| 8935feb | Ko Sugawara | 22 December 2021, 22:09:10 UTC | Merge pull request #13 from elephant-track/dev ELEPHANT server v0.3.3 -> v0.3.4 | 22 December 2021, 22:09:10 UTC |
| 7811455 | Ko Sugawara | 22 December 2021, 17:50:06 UTC | Update version to 0.3.4 | 22 December 2021, 17:50:06 UTC |
| ff1b423 | Ko Sugawara | 22 December 2021, 17:27:37 UTC | Resolve an issue with installing conda/mamba packages - update environment.yml to resolve an issue with installing conda/mamba packages - update Dockerfile to make it friendly with development | 22 December 2021, 17:27:37 UTC |
| c3743a7 | Ko Sugawara | 22 December 2021, 17:25:45 UTC | Update version to 0.3.4-dev | 22 December 2021, 17:25:45 UTC |
| cfb4caa | Ko Sugawara | 08 December 2021, 16:40:47 UTC | Merge pull request #12 from elephant-track/dev ELEPHANT server v0.3.2 -> v0.3.3 | 08 December 2021, 16:40:47 UTC |
| 3a87595 | Ko Sugawara | 08 December 2021, 16:24:54 UTC | Update version to 0.3.3 | 08 December 2021, 16:24:54 UTC |
| be4de65 | Ko Sugawara | 04 December 2021, 13:21:18 UTC | Loosen the tolerance for testing model outputs - rtol: 1e-5 -> 1e-4 | 04 December 2021, 13:21:18 UTC |
| 961f470 | Ko Sugawara | 04 December 2021, 12:45:36 UTC | Use the same parameters for allclose in Numpy and PyTorch - Use the default parameters for PyTorch (rtol=1e-05, atol=1e-08) - assert is added to test_models | 04 December 2021, 12:45:36 UTC |
| 68748e7 | Ko Sugawara | 03 December 2021, 22:59:53 UTC | Update PyTorch and CudaToolKit - support Ampere architecture | 03 December 2021, 22:59:53 UTC |
| 416cdba | Ko Sugawara | 03 December 2021, 22:11:50 UTC | Update version to 0.3.3-dev | 03 December 2021, 22:11:50 UTC |
| 50e3e22 | Ko Sugawara | 03 December 2021, 19:16:17 UTC | Merge pull request #11 from elephant-track/dev ELEPHANT server v0.3.1 -> v0.3.2 | 03 December 2021, 19:16:17 UTC |
| 4feca0a | Ko Sugawara | 02 December 2021, 14:35:32 UTC | Update version to 0.3.2 | 02 December 2021, 14:35:32 UTC |
| d470fd1 | Ko Sugawara | 02 December 2021, 14:34:41 UTC | Avoid an error with groupmod | 02 December 2021, 14:34:41 UTC |
| d69e7d4 | Ko Sugawara | 02 December 2021, 00:49:20 UTC | Merge pull request #10 from elephant-track/dev v0.3.0 -> v0.3.1 | 02 December 2021, 00:49:20 UTC |
| e3beb3c | Ko Sugawara | 02 December 2021, 00:48:06 UTC | Update version to 0.3.1 | 02 December 2021, 00:48:06 UTC |
| da2442d | Ko Sugawara | 02 December 2021, 00:47:50 UTC | Add error log for model reset during process is running | 02 December 2021, 00:47:50 UTC |
| dde7163 | Ko Sugawara | 01 December 2021, 22:55:47 UTC | Merge pull request #9 from elephant-track/dev ELEPHANT server v0.2.0 -> v0.3.0 | 01 December 2021, 22:55:47 UTC |
| c49efe4 | Ko Sugawara | 01 December 2021, 22:37:13 UTC | Update version to 0.3.0 | 01 December 2021, 22:40:32 UTC |
| 0bad3f5 | Ko Sugawara | 01 December 2021, 22:04:27 UTC | Merge branch 'v0.2.x' into dev | 01 December 2021, 22:04:27 UTC |
| bc80e1d | Ko Sugawara | 24 November 2021, 18:02:01 UTC | Explicitly kill all services before restart - the services can remain working when the notebook does not respond | 24 November 2021, 18:02:01 UTC |
| 46be9e8 | Ko Sugawara | 24 November 2021, 17:08:22 UTC | Avoid errors on the CPU environment | 24 November 2021, 17:08:22 UTC |
| 4a1adda | Ko Sugawara | 24 November 2021, 16:04:26 UTC | Avoid error on CPU environment - make the gpus/ endpoint compatible with the CPU environment | 24 November 2021, 16:04:26 UTC |
| 7d48c3f | Ko Sugawara | 24 November 2021, 16:02:56 UTC | Update version to 0.2.1-dev | 24 November 2021, 16:02:56 UTC |
| 177fdaf | Ko Sugawara | 24 November 2021, 12:20:03 UTC | Fix zip name - .zip.zip tp .zip | 24 November 2021, 12:20:03 UTC |
| 4a2a54a | Ko Sugawara | 24 November 2021, 09:30:17 UTC | Keep export results on the server | 24 November 2021, 09:39:36 UTC |
| 2107ade | Ko Sugawara | 23 November 2021, 17:06:38 UTC | Replace Conda with Mamba - replace Conda with Mamba - use the same environment.yml in all options (Docker, Singularity and Colab) | 23 November 2021, 17:06:38 UTC |
| 2f02edc | Ko Sugawara | 07 November 2021, 21:24:04 UTC | Merge branch 'tests' into dev | 07 November 2021, 21:24:04 UTC |
| 4622fd9 | Ko Sugawara | 04 November 2021, 17:21:22 UTC | Implement uploading models (via reset) - when a request is multipart/form-data, an attached file is used to load state_dict(s) for resetting a model - if the uploaded model parameters are not compatible with a model, an error will be thrown | 04 November 2021, 17:21:22 UTC |
| d94c8b6 | Ko Sugawara | 03 November 2021, 17:23:28 UTC | Implement download_model endpoint | 03 November 2021, 17:23:28 UTC |
| dd2d146 | Ko Sugawara | 03 November 2021, 14:09:18 UTC | Add/update log messages - log at before/after each request - update "waiting" log - minor fix in a comment | 03 November 2021, 14:10:03 UTC |
| 6d48e68 | Ko Sugawara | 03 November 2021, 14:07:45 UTC | Fix training epoch to log - batch_index starts from 1 in log - log at the end of log_interval instead of the beginning of it | 03 November 2021, 14:07:45 UTC |
| 21bbd85 | Ko Sugawara | 26 October 2021, 16:28:12 UTC | Add test_models | 26 October 2021, 16:28:12 UTC |
| 5216012 | Ko Sugawara | 20 October 2021, 15:15:55 UTC | Add test_dataset | 20 October 2021, 15:15:55 UTC |
| d492c52 | Ko Sugawara | 20 October 2021, 15:14:04 UTC | Dissociate build from test in Makefile | 20 October 2021, 15:14:18 UTC |
| 639a8ff | Ko Sugawara | 20 October 2021, 14:35:44 UTC | Use appropriate data for testing scaled_moments_central | 20 October 2021, 14:35:44 UTC |
| cc3a6f6 | Ko Sugawara | 20 October 2021, 14:00:22 UTC | Add test_np_linspace - debug which part causes error | 20 October 2021, 14:00:22 UTC |
| 54e9a5c | Ko Sugawara | 20 October 2021, 13:28:56 UTC | Add test_scaled_centroid - check if scaled_centroid works consistently between the local environment and GitHub's environment. | 20 October 2021, 13:28:56 UTC |
| 9d35579 | Ko Sugawara | 20 October 2021, 13:08:15 UTC | Use Conda in GitHub actions | 20 October 2021, 13:12:57 UTC |
| 1413692 | Ko Sugawara | 19 October 2021, 15:46:33 UTC | Add unit tests for util functions | 19 October 2021, 15:46:33 UTC |
| ddee2e7 | Ko Sugawara | 19 October 2021, 15:45:45 UTC | Add test command to Makefile - `make test` to run tests | 19 October 2021, 15:46:06 UTC |
| f56db82 | Ko Sugawara | 28 September 2021, 13:37:08 UTC | Merge pull request #6 from elephant-track/dev ELEPHANT server v0.1.1 -> v0.2.0 | 28 September 2021, 13:37:08 UTC |
| 633292b | Ko Sugawara | 28 September 2021, 12:27:02 UTC | Prepare for v0.3.0 | 28 September 2021, 12:27:02 UTC |
| 214a14e | Ko Sugawara | 28 September 2021, 12:12:08 UTC | Update version to 0.2.0 | 28 September 2021, 12:19:07 UTC |
| 874e1f9 | Ko Sugawara | 28 September 2021, 12:01:09 UTC | Update README | 28 September 2021, 12:01:09 UTC |
| bba57f5 | Ko Sugawara | 20 September 2021, 18:02:15 UTC | Add training config for no-aug condition | 20 September 2021, 18:02:15 UTC |
| ee70048 | Ko Sugawara | 14 September 2021, 10:22:44 UTC | Update train/val split in train.py - train/val split was done considering missing timepoints | 14 September 2021, 10:22:44 UTC |
| 29a254a | Ko Sugawara | 14 September 2021, 10:21:51 UTC | Update training scripts for checking overfitting - add a script for timepoints (t000-t100-t194) - add a script for pretrained versatile model (#002) | 14 September 2021, 10:21:51 UTC |
| 2eaf676 | Ko Sugawara | 14 September 2021, 10:20:23 UTC | Update training scripts for versatile models - scales and anisotropy are taken into account | 14 September 2021, 10:20:23 UTC |
| 7315a82 | Ko Sugawara | 11 September 2021, 13:25:49 UTC | Fix issues - arguments for init_flow_models and predict are updated | 11 September 2021, 13:25:49 UTC |
| 1a30425 | Ko Sugawara | 08 September 2021, 15:03:19 UTC | Organize model reset strategy - Versatile -> latest versatile model - Default -> self-supervised (seg) or random (flow) - URL -> load from the specified URL | 08 September 2021, 15:03:19 UTC |
| 8d14ef9 | Ko Sugawara | 08 September 2021, 09:17:54 UTC | Update Docker tag - from 0.1.0 to 0.2.0 | 08 September 2021, 09:17:54 UTC |
| 6b1d59e | Ko Sugawara | 08 September 2021, 09:17:35 UTC | Upgrade tensorflow to 2.4.0 - there was a problem of segmentation fault in v2.3.0 | 08 September 2021, 09:17:35 UTC |
| 25d9ced | Ko Sugawara | 03 September 2021, 19:35:17 UTC | Set default false_weight to 10 | 03 September 2021, 19:35:17 UTC |
| df131f5 | Ko Sugawara | 03 September 2021, 19:34:24 UTC | Move keep_axials from model to dataset - model is free from anisotropy | 03 September 2021, 19:34:45 UTC |
| 83c2a03 | Ko Sugawara | 31 August 2021, 16:53:16 UTC | Training scripts for versatile3d - 4 conditions | 31 August 2021, 16:53:16 UTC |
| 5c0b7e2 | Ko Sugawara | 29 August 2021, 21:07:40 UTC | Scripts for checking over-fitting problem - temporarily skip resetting the parameters to the best performing ones | 31 August 2021, 13:55:06 UTC |
| 90423e1 | Ko Sugawara | 29 August 2021, 21:04:50 UTC | Change loss weights for seg - use values in the paper - TODO: parameterize the weights | 29 August 2021, 21:04:50 UTC |
| c654ac9 | Ko Sugawara | 29 August 2021, 21:03:45 UTC | Stop using probability to refine spot radii - probability was used to shrink the spot radii but it is difficult to justify this operation | 29 August 2021, 21:03:45 UTC |
| 572a3a7 | Ko Sugawara | 29 August 2021, 20:50:35 UTC | Add is_trained flag to init_seg_model() - model can be reset either by pretrained model or self-supervised training | 29 August 2021, 20:50:35 UTC |
| 4fa00e2 | Ko Sugawara | 28 August 2021, 22:19:31 UTC | Fix issues - fix issues in response | 28 August 2021, 22:19:31 UTC |
| 7bb7c43 | Ko Sugawara | 26 August 2021, 23:30:02 UTC | Reorganize cancel behavior | 26 August 2021, 23:30:02 UTC |
| 4b35641 | Ko Sugawara | 26 August 2021, 23:04:39 UTC | Update training config for no-MDA condition - used for testing with the MDA231 dataset | 26 August 2021, 23:04:39 UTC |
| 84d269a | Ko Sugawara | 25 August 2021, 22:19:12 UTC | Epoch 200-249 This setting was used for the training epochs 200-249. | 25 August 2021, 22:19:12 UTC |
| 64e1956 | Ko Sugawara | 25 August 2021, 22:18:23 UTC | Epoch 100-199 This setting was used for the training epochs 100-199. | 25 August 2021, 22:18:23 UTC |
| 09c0005 | Ko Sugawara | 19 August 2021, 16:35:40 UTC | Add pretrained model parameters for detection - seg models are initialized with pretrained parameters | 19 August 2021, 16:35:40 UTC |
| 4b86c1f | Ko Sugawara | 19 August 2021, 13:30:51 UTC | Reorganizing Redis state management - avoid problems | 19 August 2021, 15:01:04 UTC |
| 102803d | Ko Sugawara | 19 August 2021, 13:28:39 UTC | Remove unused import - traceback is not used | 19 August 2021, 13:28:39 UTC |
| 7f9bcd8 | Ko Sugawara | 19 August 2021, 13:17:39 UTC | Scripts for training a versatile 3d model for detection - make generate_seg_labels.py compatible with 3d datasets - add config files | 19 August 2021, 13:17:39 UTC |
| 2e6dee2 | Ko Sugawara | 19 August 2021, 13:13:15 UTC | Ensure a model before prediction of flow - initialize if model does not exist | 19 August 2021, 13:13:15 UTC |
| 03706e1 | Ko Sugawara | 19 August 2021, 13:10:40 UTC | Minor fixes - ellipse and ellipsoid accept numpy arrays for center, radii and rotation parameters | 19 August 2021, 13:10:40 UTC |
| dbbf0ab | Ko Sugawara | 19 August 2021, 13:09:42 UTC | Use log_interval in train.py - log_interval is loaded from a config file | 19 August 2021, 13:09:42 UTC |
| d1e642d | Ko Sugawara | 19 August 2021, 13:07:43 UTC | Add patch_size to evaluation - use _patch_predict for large images in evaluation | 19 August 2021, 13:07:43 UTC |
| e3ff07f | Ko Sugawara | 18 August 2021, 17:55:48 UTC | Parameterize log_interval - log_interval can be set as a parameter | 18 August 2021, 17:55:48 UTC |
| 3d11c21 | Ko Sugawara | 18 August 2021, 09:55:15 UTC | Average training losses to show - training losses are averaged between record points | 18 August 2021, 09:55:15 UTC |
| ebcd2e7 | Ko Sugawara | 14 August 2021, 13:13:38 UTC | Automatically reset a model when a model file is missing - automatically reset a model when a model file is missing | 14 August 2021, 13:13:38 UTC |
| d662ba6 | Ko Sugawara | 11 August 2021, 17:16:15 UTC | Update Colab notebook - better instructions for SSH | 11 August 2021, 17:16:15 UTC |
| d955db9 | Ko Sugawara | 11 August 2021, 15:10:30 UTC | Update version to 0.2.0-dev | 11 August 2021, 15:10:30 UTC |
| 65e9e06 | Ko Sugawara | 11 August 2021, 15:10:09 UTC | Update Colab notebook - for dev version | 11 August 2021, 15:10:09 UTC |
| 084f4f4 | Ko Sugawara | 10 August 2021, 19:44:09 UTC | Implement logger for RabbitMQ - logger with a handler that delegates logs to RabbitMQ | 10 August 2021, 19:44:09 UTC |
| f5d9a90 | Ko Sugawara | 10 August 2021, 13:18:04 UTC | Update Colab notebook | 10 August 2021, 13:21:28 UTC |
| b720f9e | Ko Sugawara | 10 August 2021, 13:00:35 UTC | Minor fix - remove old docstring | 10 August 2021, 13:00:35 UTC |
| cd0b0d8 | Ko Sugawara | 10 August 2021, 09:28:19 UTC | Fix an issue in upload - mkdir if the datasets directory is not exists | 10 August 2021, 10:25:45 UTC |
| 47d4390 | Ko Sugawara | 09 August 2021, 20:52:50 UTC | Add environment.yml - add environment.yml to reproduce the conda environment | 09 August 2021, 21:28:20 UTC |
| 781f08e | Ko Sugawara | 09 August 2021, 20:51:57 UTC | Concatenate logs - concatenate training logs | 09 August 2021, 20:51:57 UTC |
| f6da231 | Ko Sugawara | 09 August 2021, 20:41:02 UTC | Update dataset endpoints - report progress of generating dataset using pika (RabbitMQ) - enrich error handling | 09 August 2021, 20:41:02 UTC |
| 095d1c5 | Ko Sugawara | 03 August 2021, 17:23:28 UTC | Implement dataset endpoints - dataset/generate to generate dataset from h5 - dataset/check to check if dataset is valid | 03 August 2021, 17:23:28 UTC |
| 6437521 | Ko Sugawara | 03 August 2021, 15:23:20 UTC | Update upload endpoint - use chunks to enable cancel | 03 August 2021, 15:23:20 UTC |
| 4681932 | Ko Sugawara | 27 July 2021, 22:03:27 UTC | Ignore gpus/ logs - health check logs are ignored | 27 July 2021, 22:03:27 UTC |
| e236045 | Ko Sugawara | 27 July 2021, 22:02:55 UTC | Implement dataset endpoint - generate_dataset() is implemented in elephant.tool - dataset_generator.py calls elephant.tool.generate_dataset() - gen_dataset() (dataset endpoint) calls elephant.tool.generate_dataset() | 27 July 2021, 22:02:55 UTC |
| 58740e1 | Ko Sugawara | 27 July 2021, 22:00:19 UTC | Implement upload endpoint - upload an image data to the dataset directory - USE_CISTOM_STREAM is a memory-friendly (and faster) option (default) | 27 July 2021, 22:00:19 UTC |
| fb15819 | Ko Sugawara | 17 July 2021, 20:18:06 UTC | Add get_gpus() - get gpu information using nvsmi | 17 July 2021, 20:18:06 UTC |
| 89c3995 | Ko Sugawara | 13 July 2021, 20:04:24 UTC | Remove unused import | 13 July 2021, 20:04:24 UTC |
| b55424d | Ko Sugawara | 13 July 2021, 20:04:05 UTC | Add warmup to Makefile - warmup is required to find GPUs sometimes | 13 July 2021, 20:04:05 UTC |
| b07c65a | Ko Sugawara | 29 June 2021, 16:35:03 UTC | Merge branch 'tests' into dev | 29 June 2021, 16:35:03 UTC |
| 3f7ad3b | Ko Sugawara | 29 June 2021, 16:30:24 UTC | Add tests for utils | 29 June 2021, 16:30:24 UTC |
| a3fbed2 | Ko Sugawara | 29 June 2021, 15:50:25 UTC | Merge pull request #5 from elephant-track/github-actions Add GitHub actions | 29 June 2021, 15:50:25 UTC |
| 70cb4f2 | Ko Sugawara | 29 June 2021, 15:45:16 UTC | Fix a workflow file | 29 June 2021, 15:45:16 UTC |
| 9c40185 | Ko Sugawara | 29 June 2021, 15:38:53 UTC | Fix requirements.txt | 29 June 2021, 15:38:53 UTC |
| 187665f | Ko Sugawara | 29 June 2021, 15:28:55 UTC | Update a workflow for test - install dependencies - install project - add a mock test | 29 June 2021, 15:28:55 UTC |
| adc6f28 | Ko Sugawara | 29 June 2021, 14:00:27 UTC | Update config file organization - config files are organized | 29 June 2021, 14:00:27 UTC |
