026.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:660785101e5d14c12fb6c563a37a3f0bdd23986b466785fa34722eef1871c94a
-size 12715898880
+oid sha256:114d36b599fa7b20db27ee0a2bd84fcdb900fba4bb6031f8424ecb084e8bf733
+size 12086886400

data/nasar-train-000027.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f11753555d49a271623cba453655aa7f3c33e457d84c41b16ecf90c6e1fde56e
-size 12700805120
+oid sha256:709d40fa25aae3179dd0eca23063d53428494969f6781e3c11accdff6a2d7da5
+size 12083988480

data/nasar-train-000028.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7963b01bb4093fb43f61441e61c9170f067ec68eac04283ab8098a8542f6d9bc
-size 12743147520
+oid sha256:6bcf851936b671ac75ea3bae06fb259a1befb785f02d48ee4ff9253e838271bc
+size 12107806720

data/nasar-train-000029.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ca4ec77df89d97e452083c336a7a608368fae0b32f4563360a1e40c11d25484
-size 12717783040
+oid sha256:98c3828319b3531d4c0ffaa03ff9186cbec0529ea399acc9a2da4d6f8a90412e
+size 12090972160

data/nasar-train-000030.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90955b3a663ae921673692b065881b2c8bfef89b875c4581de507e5a3f47b9c2
-size 12694231040
+oid sha256:d1cefa6ce0c7f5052cb3eca5c88552a4f81f5ea7b1c8391b7c43ce1b1a990e85
+size 12077905920

data/nasar-train-000031.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7517de607483d90e74a34c2a936afda58ede153663d50d1ed13b36e103574d9
-size 12706406400
+oid sha256:f2fd50f7fd7740f64678efd6561dd22c496d4e9f338b8139a9ea1a7a2a5ce2dd
+size 12083148800

data/nasar-train-000032.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1333d0e20f4ed459f9acba1970e196e73a096e7ebc497acc7c34587ebed82415
-size 12702883840
+oid sha256:d01e8d265f5a033325fafced600cfecb98031fa0446a6ad7a262217a3e3ea25f
+size 12082872320

data/nasar-train-000033.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d2396bee661c213bf0143f187a6caa6dd2ca93e6851d3ff09b7bdb9394929ea
-size 12722626560
+oid sha256:18f47393afd2636f0b509db2930bdb80b70cb9b53b3944edb6c098f94fc5b848
+size 12096645120

data/nasar-train-000034.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fa01dcda965dc3967b54a6b7d7653f4aa990623080bf116b67a8da1873f208f5
-size 12723496960
+oid sha256:e6a2d15e7cb4875fab525901fb64765fdc082240dc5ab87dac84f3b90ee5e942
+size 12094351360

data/nasar-train-000035.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f064c8acc170ba41224bfdaebf7c44280ece8d2cb3849a0041bff675fed081be
-size 12730736640
+oid sha256:13f0e753e691a65331c775461c836f286062032a5bd031cc8da1aaee83424eee
+size 12104499200

data/nasar-train-000036.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b104229e88feb1c5807a40a3f243364edb2cf3db978997c7ba903473a3d10195
-size 12723476480
+oid sha256:f07f6d3c81f92f8da44c3a979944c6cdfa7810b92b6d98120b9cccbb22964a64
+size 12094976000

data/nasar-train-000037.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b489b909c3ffd3f50710fcaff392bae1983c18a9a954404fd720c5c032347bb3
-size 12697221120
+oid sha256:b78bae62cb5381ea7b517699c972e2c1690edd4405cefd225fd6e81b1d97591b
+size 12079370240

data/nasar-train-000038.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:165a5bebb1daa5b53c9faf9cd798f38b536d4f71de52796b0226d7f590e5e3e5
-size 12714506240
+oid sha256:16d6c83ec6a36058d9c95a6dde67e314bbf3585b861188e449df1035779c7391
+size 12086353920

data/nasar-train-000039.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46acc4b27e9aa72c8efcd54874cf76a60cdccd1564bd31e36941febbe0f4389f
-size 12701982720
+oid sha256:5536f5cf214865520d1aa9c6bc7830e2d60c353085a6c941250f2af755381789
+size 12079073280

data/nasar-train-000040.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e1b2ab6bb8335d8a177a4243add17ef8dd88a83cc74c7af15358633a261769ae
-size 12698163200
+oid sha256:db77de14fc9638edd171b22c297276d6240b4c3ad3bd2ba21912d20a3db4f498
+size 12071393280

data/nasar-train-000041.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:779574665869c30823c5c65e13aa425223516519268419d4497a674d2f5781b9
-size 12660070400
+oid sha256:30ebf32c8a4d4bc74ff2c6727820ea6494f779a205c4e752fc332dfe81e010b6
+size 12046714880

data/nasar-train-000042.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a5af7aaededb0fe33c5b9f15af02b7de3bae3029cb2daad5fa938eb3c0fee20
-size 12687042560
+oid sha256:de0330793ad336725693b117616591d6c3e09e57e97f313b1484df987920fbc5
+size 12058900480

data/nasar-train-000043.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3192b4dd59b9baf9a023a49f93edf35f921774ce9d7450bfeafc36cce9ac830
-size 12689479680
+oid sha256:fca1008148c0952f30e73686a87f1e7fa66513914a55072260ed20bdbbf5361a
+size 12065761280

data/nasar-train-000044.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bce1101103582f3d75eba77821ff468003144f0c86252b595ef0eb4c5c51afb
-size 12680949760
+oid sha256:b6bb4c248c70d1cd1a9b7a1a1ff0f2ffe565207b14ea488fa1013630cd862f84
+size 12053483520

data/nasar-train-000045.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b668dd77dfd4be5b013beada78394f9ba84e6af5adb29bee1b1d9ebd53bad974
-size 12683417600
+oid sha256:ad93291c25ed8c0bb810958973342328bbf82e9669f0443d85597fcab12e2d0d
+size 12057978880

data/nasar-train-000046.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d52896dfe09889cf67910793bfa084763b8e8620c00a1fea82674f17419e166e
-size 12715724800
+oid sha256:b4106f1631be8b055f8db681c07ea4ef1f51d3af542ed1685be7530c275316e5
+size 12084490240

data/nasar-train-000047.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:106e1310d99c8304c834208acac58dce4be74286668d7c1b7794246c25d1a765
-size 12709079040
+oid sha256:439314b4dc7e0a5183499356850a0b7375875a0cc5b8f6e2a28fbd04390ff3c9
+size 12077271040

data/nasar-train-000048.tar

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:432491742c50c7e3d39ac2edf184d92c1abd5bb426481ff0dac77912f8e9c6ee
-size 12688414720
+oid sha256:98c50a978dbb07c792db299a2e1de1668b4c2f00b47967a335837149458560f3
+size 12063201280