hypergan 1.0.6__tar.gz → 2.0.0a2__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (800) hide show
  1. hypergan-2.0.0a2/LICENSE +21 -0
  2. hypergan-2.0.0a2/MANIFEST.in +34 -0
  3. hypergan-2.0.0a2/PKG-INFO +390 -0
  4. hypergan-2.0.0a2/README.md +319 -0
  5. hypergan-2.0.0a2/examples/README.md +23 -0
  6. hypergan-2.0.0a2/examples/api/config_first/critic.hndl +6 -0
  7. hypergan-2.0.0a2/examples/api/config_first/demo_1_simple.py +32 -0
  8. hypergan-2.0.0a2/examples/api/config_first/demo_2_multi_network.py +37 -0
  9. hypergan-2.0.0a2/examples/api/config_first/demo_3_extensions.py +41 -0
  10. hypergan-2.0.0a2/examples/api/config_first/demo_4_round_trip.py +71 -0
  11. hypergan-2.0.0a2/examples/api/config_first/demo_5_read_run.py +51 -0
  12. hypergan-2.0.0a2/examples/api/config_first/demo_replicated.py +33 -0
  13. hypergan-2.0.0a2/examples/api/config_first/extensions.toml +77 -0
  14. hypergan-2.0.0a2/examples/api/config_first/simple.toml +47 -0
  15. hypergan-2.0.0a2/examples/api/config_first/toy_project.py +107 -0
  16. hypergan-2.0.0a2/examples/api/config_first/two_critics.toml +91 -0
  17. hypergan-2.0.0a2/examples/archive/README.md +12 -0
  18. hypergan-2.0.0a2/examples/archive/cifar-pretrained-sagan.toml +151 -0
  19. hypergan-2.0.0a2/examples/archive/cifar-tiny-transformer-resnet-features-pg07-lr2e-4-demo.toml +129 -0
  20. hypergan-2.0.0a2/examples/archive/cifar-tiny-transformer-resnet-features.toml +129 -0
  21. hypergan-2.0.0a2/examples/archive/cifar-tiny-transformer.toml +95 -0
  22. hypergan-2.0.0a2/examples/archive/cifar-transgan.toml +153 -0
  23. hypergan-2.0.0a2/examples/archive/dcgan-128.toml +120 -0
  24. hypergan-2.0.0a2/examples/archive/dcgan-resnet-128.toml +111 -0
  25. hypergan-2.0.0a2/examples/archive/dcgan-resnet-multiscale-128.toml +113 -0
  26. hypergan-2.0.0a2/examples/archive/logos-ae-gan-256.toml +225 -0
  27. hypergan-2.0.0a2/examples/archive/logos-colorization-128.toml +198 -0
  28. hypergan-2.0.0a2/examples/archive/logos-colorization-256.toml +258 -0
  29. hypergan-2.0.0a2/examples/archive/logos-tiny-transformer-dinov3-projected-128.toml +186 -0
  30. hypergan-2.0.0a2/examples/archive/logos-tiny-transformer-resnet-features-128.toml +204 -0
  31. hypergan-2.0.0a2/examples/archive/networks/cifar-discriminator-32.hndl +147 -0
  32. hypergan-2.0.0a2/examples/archive/networks/colorization-generator-128.hndl +115 -0
  33. hypergan-2.0.0a2/examples/archive/networks/dinov3-multidepth-discriminator-128.hndl +193 -0
  34. hypergan-2.0.0a2/examples/archive/networks/dinov3-projected-discriminator-128-no-diffaug.hndl +70 -0
  35. hypergan-2.0.0a2/examples/archive/networks/dinov3-projected-discriminator-128-stable.hndl +69 -0
  36. hypergan-2.0.0a2/examples/archive/networks/dinov3-projected-discriminator-128.hndl +94 -0
  37. hypergan-2.0.0a2/examples/archive/networks/resnet18-discriminator.hndl +29 -0
  38. hypergan-2.0.0a2/examples/archive/networks/resnet18-features-discriminator-32.hndl +70 -0
  39. hypergan-2.0.0a2/examples/archive/networks/resnet18-multiscale-discriminator-128.hndl +177 -0
  40. hypergan-2.0.0a2/examples/archive/networks/sagan-adain-generator-128.hndl +110 -0
  41. hypergan-2.0.0a2/examples/archive/networks/simple-styletransformer-generator-128.hndl +173 -0
  42. hypergan-2.0.0a2/examples/archive/networks/tiny-transformer-generator-128.hndl +43 -0
  43. hypergan-2.0.0a2/examples/archive/networks/tiny-transformer-generator-32.hndl +30 -0
  44. hypergan-2.0.0a2/examples/archive/networks/transgan-discriminator-32.hndl +88 -0
  45. hypergan-2.0.0a2/examples/archive/networks/transgan-generator-128-e2.hndl +439 -0
  46. hypergan-2.0.0a2/examples/archive/networks/transgan-generator-128-e3.hndl +408 -0
  47. hypergan-2.0.0a2/examples/archive/networks/transgan-generator-128-equalized.hndl +389 -0
  48. hypergan-2.0.0a2/examples/archive/networks/transgan-generator-128-stable.hndl +395 -0
  49. hypergan-2.0.0a2/examples/archive/networks/transgan-generator-128.hndl +241 -0
  50. hypergan-2.0.0a2/examples/archive/networks/transgan-generator-32.hndl +115 -0
  51. hypergan-2.0.0a2/examples/archive/sagan-adain-dinov3-multidepth-128.toml +97 -0
  52. hypergan-2.0.0a2/examples/archive/sagan-adain-resnet-multiscale-128.toml +93 -0
  53. hypergan-2.0.0a2/examples/archive/simple-styletransformer-resnet-128-low-g-lr.toml +92 -0
  54. hypergan-2.0.0a2/examples/archive/simple-styletransformer-resnet-128-normal-g-lr.toml +92 -0
  55. hypergan-2.0.0a2/examples/archive/tiny-transformer-resnet-128.toml +92 -0
  56. hypergan-2.0.0a2/examples/archive/transgan-dinov3-multidepth-128.toml +96 -0
  57. hypergan-2.0.0a2/examples/archive/transgan-projected-dinov3-128-no-diffaug.toml +99 -0
  58. hypergan-2.0.0a2/examples/archive/transgan-projected-dinov3-128-stable.toml +98 -0
  59. hypergan-2.0.0a2/examples/archive/transgan-resnet-multiscale-128-e2.toml +96 -0
  60. hypergan-2.0.0a2/examples/archive/transgan-resnet-multiscale-128-e3.toml +96 -0
  61. hypergan-2.0.0a2/examples/archive/transgan-resnet-multiscale-128-equalized.toml +96 -0
  62. hypergan-2.0.0a2/examples/archive/transgan-resnet-multiscale-128-stable.toml +95 -0
  63. hypergan-2.0.0a2/examples/cifar-transgan-resnet18.toml +134 -0
  64. hypergan-2.0.0a2/examples/cifar-transgan-scratch.toml +131 -0
  65. hypergan-2.0.0a2/examples/cifar10-tiny-transformer.toml +131 -0
  66. hypergan-2.0.0a2/examples/cifar10-transgan.toml +132 -0
  67. hypergan-2.0.0a2/examples/color-words.toml +174 -0
  68. hypergan-2.0.0a2/examples/execution/cpu-replicated.toml +11 -0
  69. hypergan-2.0.0a2/examples/execution/cpu-single.toml +10 -0
  70. hypergan-2.0.0a2/examples/execution/cuda-replicated-nccl.toml +10 -0
  71. hypergan-2.0.0a2/examples/gaussian-grid.toml +64 -0
  72. hypergan-2.0.0a2/examples/image-folder-colorization.toml +206 -0
  73. hypergan-2.0.0a2/examples/image-folder-dcgan.toml +99 -0
  74. hypergan-2.0.0a2/examples/image-folder-tiny-transformer.toml +68 -0
  75. hypergan-2.0.0a2/examples/networks/cifar-discriminator-32.hndl +147 -0
  76. hypergan-2.0.0a2/examples/networks/colorization-generator-128.hndl +115 -0
  77. hypergan-2.0.0a2/examples/networks/dinov3-multidepth-discriminator-128.hndl +193 -0
  78. hypergan-2.0.0a2/examples/networks/dinov3-projected-discriminator-128-no-diffaug.hndl +70 -0
  79. hypergan-2.0.0a2/examples/networks/dinov3-projected-discriminator-128-stable.hndl +69 -0
  80. hypergan-2.0.0a2/examples/networks/dinov3-projected-discriminator-128.hndl +94 -0
  81. hypergan-2.0.0a2/examples/networks/resnet18-discriminator.hndl +29 -0
  82. hypergan-2.0.0a2/examples/networks/resnet18-features-discriminator-32.hndl +70 -0
  83. hypergan-2.0.0a2/examples/networks/resnet18-multiscale-discriminator-128.hndl +177 -0
  84. hypergan-2.0.0a2/examples/networks/sagan-adain-generator-128.hndl +110 -0
  85. hypergan-2.0.0a2/examples/networks/simple-styletransformer-generator-128.hndl +173 -0
  86. hypergan-2.0.0a2/examples/networks/tiny-transformer-generator-128.hndl +43 -0
  87. hypergan-2.0.0a2/examples/networks/tiny-transformer-generator-32.hndl +30 -0
  88. hypergan-2.0.0a2/examples/networks/transgan-discriminator-32.hndl +88 -0
  89. hypergan-2.0.0a2/examples/networks/transgan-generator-128-e2.hndl +439 -0
  90. hypergan-2.0.0a2/examples/networks/transgan-generator-128-e3.hndl +408 -0
  91. hypergan-2.0.0a2/examples/networks/transgan-generator-128-equalized.hndl +389 -0
  92. hypergan-2.0.0a2/examples/networks/transgan-generator-128-stable.hndl +395 -0
  93. hypergan-2.0.0a2/examples/networks/transgan-generator-128.hndl +241 -0
  94. hypergan-2.0.0a2/examples/networks/transgan-generator-32.hndl +115 -0
  95. hypergan-2.0.0a2/examples/paired-linear.toml +88 -0
  96. hypergan-2.0.0a2/frontend/API.md +145 -0
  97. hypergan-2.0.0a2/frontend/README.md +117 -0
  98. hypergan-2.0.0a2/frontend/build.mjs +90 -0
  99. hypergan-2.0.0a2/frontend/chart.test.mjs +57 -0
  100. hypergan-2.0.0a2/frontend/coarsen.test.mjs +43 -0
  101. hypergan-2.0.0a2/frontend/package-lock.json +527 -0
  102. hypergan-2.0.0a2/frontend/package.json +18 -0
  103. hypergan-2.0.0a2/frontend/presets.test.mjs +188 -0
  104. hypergan-2.0.0a2/frontend/src/app.js +2149 -0
  105. hypergan-2.0.0a2/frontend/src/chart.js +84 -0
  106. hypergan-2.0.0a2/frontend/src/coarsen.js +28 -0
  107. hypergan-2.0.0a2/frontend/src/evaluations.js +416 -0
  108. hypergan-2.0.0a2/frontend/src/model.js +364 -0
  109. hypergan-2.0.0a2/frontend/src/presets.js +254 -0
  110. hypergan-2.0.0a2/pyproject.toml +98 -0
  111. hypergan-2.0.0a2/scripts/benchmark_particlegan_cifar.py +166 -0
  112. hypergan-2.0.0a2/scripts/benchmark_scaling.py +373 -0
  113. hypergan-2.0.0a2/scripts/benchmark_transgan_critics.sh +94 -0
  114. hypergan-2.0.0a2/scripts/foldlite_integration_proof.py +120 -0
  115. hypergan-2.0.0a2/scripts/generate_logo.py +364 -0
  116. hypergan-2.0.0a2/scripts/generator_signal_proof.py +267 -0
  117. hypergan-2.0.0a2/scripts/metrics_server_proof.py +132 -0
  118. hypergan-2.0.0a2/scripts/metrics_training_proof.py +405 -0
  119. hypergan-2.0.0a2/scripts/modal_multinode_smoke.py +115 -0
  120. hypergan-2.0.0a2/scripts/modal_scaling.py +131 -0
  121. hypergan-2.0.0a2/scripts/multinode_smoke_worker.py +109 -0
  122. hypergan-2.0.0a2/scripts/reducer_proof.py +54 -0
  123. hypergan-2.0.0a2/scripts/rocm_preflight.sh +252 -0
  124. hypergan-2.0.0a2/scripts/rocm_preflight_helper.py +963 -0
  125. hypergan-2.0.0a2/scripts/start-cifar-tiny-transformer-resnet-features-pg07-lr2e-4-demo.sh +53 -0
  126. hypergan-2.0.0a2/setup.cfg +4 -0
  127. hypergan-2.0.0a2/setup.py +47 -0
  128. hypergan-2.0.0a2/src/hypergan/__init__.py +131 -0
  129. hypergan-2.0.0a2/src/hypergan/__main__.py +5 -0
  130. hypergan-2.0.0a2/src/hypergan/_build_provenance.json +1 -0
  131. hypergan-2.0.0a2/src/hypergan/_file_lock.py +55 -0
  132. hypergan-2.0.0a2/src/hypergan/aggregation.py +651 -0
  133. hypergan-2.0.0a2/src/hypergan/api.py +2709 -0
  134. hypergan-2.0.0a2/src/hypergan/artifacts.py +674 -0
  135. hypergan-2.0.0a2/src/hypergan/autoencoder_components.py +60 -0
  136. hypergan-2.0.0a2/src/hypergan/background_poll.py +74 -0
  137. hypergan-2.0.0a2/src/hypergan/bounded_cli_output.py +576 -0
  138. hypergan-2.0.0a2/src/hypergan/bounded_observer.py +278 -0
  139. hypergan-2.0.0a2/src/hypergan/cache.py +247 -0
  140. hypergan-2.0.0a2/src/hypergan/checkpoint_compatibility.py +280 -0
  141. hypergan-2.0.0a2/src/hypergan/checkpoints.py +1242 -0
  142. hypergan-2.0.0a2/src/hypergan/cli.py +2394 -0
  143. hypergan-2.0.0a2/src/hypergan/color_words.py +297 -0
  144. hypergan-2.0.0a2/src/hypergan/colorization_components.py +360 -0
  145. hypergan-2.0.0a2/src/hypergan/colorization_data.py +270 -0
  146. hypergan-2.0.0a2/src/hypergan/colorization_metrics.py +158 -0
  147. hypergan-2.0.0a2/src/hypergan/colorspace.py +9 -0
  148. hypergan-2.0.0a2/src/hypergan/config.py +950 -0
  149. hypergan-2.0.0a2/src/hypergan/console_settings.py +87 -0
  150. hypergan-2.0.0a2/src/hypergan/cpu_worker_service.py +696 -0
  151. hypergan-2.0.0a2/src/hypergan/cpu_workers.py +125 -0
  152. hypergan-2.0.0a2/src/hypergan/data.py +602 -0
  153. hypergan-2.0.0a2/src/hypergan/dataset_download.py +568 -0
  154. hypergan-2.0.0a2/src/hypergan/datasets.py +6 -0
  155. hypergan-2.0.0a2/src/hypergan/diff_augment.py +131 -0
  156. hypergan-2.0.0a2/src/hypergan/distributed.py +120 -0
  157. hypergan-2.0.0a2/src/hypergan/distributed_checkpoints.py +858 -0
  158. hypergan-2.0.0a2/src/hypergan/distributed_commit.py +319 -0
  159. hypergan-2.0.0a2/src/hypergan/distributed_training.py +1271 -0
  160. hypergan-2.0.0a2/src/hypergan/diversity_metrics.py +187 -0
  161. hypergan-2.0.0a2/src/hypergan/evaluation_cli.py +22 -0
  162. hypergan-2.0.0a2/src/hypergan/evaluation_snapshot.py +28 -0
  163. hypergan-2.0.0a2/src/hypergan/event_views.py +500 -0
  164. hypergan-2.0.0a2/src/hypergan/execution.py +852 -0
  165. hypergan-2.0.0a2/src/hypergan/execution_preflight.py +286 -0
  166. hypergan-2.0.0a2/src/hypergan/execution_profiles.py +310 -0
  167. hypergan-2.0.0a2/src/hypergan/gpu_inventory.py +690 -0
  168. hypergan-2.0.0a2/src/hypergan/hndl_augmentation.py +40 -0
  169. hypergan-2.0.0a2/src/hypergan/hndl_networks.py +42 -0
  170. hypergan-2.0.0a2/src/hypergan/image_cache.py +66 -0
  171. hypergan-2.0.0a2/src/hypergan/image_components.py +319 -0
  172. hypergan-2.0.0a2/src/hypergan/image_data.py +91 -0
  173. hypergan-2.0.0a2/src/hypergan/image_grids.py +151 -0
  174. hypergan-2.0.0a2/src/hypergan/image_metrics.py +117 -0
  175. hypergan-2.0.0a2/src/hypergan/inspection.py +2324 -0
  176. hypergan-2.0.0a2/src/hypergan/interval_evaluation.py +314 -0
  177. hypergan-2.0.0a2/src/hypergan/item_data.py +169 -0
  178. hypergan-2.0.0a2/src/hypergan/main_guard.py +71 -0
  179. hypergan-2.0.0a2/src/hypergan/metric_evaluation.py +456 -0
  180. hypergan-2.0.0a2/src/hypergan/metric_evaluation_worker.py +151 -0
  181. hypergan-2.0.0a2/src/hypergan/metric_examples.py +81 -0
  182. hypergan-2.0.0a2/src/hypergan/metric_plugins.py +339 -0
  183. hypergan-2.0.0a2/src/hypergan/metrics.py +447 -0
  184. hypergan-2.0.0a2/src/hypergan/metrics_reducer/__init__.py +12 -0
  185. hypergan-2.0.0a2/src/hypergan/model_description.py +659 -0
  186. hypergan-2.0.0a2/src/hypergan/model_file.py +1064 -0
  187. hypergan-2.0.0a2/src/hypergan/models/cifar10-tiny-transformer.toml +131 -0
  188. hypergan-2.0.0a2/src/hypergan/models/cifar10-transgan.toml +326 -0
  189. hypergan-2.0.0a2/src/hypergan/models/colorization-128.toml +206 -0
  190. hypergan-2.0.0a2/src/hypergan/models/gaussian-grid.toml +64 -0
  191. hypergan-2.0.0a2/src/hypergan/models/image-folder-128.toml +99 -0
  192. hypergan-2.0.0a2/src/hypergan/network_config.py +208 -0
  193. hypergan-2.0.0a2/src/hypergan/networks/autoencoder_offset.hndl +1 -0
  194. hypergan-2.0.0a2/src/hypergan/networks/colorization_conv_head.hndl +6 -0
  195. hypergan-2.0.0a2/src/hypergan/networks/colorization_dcgan256.hndl +14 -0
  196. hypergan-2.0.0a2/src/hypergan/networks/colorization_dinov3_multidepth.hndl +20 -0
  197. hypergan-2.0.0a2/src/hypergan/networks/colorization_dinov3_patch_tokens.hndl +19 -0
  198. hypergan-2.0.0a2/src/hypergan/networks/colorization_encoder_features.hndl +19 -0
  199. hypergan-2.0.0a2/src/hypergan/networks/colorization_feature_concat.hndl +2 -0
  200. hypergan-2.0.0a2/src/hypergan/networks/colorization_feature_project.hndl +3 -0
  201. hypergan-2.0.0a2/src/hypergan/networks/colorization_generator16.hndl +69 -0
  202. hypergan-2.0.0a2/src/hypergan/networks/colorization_generator32.hndl +69 -0
  203. hypergan-2.0.0a2/src/hypergan/networks/colorization_joint_score.hndl +2 -0
  204. hypergan-2.0.0a2/src/hypergan/networks/colorization_linear_head.hndl +3 -0
  205. hypergan-2.0.0a2/src/hypergan/networks/colorization_multidepth_projection.hndl +18 -0
  206. hypergan-2.0.0a2/src/hypergan/networks/colorization_multiscale_score.hndl +2 -0
  207. hypergan-2.0.0a2/src/hypergan/networks/colorization_pixel_features.hndl +8 -0
  208. hypergan-2.0.0a2/src/hypergan/networks/colorization_pixel_head.hndl +17 -0
  209. hypergan-2.0.0a2/src/hypergan/networks/colorization_query.hndl +2 -0
  210. hypergan-2.0.0a2/src/hypergan/networks/colorization_random_project.hndl +2 -0
  211. hypergan-2.0.0a2/src/hypergan/networks/colorization_scale_head16.hndl +8 -0
  212. hypergan-2.0.0a2/src/hypergan/networks/colorization_scale_head32.hndl +10 -0
  213. hypergan-2.0.0a2/src/hypergan/networks/colorization_scale_head4.hndl +4 -0
  214. hypergan-2.0.0a2/src/hypergan/networks/colorization_scale_head8.hndl +6 -0
  215. hypergan-2.0.0a2/src/hypergan/networks/image_attention.hndl +15 -0
  216. hypergan-2.0.0a2/src/hypergan/networks/image_critic_score.hndl +6 -0
  217. hypergan-2.0.0a2/src/hypergan/networks/image_encoder.hndl +17 -0
  218. hypergan-2.0.0a2/src/hypergan/networks/image_feature_head.hndl +9 -0
  219. hypergan-2.0.0a2/src/hypergan/networks/image_generator.hndl +13 -0
  220. hypergan-2.0.0a2/src/hypergan/networks/image_pixel.hndl +7 -0
  221. hypergan-2.0.0a2/src/hypergan/networks/image_pixel_block.hndl +16 -0
  222. hypergan-2.0.0a2/src/hypergan/networks/image_resnet_stage1.hndl +7 -0
  223. hypergan-2.0.0a2/src/hypergan/networks/image_resnet_stage2.hndl +2 -0
  224. hypergan-2.0.0a2/src/hypergan/networks/image_resnet_stage3.hndl +2 -0
  225. hypergan-2.0.0a2/src/hypergan/networks/reference.hndl +6 -0
  226. hypergan-2.0.0a2/src/hypergan/networks/transgan-discriminator-32.hndl +88 -0
  227. hypergan-2.0.0a2/src/hypergan/numerical_policy.py +95 -0
  228. hypergan-2.0.0a2/src/hypergan/objective_program.py +506 -0
  229. hypergan-2.0.0a2/src/hypergan/observation_io.py +150 -0
  230. hypergan-2.0.0a2/src/hypergan/ops_info.py +66 -0
  231. hypergan-2.0.0a2/src/hypergan/particlegan_integration.py +143 -0
  232. hypergan-2.0.0a2/src/hypergan/plugin_functions.py +94 -0
  233. hypergan-2.0.0a2/src/hypergan/ports.py +9 -0
  234. hypergan-2.0.0a2/src/hypergan/pretrained_providers.py +144 -0
  235. hypergan-2.0.0a2/src/hypergan/preview_snapshot.py +229 -0
  236. hypergan-2.0.0a2/src/hypergan/preview_worker.py +128 -0
  237. hypergan-2.0.0a2/src/hypergan/previews.py +624 -0
  238. hypergan-2.0.0a2/src/hypergan/provenance.py +187 -0
  239. hypergan-2.0.0a2/src/hypergan/py.typed +0 -0
  240. hypergan-2.0.0a2/src/hypergan/qualify.py +887 -0
  241. hypergan-2.0.0a2/src/hypergan/queue.py +1190 -0
  242. hypergan-2.0.0a2/src/hypergan/recipe_generator.py +927 -0
  243. hypergan-2.0.0a2/src/hypergan/recipes.py +279 -0
  244. hypergan-2.0.0a2/src/hypergan/registry.py +1813 -0
  245. hypergan-2.0.0a2/src/hypergan/replicated_execution.py +598 -0
  246. hypergan-2.0.0a2/src/hypergan/replicated_worker.py +216 -0
  247. hypergan-2.0.0a2/src/hypergan/run_controller.py +1204 -0
  248. hypergan-2.0.0a2/src/hypergan/run_events.py +185 -0
  249. hypergan-2.0.0a2/src/hypergan/run_info.py +147 -0
  250. hypergan-2.0.0a2/src/hypergan/run_requests.py +360 -0
  251. hypergan-2.0.0a2/src/hypergan/run_signals.py +42 -0
  252. hypergan-2.0.0a2/src/hypergan/run_state.py +80 -0
  253. hypergan-2.0.0a2/src/hypergan/schema_info.py +327 -0
  254. hypergan-2.0.0a2/src/hypergan/single_execution.py +291 -0
  255. hypergan-2.0.0a2/src/hypergan/snapshot_renderer.py +71 -0
  256. hypergan-2.0.0a2/src/hypergan/summary.py +450 -0
  257. hypergan-2.0.0a2/src/hypergan/toml_writer.py +125 -0
  258. hypergan-2.0.0a2/src/hypergan/training.py +521 -0
  259. hypergan-2.0.0a2/src/hypergan/web_assets/THIRD_PARTY_LICENSES.txt +289 -0
  260. hypergan-2.0.0a2/src/hypergan/web_assets/__init__.py +25 -0
  261. hypergan-2.0.0a2/src/hypergan/web_assets/app.js +22 -0
  262. hypergan-2.0.0a2/src/hypergan/web_assets/index.html +294 -0
  263. hypergan-2.0.0a2/src/hypergan/web_assets/logo.png +0 -0
  264. hypergan-2.0.0a2/src/hypergan/web_assets/runs.css +758 -0
  265. hypergan-2.0.0a2/src/hypergan/web_assets/runs.js +530 -0
  266. hypergan-2.0.0a2/src/hypergan/web_assets/style.css +1189 -0
  267. hypergan-2.0.0a2/src/hypergan/web_assets/view-worker.js +204 -0
  268. hypergan-2.0.0a2/src/hypergan/web_autostart.py +1490 -0
  269. hypergan-2.0.0a2/src/hypergan/web_dev.py +132 -0
  270. hypergan-2.0.0a2/src/hypergan/web_files.py +83 -0
  271. hypergan-2.0.0a2/src/hypergan/web_launch.py +293 -0
  272. hypergan-2.0.0a2/src/hypergan/web_schema.py +165 -0
  273. hypergan-2.0.0a2/src/hypergan/web_server.py +1665 -0
  274. hypergan-2.0.0a2/src/hypergan/web_service.py +1310 -0
  275. hypergan-2.0.0a2/src/hypergan/web_session.py +306 -0
  276. hypergan-2.0.0a2/src/hypergan/workspace.py +346 -0
  277. hypergan-2.0.0a2/src/hypergan.egg-info/PKG-INFO +390 -0
  278. hypergan-2.0.0a2/src/hypergan.egg-info/SOURCES.txt +562 -0
  279. hypergan-2.0.0a2/src/hypergan.egg-info/entry_points.txt +2 -0
  280. hypergan-2.0.0a2/src/hypergan.egg-info/requires.txt +57 -0
  281. hypergan-2.0.0a2/tests/__init__.py +1 -0
  282. hypergan-2.0.0a2/tests/browser/test_reducer_browser.py +158 -0
  283. hypergan-2.0.0a2/tests/browser/test_run_index.py +408 -0
  284. hypergan-2.0.0a2/tests/browser/test_viewer_integration.py +878 -0
  285. hypergan-2.0.0a2/tests/browser/test_viewer_ui.py +894 -0
  286. hypergan-2.0.0a2/tests/config_fixtures.py +150 -0
  287. hypergan-2.0.0a2/tests/cuda/nccl_smoke.py +221 -0
  288. hypergan-2.0.0a2/tests/cuda/test_aggregation_2gpu_nccl.py +491 -0
  289. hypergan-2.0.0a2/tests/cuda/test_checkpoint_continuity_2gpu_nccl_challenges.py +444 -0
  290. hypergan-2.0.0a2/tests/cuda/test_cifar_replicated_qualification.py +239 -0
  291. hypergan-2.0.0a2/tests/cuda/test_cuda_two_devices.py +61 -0
  292. hypergan-2.0.0a2/tests/cuda/test_custom_metrics.py +87 -0
  293. hypergan-2.0.0a2/tests/cuda/test_hndl_images.py +54 -0
  294. hypergan-2.0.0a2/tests/cuda/test_interval_evaluation.py +34 -0
  295. hypergan-2.0.0a2/tests/cuda/test_m2_adversarial_2gpu_nccl.py +284 -0
  296. hypergan-2.0.0a2/tests/cuda/test_m6_empirical_challenges.py +516 -0
  297. hypergan-2.0.0a2/tests/cuda/test_metric_scalar_transfer.py +32 -0
  298. hypergan-2.0.0a2/tests/cuda/test_metrics.py +32 -0
  299. hypergan-2.0.0a2/tests/cuda/test_native_cuda.py +207 -0
  300. hypergan-2.0.0a2/tests/cuda/test_nccl.py +35 -0
  301. hypergan-2.0.0a2/tests/cuda/test_nccl_service.py +97 -0
  302. hypergan-2.0.0a2/tests/cuda/test_public_execution_acceptance.py +39 -0
  303. hypergan-2.0.0a2/tests/cuda/test_replicated_images.py +120 -0
  304. hypergan-2.0.0a2/tests/cuda/test_replicated_numerics.py +207 -0
  305. hypergan-2.0.0a2/tests/cuda/test_replicated_recovery.py +324 -0
  306. hypergan-2.0.0a2/tests/cuda/test_trainer_host_syncs.py +182 -0
  307. hypergan-2.0.0a2/tests/dinov3_fixtures.py +20 -0
  308. hypergan-2.0.0a2/tests/distributed_fixtures.py +119 -0
  309. hypergan-2.0.0a2/tests/e2e/__init__.py +1 -0
  310. hypergan-2.0.0a2/tests/e2e/conftest.py +45 -0
  311. hypergan-2.0.0a2/tests/e2e/harness.py +296 -0
  312. hypergan-2.0.0a2/tests/e2e/harness_v2.py +227 -0
  313. hypergan-2.0.0a2/tests/e2e/test_playbook_v2_e2e.py +894 -0
  314. hypergan-2.0.0a2/tests/e2e/test_tier1_aggregation.py +249 -0
  315. hypergan-2.0.0a2/tests/e2e/test_tier1_multinode.py +337 -0
  316. hypergan-2.0.0a2/tests/e2e/test_tier1_profile_replicated.py +491 -0
  317. hypergan-2.0.0a2/tests/e2e/test_tier2_boundaries.py +249 -0
  318. hypergan-2.0.0a2/tests/e2e/test_tier3_combinations.py +243 -0
  319. hypergan-2.0.0a2/tests/e2e/test_tier4_scenarios.py +197 -0
  320. hypergan-2.0.0a2/tests/e2e/test_tier5_adversarial_distributed.py +833 -0
  321. hypergan-2.0.0a2/tests/e2e_integration/__init__.py +1 -0
  322. hypergan-2.0.0a2/tests/e2e_integration/test_foldlite_e2e.py +836 -0
  323. hypergan-2.0.0a2/tests/fixtures/networks.toml +94 -0
  324. hypergan-2.0.0a2/tests/foundation/test_adversarial_m1.py +908 -0
  325. hypergan-2.0.0a2/tests/foundation/test_api.py +313 -0
  326. hypergan-2.0.0a2/tests/foundation/test_api_poc_config_first.py +248 -0
  327. hypergan-2.0.0a2/tests/foundation/test_background_poll.py +57 -0
  328. hypergan-2.0.0a2/tests/foundation/test_bounded_cli_output.py +343 -0
  329. hypergan-2.0.0a2/tests/foundation/test_bounded_observer.py +326 -0
  330. hypergan-2.0.0a2/tests/foundation/test_cache.py +225 -0
  331. hypergan-2.0.0a2/tests/foundation/test_cifar_tiny_feature_recipe.py +23 -0
  332. hypergan-2.0.0a2/tests/foundation/test_cli.py +468 -0
  333. hypergan-2.0.0a2/tests/foundation/test_cli_api_review.py +416 -0
  334. hypergan-2.0.0a2/tests/foundation/test_cli_inspection.py +411 -0
  335. hypergan-2.0.0a2/tests/foundation/test_cli_viewer_options.py +184 -0
  336. hypergan-2.0.0a2/tests/foundation/test_config.py +230 -0
  337. hypergan-2.0.0a2/tests/foundation/test_config_loading.py +169 -0
  338. hypergan-2.0.0a2/tests/foundation/test_console_ergonomics.py +242 -0
  339. hypergan-2.0.0a2/tests/foundation/test_console_settings.py +117 -0
  340. hypergan-2.0.0a2/tests/foundation/test_controller_fences.py +531 -0
  341. hypergan-2.0.0a2/tests/foundation/test_cpu_service_protocol.py +113 -0
  342. hypergan-2.0.0a2/tests/foundation/test_dataset_download.py +346 -0
  343. hypergan-2.0.0a2/tests/foundation/test_destructive_commands.py +582 -0
  344. hypergan-2.0.0a2/tests/foundation/test_distributed_commit.py +173 -0
  345. hypergan-2.0.0a2/tests/foundation/test_distribution.py +34 -0
  346. hypergan-2.0.0a2/tests/foundation/test_docker_profile.py +248 -0
  347. hypergan-2.0.0a2/tests/foundation/test_event_boundary.py +103 -0
  348. hypergan-2.0.0a2/tests/foundation/test_event_views.py +370 -0
  349. hypergan-2.0.0a2/tests/foundation/test_execution_profiles.py +422 -0
  350. hypergan-2.0.0a2/tests/foundation/test_extras_runtime_pins.py +30 -0
  351. hypergan-2.0.0a2/tests/foundation/test_file_lock.py +42 -0
  352. hypergan-2.0.0a2/tests/foundation/test_group_free_service.py +49 -0
  353. hypergan-2.0.0a2/tests/foundation/test_hndl_config.py +70 -0
  354. hypergan-2.0.0a2/tests/foundation/test_hndl_provider_config.py +112 -0
  355. hypergan-2.0.0a2/tests/foundation/test_image_grids.py +39 -0
  356. hypergan-2.0.0a2/tests/foundation/test_inheritance.py +270 -0
  357. hypergan-2.0.0a2/tests/foundation/test_interval_cancellation.py +106 -0
  358. hypergan-2.0.0a2/tests/foundation/test_lifecycle_acceptance.py +150 -0
  359. hypergan-2.0.0a2/tests/foundation/test_main_guard.py +128 -0
  360. hypergan-2.0.0a2/tests/foundation/test_metric_evaluation_files.py +85 -0
  361. hypergan-2.0.0a2/tests/foundation/test_metric_plugins.py +283 -0
  362. hypergan-2.0.0a2/tests/foundation/test_metrics_cli.py +67 -0
  363. hypergan-2.0.0a2/tests/foundation/test_metrics_config.py +229 -0
  364. hypergan-2.0.0a2/tests/foundation/test_model_description.py +189 -0
  365. hypergan-2.0.0a2/tests/foundation/test_model_file.py +328 -0
  366. hypergan-2.0.0a2/tests/foundation/test_observation_io.py +119 -0
  367. hypergan-2.0.0a2/tests/foundation/test_observer_cli.py +56 -0
  368. hypergan-2.0.0a2/tests/foundation/test_particlegan_upgrade.py +64 -0
  369. hypergan-2.0.0a2/tests/foundation/test_preflight_acceptance.py +104 -0
  370. hypergan-2.0.0a2/tests/foundation/test_preview_failure_boundaries.py +83 -0
  371. hypergan-2.0.0a2/tests/foundation/test_preview_index.py +68 -0
  372. hypergan-2.0.0a2/tests/foundation/test_profile_m1_challenges.py +299 -0
  373. hypergan-2.0.0a2/tests/foundation/test_progress_failure_severity.py +42 -0
  374. hypergan-2.0.0a2/tests/foundation/test_provenance.py +184 -0
  375. hypergan-2.0.0a2/tests/foundation/test_public_execution.py +564 -0
  376. hypergan-2.0.0a2/tests/foundation/test_queue.py +319 -0
  377. hypergan-2.0.0a2/tests/foundation/test_queue_lifecycle.py +555 -0
  378. hypergan-2.0.0a2/tests/foundation/test_queue_stress.py +644 -0
  379. hypergan-2.0.0a2/tests/foundation/test_recipe_generator.py +81 -0
  380. hypergan-2.0.0a2/tests/foundation/test_reducer_import.py +20 -0
  381. hypergan-2.0.0a2/tests/foundation/test_registry_resilience.py +454 -0
  382. hypergan-2.0.0a2/tests/foundation/test_replicated_execution.py +188 -0
  383. hypergan-2.0.0a2/tests/foundation/test_repository_consistency.py +276 -0
  384. hypergan-2.0.0a2/tests/foundation/test_run_signals.py +32 -0
  385. hypergan-2.0.0a2/tests/foundation/test_run_state.py +696 -0
  386. hypergan-2.0.0a2/tests/foundation/test_runtime_compatibility.py +242 -0
  387. hypergan-2.0.0a2/tests/foundation/test_sampling_particle_binding.py +22 -0
  388. hypergan-2.0.0a2/tests/foundation/test_scaling_benchmark.py +121 -0
  389. hypergan-2.0.0a2/tests/foundation/test_summary.py +317 -0
  390. hypergan-2.0.0a2/tests/foundation/test_training_runs_port.py +162 -0
  391. hypergan-2.0.0a2/tests/foundation/test_transgan_recipe.py +53 -0
  392. hypergan-2.0.0a2/tests/foundation/test_transgan_stable_recipe.py +68 -0
  393. hypergan-2.0.0a2/tests/foundation/test_typing_pep561.py +154 -0
  394. hypergan-2.0.0a2/tests/foundation/test_web_files.py +56 -0
  395. hypergan-2.0.0a2/tests/foundation/test_web_session.py +117 -0
  396. hypergan-2.0.0a2/tests/foundation/test_workspace.py +160 -0
  397. hypergan-2.0.0a2/tests/hndl_fixtures.py +41 -0
  398. hypergan-2.0.0a2/tests/playbooks/test_playbook_greenfield.py +325 -0
  399. hypergan-2.0.0a2/tests/playbooks/test_playbook_observatory_and_api.py +407 -0
  400. hypergan-2.0.0a2/tests/playbooks/test_playbook_power_cli.py +598 -0
  401. hypergan-2.0.0a2/tests/reducers/test_shared_reducer.py +174 -0
  402. hypergan-2.0.0a2/tests/reference/legacy_native_oracle.py +88 -0
  403. hypergan-2.0.0a2/tests/reference/test_accumulation_acceptance.py +434 -0
  404. hypergan-2.0.0a2/tests/reference/test_accumulation_replay.py +204 -0
  405. hypergan-2.0.0a2/tests/reference/test_accumulation_stochastic_loss.py +121 -0
  406. hypergan-2.0.0a2/tests/reference/test_adversarial_bindings.py +298 -0
  407. hypergan-2.0.0a2/tests/reference/test_adversarial_terms.py +268 -0
  408. hypergan-2.0.0a2/tests/reference/test_api_typing.py +68 -0
  409. hypergan-2.0.0a2/tests/reference/test_async_observer_signals.py +88 -0
  410. hypergan-2.0.0a2/tests/reference/test_bad_image_recovery.py +219 -0
  411. hypergan-2.0.0a2/tests/reference/test_cache.py +22 -0
  412. hypergan-2.0.0a2/tests/reference/test_checkpoint_aggregation_continuity_challenges.py +519 -0
  413. hypergan-2.0.0a2/tests/reference/test_checkpoint_best_and_branches.py +186 -0
  414. hypergan-2.0.0a2/tests/reference/test_checkpoint_compatibility.py +220 -0
  415. hypergan-2.0.0a2/tests/reference/test_checkpoint_continuity_matrix_challenges.py +531 -0
  416. hypergan-2.0.0a2/tests/reference/test_checkpoint_event_commit.py +118 -0
  417. hypergan-2.0.0a2/tests/reference/test_checkpoint_metric_evaluation.py +274 -0
  418. hypergan-2.0.0a2/tests/reference/test_checkpoint_policy.py +363 -0
  419. hypergan-2.0.0a2/tests/reference/test_cifar_components.py +105 -0
  420. hypergan-2.0.0a2/tests/reference/test_cifar_data.py +90 -0
  421. hypergan-2.0.0a2/tests/reference/test_cifar_feature_only_discriminator.py +83 -0
  422. hypergan-2.0.0a2/tests/reference/test_cifar_hndl_discriminator.py +177 -0
  423. hypergan-2.0.0a2/tests/reference/test_cli_new.py +328 -0
  424. hypergan-2.0.0a2/tests/reference/test_color_words.py +327 -0
  425. hypergan-2.0.0a2/tests/reference/test_colorization_128.py +304 -0
  426. hypergan-2.0.0a2/tests/reference/test_colorization_components.py +383 -0
  427. hypergan-2.0.0a2/tests/reference/test_colorization_data_metrics.py +208 -0
  428. hypergan-2.0.0a2/tests/reference/test_colorization_dcgan.py +57 -0
  429. hypergan-2.0.0a2/tests/reference/test_colorization_multiscale.py +121 -0
  430. hypergan-2.0.0a2/tests/reference/test_colorization_pixel_features.py +113 -0
  431. hypergan-2.0.0a2/tests/reference/test_core_cli.py +393 -0
  432. hypergan-2.0.0a2/tests/reference/test_cpu_job.py +82 -0
  433. hypergan-2.0.0a2/tests/reference/test_cpu_worker_service.py +212 -0
  434. hypergan-2.0.0a2/tests/reference/test_cpu_workers.py +108 -0
  435. hypergan-2.0.0a2/tests/reference/test_custom_metrics.py +305 -0
  436. hypergan-2.0.0a2/tests/reference/test_data_cli.py +35 -0
  437. hypergan-2.0.0a2/tests/reference/test_diff_augment.py +172 -0
  438. hypergan-2.0.0a2/tests/reference/test_dinov3_hndl_discriminator.py +180 -0
  439. hypergan-2.0.0a2/tests/reference/test_dinov3_projected_discriminator.py +115 -0
  440. hypergan-2.0.0a2/tests/reference/test_distributed.py +200 -0
  441. hypergan-2.0.0a2/tests/reference/test_distributed_acceptance.py +267 -0
  442. hypergan-2.0.0a2/tests/reference/test_distributed_checkpoint_acceptance.py +108 -0
  443. hypergan-2.0.0a2/tests/reference/test_distributed_checkpoints.py +432 -0
  444. hypergan-2.0.0a2/tests/reference/test_distributed_commit_paths.py +36 -0
  445. hypergan-2.0.0a2/tests/reference/test_distributed_training.py +222 -0
  446. hypergan-2.0.0a2/tests/reference/test_diversity_metrics.py +215 -0
  447. hypergan-2.0.0a2/tests/reference/test_execution_launch.py +68 -0
  448. hypergan-2.0.0a2/tests/reference/test_execution_preflight.py +207 -0
  449. hypergan-2.0.0a2/tests/reference/test_execution_profiles_particlegan.py +17 -0
  450. hypergan-2.0.0a2/tests/reference/test_hndl_augmentation.py +207 -0
  451. hypergan-2.0.0a2/tests/reference/test_hndl_networks.py +73 -0
  452. hypergan-2.0.0a2/tests/reference/test_image_cache.py +199 -0
  453. hypergan-2.0.0a2/tests/reference/test_image_data.py +266 -0
  454. hypergan-2.0.0a2/tests/reference/test_image_folder_inventory.py +232 -0
  455. hypergan-2.0.0a2/tests/reference/test_image_metrics.py +104 -0
  456. hypergan-2.0.0a2/tests/reference/test_image_phase_policy.py +303 -0
  457. hypergan-2.0.0a2/tests/reference/test_image_previews.py +643 -0
  458. hypergan-2.0.0a2/tests/reference/test_image_recovery.py +142 -0
  459. hypergan-2.0.0a2/tests/reference/test_image_workers.py +193 -0
  460. hypergan-2.0.0a2/tests/reference/test_interval_evaluation.py +208 -0
  461. hypergan-2.0.0a2/tests/reference/test_interval_schedule.py +80 -0
  462. hypergan-2.0.0a2/tests/reference/test_item_data.py +44 -0
  463. hypergan-2.0.0a2/tests/reference/test_lifecycle_acceptance.py +131 -0
  464. hypergan-2.0.0a2/tests/reference/test_logo_autoencoder.py +101 -0
  465. hypergan-2.0.0a2/tests/reference/test_metric_cli_observation.py +52 -0
  466. hypergan-2.0.0a2/tests/reference/test_metric_evaluation.py +228 -0
  467. hypergan-2.0.0a2/tests/reference/test_metric_generated_binding.py +115 -0
  468. hypergan-2.0.0a2/tests/reference/test_metric_scalar_transfer.py +28 -0
  469. hypergan-2.0.0a2/tests/reference/test_metrics_replicated.py +103 -0
  470. hypergan-2.0.0a2/tests/reference/test_metrics_runtime.py +120 -0
  471. hypergan-2.0.0a2/tests/reference/test_model_record.py +51 -0
  472. hypergan-2.0.0a2/tests/reference/test_multinode_checkpoint_continuity_matrix_challenges.py +719 -0
  473. hypergan-2.0.0a2/tests/reference/test_multinode_tcp.py +908 -0
  474. hypergan-2.0.0a2/tests/reference/test_native_devices.py +41 -0
  475. hypergan-2.0.0a2/tests/reference/test_native_observer_restoration.py +42 -0
  476. hypergan-2.0.0a2/tests/reference/test_native_step_oracle.py +260 -0
  477. hypergan-2.0.0a2/tests/reference/test_nonfinite_refusal_and_ema_fusion.py +285 -0
  478. hypergan-2.0.0a2/tests/reference/test_observation.py +396 -0
  479. hypergan-2.0.0a2/tests/reference/test_partial_freezing.py +136 -0
  480. hypergan-2.0.0a2/tests/reference/test_particlegan_integration.py +262 -0
  481. hypergan-2.0.0a2/tests/reference/test_particlegan_recipe_restoration.py +38 -0
  482. hypergan-2.0.0a2/tests/reference/test_preflight_acceptance.py +161 -0
  483. hypergan-2.0.0a2/tests/reference/test_pretrained_providers.py +303 -0
  484. hypergan-2.0.0a2/tests/reference/test_preview_snapshot.py +184 -0
  485. hypergan-2.0.0a2/tests/reference/test_preview_worker.py +343 -0
  486. hypergan-2.0.0a2/tests/reference/test_prior_sigma.py +95 -0
  487. hypergan-2.0.0a2/tests/reference/test_projected_dinov3_stable.py +133 -0
  488. hypergan-2.0.0a2/tests/reference/test_public_execution_acceptance.py +292 -0
  489. hypergan-2.0.0a2/tests/reference/test_public_runtime_boundaries.py +358 -0
  490. hypergan-2.0.0a2/tests/reference/test_recovery.py +375 -0
  491. hypergan-2.0.0a2/tests/reference/test_reference.py +181 -0
  492. hypergan-2.0.0a2/tests/reference/test_released_recipe_config.py +119 -0
  493. hypergan-2.0.0a2/tests/reference/test_replicated_evaluation_snapshot.py +145 -0
  494. hypergan-2.0.0a2/tests/reference/test_replicated_fault_acceptance.py +190 -0
  495. hypergan-2.0.0a2/tests/reference/test_replicated_inference.py +67 -0
  496. hypergan-2.0.0a2/tests/reference/test_replicated_interval_evaluation.py +183 -0
  497. hypergan-2.0.0a2/tests/reference/test_replicated_job_acceptance.py +501 -0
  498. hypergan-2.0.0a2/tests/reference/test_replicated_observer_acceptance.py +342 -0
  499. hypergan-2.0.0a2/tests/reference/test_replicated_signals.py +67 -0
  500. hypergan-2.0.0a2/tests/reference/test_replicated_strategy_alignment_challenges.py +460 -0
  501. hypergan-2.0.0a2/tests/reference/test_resnet_colorization.py +190 -0
  502. hypergan-2.0.0a2/tests/reference/test_runtime_model_description.py +103 -0
  503. hypergan-2.0.0a2/tests/reference/test_sagan_adain_generator.py +165 -0
  504. hypergan-2.0.0a2/tests/reference/test_sample_diversity.py +64 -0
  505. hypergan-2.0.0a2/tests/reference/test_scaling_benchmark.py +27 -0
  506. hypergan-2.0.0a2/tests/reference/test_simple_styletransformer_generator.py +118 -0
  507. hypergan-2.0.0a2/tests/reference/test_source_provenance.py +92 -0
  508. hypergan-2.0.0a2/tests/reference/test_tiny_transformer_generator.py +45 -0
  509. hypergan-2.0.0a2/tests/reference/test_training_signals.py +87 -0
  510. hypergan-2.0.0a2/tests/reference/test_transgan_discriminator_32.py +81 -0
  511. hypergan-2.0.0a2/tests/reference/test_transgan_e2_generator.py +100 -0
  512. hypergan-2.0.0a2/tests/reference/test_transgan_e3_generator.py +107 -0
  513. hypergan-2.0.0a2/tests/reference/test_transgan_equalized_generator.py +77 -0
  514. hypergan-2.0.0a2/tests/reference/test_transgan_generator.py +319 -0
  515. hypergan-2.0.0a2/tests/reference/test_transgan_generator_32.py +110 -0
  516. hypergan-2.0.0a2/tests/reference/test_transgan_stable_generator.py +113 -0
  517. hypergan-2.0.0a2/tests/reference/test_worker_service_acceptance.py +414 -0
  518. hypergan-2.0.0a2/tests/unit/test_aggregation.py +323 -0
  519. hypergan-2.0.0a2/tests/unit/test_aggregation_m2_challenges.py +557 -0
  520. hypergan-2.0.0a2/tests/unit/test_challenger_m2_adversarial.py +804 -0
  521. hypergan-2.0.0a2/tests/unit/test_challenger_m3_2_1_adversarial.py +160 -0
  522. hypergan-2.0.0a2/tests/unit/test_challenger_m3_2_adversarial.py +262 -0
  523. hypergan-2.0.0a2/tests/unit/test_challenger_m3_adversarial.py +389 -0
  524. hypergan-2.0.0a2/tests/unit/test_challenger_m4_1_adversarial.py +573 -0
  525. hypergan-2.0.0a2/tests/unit/test_challenger_m4_2_adversarial.py +587 -0
  526. hypergan-2.0.0a2/tests/unit/test_challenger_m4_it2_adversarial.py +518 -0
  527. hypergan-2.0.0a2/tests/unit/test_cluster_autodetection.py +152 -0
  528. hypergan-2.0.0a2/tests/unit/test_coalesced_collectives.py +117 -0
  529. hypergan-2.0.0a2/tests/unit/test_conditional_sample.py +195 -0
  530. hypergan-2.0.0a2/tests/unit/test_distributed_review_fixes.py +299 -0
  531. hypergan-2.0.0a2/tests/unit/test_documentation_snippets.py +238 -0
  532. hypergan-2.0.0a2/tests/unit/test_elastic_topology_resume.py +497 -0
  533. hypergan-2.0.0a2/tests/unit/test_gpu_inventory.py +792 -0
  534. hypergan-2.0.0a2/tests/unit/test_m1_adversarial_sweep.py +463 -0
  535. hypergan-2.0.0a2/tests/unit/test_m1_it2_adversarial_deep_stress.py +426 -0
  536. hypergan-2.0.0a2/tests/unit/test_m1_it2_state_isolation.py +285 -0
  537. hypergan-2.0.0a2/tests/unit/test_m2_adversarial_collectives.py +443 -0
  538. hypergan-2.0.0a2/tests/unit/test_milestone3_features.py +244 -0
  539. hypergan-2.0.0a2/tests/unit/test_milestone4_features.py +449 -0
  540. hypergan-2.0.0a2/tests/unit/test_model_file_multinode_toml.py +110 -0
  541. hypergan-2.0.0a2/tests/unit/test_multi_rank_parity_challenges.py +511 -0
  542. hypergan-2.0.0a2/tests/unit/test_multigpu_readiness.py +47 -0
  543. hypergan-2.0.0a2/tests/unit/test_multigpu_telemetry_and_packaged_cli.py +615 -0
  544. hypergan-2.0.0a2/tests/unit/test_multinode_cli.py +153 -0
  545. hypergan-2.0.0a2/tests/unit/test_multinode_topology.py +271 -0
  546. hypergan-2.0.0a2/tests/unit/test_nfs_free_checkpoints.py +85 -0
  547. hypergan-2.0.0a2/tests/unit/test_qualify.py +452 -0
  548. hypergan-2.0.0a2/tests/unit/test_qualify_gpu_selection.py +97 -0
  549. hypergan-2.0.0a2/tests/unit/test_qualify_m4_challenges.py +263 -0
  550. hypergan-2.0.0a2/tests/unit/test_replicated_recipe_parity.py +292 -0
  551. hypergan-2.0.0a2/tests/unit/test_rocm_determinism.py +156 -0
  552. hypergan-2.0.0a2/tests/unit/test_rocm_preflight.py +475 -0
  553. hypergan-2.0.0a2/tests/unit/test_rocm_runtime.py +254 -0
  554. hypergan-2.0.0a2/tests/unit/test_tier5_qualification_challenges.py +327 -0
  555. hypergan-2.0.0a2/tests/unit/test_viewer_cache_isolation.py +15 -0
  556. hypergan-2.0.0a2/tests/web/test_metric_consistency.py +72 -0
  557. hypergan-2.0.0a2/tests/web/test_multi_run_dashboard.py +809 -0
  558. hypergan-2.0.0a2/tests/web/test_observatory_cache.py +453 -0
  559. hypergan-2.0.0a2/tests/web/test_observatory_telemetry.py +709 -0
  560. hypergan-2.0.0a2/tests/web/test_public_origin_proxy.py +156 -0
  561. hypergan-2.0.0a2/tests/web/test_viewer_dev_mode.py +125 -0
  562. hypergan-2.0.0a2/tests/web/test_web_autostart.py +787 -0
  563. hypergan-2.0.0a2/tests/web/test_web_service.py +1721 -0
  564. hypergan-1.0.6/MANIFEST.in +0 -1
  565. hypergan-1.0.6/PKG-INFO +0 -29
  566. hypergan-1.0.6/README.md +0 -227
  567. hypergan-1.0.6/bin/hypergan +0 -146
  568. hypergan-1.0.6/hypergan/__init__.py +0 -22
  569. hypergan-1.0.6/hypergan/accelerators/gpu_backend.py +0 -2
  570. hypergan-1.0.6/hypergan/cli.py +0 -215
  571. hypergan-1.0.6/hypergan/configurable_component.py +0 -753
  572. hypergan-1.0.6/hypergan/configuration.py +0 -46
  573. hypergan-1.0.6/hypergan/configurations/__init__.py +0 -3
  574. hypergan-1.0.6/hypergan/configurations/aligned_interpolated.json +0 -90
  575. hypergan-1.0.6/hypergan/configurations/classification.json +0 -79
  576. hypergan-1.0.6/hypergan/configurations/colorizer.json +0 -84
  577. hypergan-1.0.6/hypergan/configurations/components/losses/ali_loss/ali_loss_1.json +0 -64
  578. hypergan-1.0.6/hypergan/configurations/components/losses/f_divergence_loss/f_divergence_loss_1.json +0 -65
  579. hypergan-1.0.6/hypergan/configurations/components/losses/f_divergence_loss/f_divergence_loss_2.json +0 -67
  580. hypergan-1.0.6/hypergan/configurations/components/losses/f_divergence_loss/f_divergence_loss_3.json +0 -65
  581. hypergan-1.0.6/hypergan/configurations/components/losses/least_squares_loss/least_squares_loss_1.json +0 -64
  582. hypergan-1.0.6/hypergan/configurations/components/losses/logistic_loss/logistic_loss_1.json +0 -64
  583. hypergan-1.0.6/hypergan/configurations/components/losses/qp_loss/qp_loss_1.json +0 -64
  584. hypergan-1.0.6/hypergan/configurations/components/losses/ragan_loss/ragan_loss_1.json +0 -65
  585. hypergan-1.0.6/hypergan/configurations/components/losses/ragan_loss/ragan_loss_2.json +0 -65
  586. hypergan-1.0.6/hypergan/configurations/components/losses/realness_loss/realness_loss_1.json +0 -81
  587. hypergan-1.0.6/hypergan/configurations/components/losses/softmax_loss/softmax_loss_1.json +0 -64
  588. hypergan-1.0.6/hypergan/configurations/components/losses/standard_loss/standard_loss_1.json +0 -64
  589. hypergan-1.0.6/hypergan/configurations/components/losses/wasserstein_loss/wasserstein_loss_1.json +0 -64
  590. hypergan-1.0.6/hypergan/configurations/components/losses/wasserstein_loss/wasserstein_loss_2.json +0 -65
  591. hypergan-1.0.6/hypergan/configurations/components/trainers/accumulate_gradients/accumulate_gradients_1.json +0 -77
  592. hypergan-1.0.6/hypergan/configurations/components/trainers/accumulate_gradients/accumulate_gradients_2.json +0 -78
  593. hypergan-1.0.6/hypergan/configurations/components/trainers/alternating_trainer/alternating_trainer_1.json +0 -86
  594. hypergan-1.0.6/hypergan/configurations/components/trainers/balanced_trainer/balanced_trainer_1.json +0 -93
  595. hypergan-1.0.6/hypergan/configurations/components/trainers/balanced_trainer/balanced_trainer_2.json +0 -114
  596. hypergan-1.0.6/hypergan/configurations/components/trainers/simultaneous_trainer/simultaneous_trainer_1.json +0 -76
  597. hypergan-1.0.6/hypergan/configurations/configurable.json +0 -91
  598. hypergan-1.0.6/hypergan/configurations/dcgan.json +0 -83
  599. hypergan-1.0.6/hypergan/configurations/default.json +0 -88
  600. hypergan-1.0.6/hypergan/configurations/generate_readme.py +0 -17
  601. hypergan-1.0.6/hypergan/configurations/generate_samples.py +0 -23
  602. hypergan-1.0.6/hypergan/configurations/model_scaling.json +0 -88
  603. hypergan-1.0.6/hypergan/configurations/replace_hyperparms.py +0 -14
  604. hypergan-1.0.6/hypergan/configurations/run_all.py +0 -89
  605. hypergan-1.0.6/hypergan/configurations/stylegan.json +0 -85
  606. hypergan-1.0.6/hypergan/discriminators/__init__.py +0 -7
  607. hypergan-1.0.6/hypergan/discriminators/base_discriminator.py +0 -11
  608. hypergan-1.0.6/hypergan/discriminators/configurable_discriminator.py +0 -14
  609. hypergan-1.0.6/hypergan/discriminators/dcgan_discriminator.py +0 -28
  610. hypergan-1.0.6/hypergan/distributions/__init__.py +0 -7
  611. hypergan-1.0.6/hypergan/distributions/base_distribution.py +0 -4
  612. hypergan-1.0.6/hypergan/distributions/fitness_distribution.py +0 -57
  613. hypergan-1.0.6/hypergan/distributions/optimize_distribution.py +0 -70
  614. hypergan-1.0.6/hypergan/distributions/truncated_normal_distribution.py +0 -48
  615. hypergan-1.0.6/hypergan/distributions/uniform_distribution.py +0 -79
  616. hypergan-1.0.6/hypergan/gan.py +0 -16
  617. hypergan-1.0.6/hypergan/gan_component.py +0 -99
  618. hypergan-1.0.6/hypergan/gans/__init__.py +0 -3
  619. hypergan-1.0.6/hypergan/gans/ali_gan.py +0 -141
  620. hypergan-1.0.6/hypergan/gans/aligned_gan.py +0 -65
  621. hypergan-1.0.6/hypergan/gans/aligned_interpolate_gan.py +0 -103
  622. hypergan-1.0.6/hypergan/gans/base_gan.py +0 -270
  623. hypergan-1.0.6/hypergan/gans/configurable_gan.py +0 -143
  624. hypergan-1.0.6/hypergan/gans/encoder_gan.py +0 -69
  625. hypergan-1.0.6/hypergan/gans/needs_pytorch/ali_gan_combined.py +0 -271
  626. hypergan-1.0.6/hypergan/gans/needs_pytorch/ali_style_gan.py +0 -268
  627. hypergan-1.0.6/hypergan/gans/needs_pytorch/ali_vib_gan.py +0 -313
  628. hypergan-1.0.6/hypergan/gans/needs_pytorch/alialpha_gan.py +0 -236
  629. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_gan.py +0 -234
  630. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_gan3.py +0 -259
  631. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_gan6.py +0 -282
  632. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_gan7.py +0 -336
  633. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_gan8.py +0 -419
  634. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_gan_test.py +0 -338
  635. hypergan-1.0.6/hypergan/gans/needs_pytorch/aligned_ali_one_gan.py +0 -364
  636. hypergan-1.0.6/hypergan/gans/needs_pytorch/alpha_gan.py +0 -269
  637. hypergan-1.0.6/hypergan/gans/needs_pytorch/autoencoder_gan.py +0 -52
  638. hypergan-1.0.6/hypergan/gans/needs_pytorch/conditional_gan.py +0 -259
  639. hypergan-1.0.6/hypergan/gans/needs_pytorch/distribution_filtering_gan.py +0 -69
  640. hypergan-1.0.6/hypergan/gans/needs_pytorch/multi_generator_gan.py +0 -201
  641. hypergan-1.0.6/hypergan/gans/standard_gan.py +0 -100
  642. hypergan-1.0.6/hypergan/generators/__init__.py +0 -7
  643. hypergan-1.0.6/hypergan/generators/base_generator.py +0 -33
  644. hypergan-1.0.6/hypergan/generators/configurable_generator.py +0 -14
  645. hypergan-1.0.6/hypergan/generators/dcgan_generator.py +0 -34
  646. hypergan-1.0.6/hypergan/inputs/__init__.py +0 -7
  647. hypergan-1.0.6/hypergan/inputs/celeba_loader.py +0 -77
  648. hypergan-1.0.6/hypergan/inputs/crop_resize_transform.py +0 -50
  649. hypergan-1.0.6/hypergan/inputs/fitness_image_loader.py +0 -51
  650. hypergan-1.0.6/hypergan/inputs/image_loader.py +0 -77
  651. hypergan-1.0.6/hypergan/inputs/unsupervised_image_folder.py +0 -63
  652. hypergan-1.0.6/hypergan/layer.py +0 -13
  653. hypergan-1.0.6/hypergan/layer_shape.py +0 -35
  654. hypergan-1.0.6/hypergan/layers/__init__.py +0 -25
  655. hypergan-1.0.6/hypergan/layers/add.py +0 -37
  656. hypergan-1.0.6/hypergan/layers/cat.py +0 -39
  657. hypergan-1.0.6/hypergan/layers/channel_attention.py +0 -41
  658. hypergan-1.0.6/hypergan/layers/ez_norm.py +0 -72
  659. hypergan-1.0.6/hypergan/layers/layer.py +0 -63
  660. hypergan-1.0.6/hypergan/layers/linear_attention.py +0 -58
  661. hypergan-1.0.6/hypergan/layers/mul.py +0 -39
  662. hypergan-1.0.6/hypergan/layers/multi_head_attention.py +0 -61
  663. hypergan-1.0.6/hypergan/layers/operation.py +0 -92
  664. hypergan-1.0.6/hypergan/layers/pixel_shuffle.py +0 -39
  665. hypergan-1.0.6/hypergan/layers/residual.py +0 -57
  666. hypergan-1.0.6/hypergan/layers/resizable_stack.py +0 -119
  667. hypergan-1.0.6/hypergan/layers/segment_softmax.py +0 -63
  668. hypergan-1.0.6/hypergan/layers/upsample.py +0 -56
  669. hypergan-1.0.6/hypergan/losses/__init__.py +0 -7
  670. hypergan-1.0.6/hypergan/losses/ali_loss.py +0 -13
  671. hypergan-1.0.6/hypergan/losses/base_loss.py +0 -23
  672. hypergan-1.0.6/hypergan/losses/f_divergence_loss.py +0 -213
  673. hypergan-1.0.6/hypergan/losses/least_squares_loss.py +0 -17
  674. hypergan-1.0.6/hypergan/losses/logistic_loss.py +0 -16
  675. hypergan-1.0.6/hypergan/losses/qp_loss.py +0 -24
  676. hypergan-1.0.6/hypergan/losses/ragan_loss.py +0 -42
  677. hypergan-1.0.6/hypergan/losses/realness_loss.py +0 -70
  678. hypergan-1.0.6/hypergan/losses/softmax_loss.py +0 -18
  679. hypergan-1.0.6/hypergan/losses/special_d_loss.py +0 -30
  680. hypergan-1.0.6/hypergan/losses/standard_loss.py +0 -20
  681. hypergan-1.0.6/hypergan/losses/wasserstein_loss.py +0 -20
  682. hypergan-1.0.6/hypergan/modules/adaptive_instance_norm.py +0 -44
  683. hypergan-1.0.6/hypergan/modules/attention.py +0 -25
  684. hypergan-1.0.6/hypergan/modules/concat_noise.py +0 -12
  685. hypergan-1.0.6/hypergan/modules/const.py +0 -10
  686. hypergan-1.0.6/hypergan/modules/hardsigmoid.py +0 -11
  687. hypergan-1.0.6/hypergan/modules/learned_noise.py +0 -16
  688. hypergan-1.0.6/hypergan/modules/linformer.py +0 -194
  689. hypergan-1.0.6/hypergan/modules/modulated_conv2d.py +0 -353
  690. hypergan-1.0.6/hypergan/modules/multi_head_attention.py +0 -28
  691. hypergan-1.0.6/hypergan/modules/no_op.py +0 -7
  692. hypergan-1.0.6/hypergan/modules/pixel_norm.py +0 -9
  693. hypergan-1.0.6/hypergan/modules/reshape.py +0 -8
  694. hypergan-1.0.6/hypergan/modules/residual.py +0 -9
  695. hypergan-1.0.6/hypergan/modules/scaled_conv2d.py +0 -38
  696. hypergan-1.0.6/hypergan/modules/styled_conv2d.py +0 -41
  697. hypergan-1.0.6/hypergan/modules/variational.py +0 -18
  698. hypergan-1.0.6/hypergan/modules/zeros.py +0 -10
  699. hypergan-1.0.6/hypergan/optimizers/adamirror.py +0 -116
  700. hypergan-1.0.6/hypergan/optimizers/amsgrad.py +0 -169
  701. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/competitive2_optimizer.py +0 -266
  702. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/competitive_optimizer.py +0 -344
  703. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/consensus_optimizer.py +0 -68
  704. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/curl_optimizer.py +0 -177
  705. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/depth_optimizer.py +0 -92
  706. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/elastic_weight_consolidation_optimizer.py +0 -137
  707. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/ema_optimizer.py +0 -79
  708. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/gan_optimizer.py +0 -72
  709. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/giga_wolf_optimizer.py +0 -19
  710. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/gradient_magnitude_optimizer.py +0 -87
  711. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/jr_optimizer.py +0 -193
  712. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/local_nash_optimizer.py +0 -173
  713. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/negative_momentum_optimizer.py +0 -77
  714. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/orthonormal_optimizer.py +0 -81
  715. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/potential_optimizer.py +0 -135
  716. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/predictive_method_optimizer.py +0 -87
  717. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/sga_optimizer.py +0 -71
  718. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/social_optimizer.py +0 -95
  719. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/sos_optimizer.py +0 -174
  720. hypergan-1.0.6/hypergan/optimizers/needs_pytorch/tpu_negative_momentum_optimizer.py +0 -43
  721. hypergan-1.0.6/hypergan/optimizers/oadam.py +0 -111
  722. hypergan-1.0.6/hypergan/parser.py +0 -49
  723. hypergan-1.0.6/hypergan/samplers/__init__.py +0 -7
  724. hypergan-1.0.6/hypergan/samplers/aligned_batch_walk_sampler.py +0 -42
  725. hypergan-1.0.6/hypergan/samplers/aligned_sampler.py +0 -25
  726. hypergan-1.0.6/hypergan/samplers/base_sampler.py +0 -78
  727. hypergan-1.0.6/hypergan/samplers/batch_sampler.py +0 -17
  728. hypergan-1.0.6/hypergan/samplers/batch_walk_sampler.py +0 -42
  729. hypergan-1.0.6/hypergan/samplers/encoder_sampler.py +0 -32
  730. hypergan-1.0.6/hypergan/samplers/grid_sampler.py +0 -33
  731. hypergan-1.0.6/hypergan/samplers/input_sampler.py +0 -11
  732. hypergan-1.0.6/hypergan/samplers/needs_pytorch/debug_sampler.py +0 -119
  733. hypergan-1.0.6/hypergan/samplers/needs_pytorch/gang_sampler.py +0 -57
  734. hypergan-1.0.6/hypergan/samplers/needs_pytorch/progressive_sampler.py +0 -77
  735. hypergan-1.0.6/hypergan/samplers/needs_pytorch/random_walk_sampler.py +0 -42
  736. hypergan-1.0.6/hypergan/samplers/needs_pytorch/sorted_sampler.py +0 -59
  737. hypergan-1.0.6/hypergan/samplers/needs_pytorch/style_walk_sampler.py +0 -84
  738. hypergan-1.0.6/hypergan/samplers/static_batch_sampler.py +0 -18
  739. hypergan-1.0.6/hypergan/samplers/y_sampler.py +0 -82
  740. hypergan-1.0.6/hypergan/search/__init__.py +0 -3
  741. hypergan-1.0.6/hypergan/search/aligned_random_search.py +0 -67
  742. hypergan-1.0.6/hypergan/search/alphagan_random_search.py +0 -53
  743. hypergan-1.0.6/hypergan/search/default_configurations.py +0 -4
  744. hypergan-1.0.6/hypergan/search/random_search.py +0 -188
  745. hypergan-1.0.6/hypergan/threaded_tk_viewer.py +0 -288
  746. hypergan-1.0.6/hypergan/train_hooks/adversarial_norm_train_hook.py +0 -78
  747. hypergan-1.0.6/hypergan/train_hooks/base_train_hook.py +0 -36
  748. hypergan-1.0.6/hypergan/train_hooks/conjecture_train_hook.py +0 -125
  749. hypergan-1.0.6/hypergan/train_hooks/experimental/rolling_memory_2_train_hook.py +0 -212
  750. hypergan-1.0.6/hypergan/train_hooks/experimental/rolling_memory_train_hook.py +0 -126
  751. hypergan-1.0.6/hypergan/train_hooks/extragradient_train_hook.py +0 -51
  752. hypergan-1.0.6/hypergan/train_hooks/gradient_penalty_train_hook.py +0 -60
  753. hypergan-1.0.6/hypergan/train_hooks/initialize_as_autoencoder.py +0 -47
  754. hypergan-1.0.6/hypergan/train_hooks/match_support_train_hook.py +0 -52
  755. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/adversarial_robust_train_hook.py +0 -83
  756. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/competitive_train_hook.py +0 -204
  757. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/force_equilibrium_train_hook.py +0 -33
  758. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/gradient_locally_stable_train_hook.py +0 -35
  759. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/imle_train_hook.py +0 -97
  760. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/k_lipschitz_train_hook.py +0 -35
  761. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/learning_rate_dropout_train_hook.py +0 -48
  762. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/max_gp_train_hook.py +0 -74
  763. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/minibatch_train_hook.py +0 -93
  764. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/progress_compress_kbgan_train_hook.py +0 -129
  765. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/progress_compress_train_hook.py +0 -92
  766. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/self_supervised_train_hook.py +0 -62
  767. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/weight_constraint_train_hook.py +0 -205
  768. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/weight_penalty_train_hook.py +0 -96
  769. hypergan-1.0.6/hypergan/train_hooks/needs_pytorch/zero_penalty_train_hook.py +0 -36
  770. hypergan-1.0.6/hypergan/train_hooks/negative_momentum_train_hook.py +0 -22
  771. hypergan-1.0.6/hypergan/train_hooks/online_ewc_train_hook.py +0 -102
  772. hypergan-1.0.6/hypergan/train_hooks/stabilizing_training_train_hook.py +0 -52
  773. hypergan-1.0.6/hypergan/trainers/__init__.py +0 -7
  774. hypergan-1.0.6/hypergan/trainers/accumulate_gradient_trainer.py +0 -77
  775. hypergan-1.0.6/hypergan/trainers/alternating_trainer.py +0 -132
  776. hypergan-1.0.6/hypergan/trainers/balanced_trainer.py +0 -67
  777. hypergan-1.0.6/hypergan/trainers/base_trainer.py +0 -82
  778. hypergan-1.0.6/hypergan/trainers/needs_pytorch/competitive_alternating_trainer.py +0 -484
  779. hypergan-1.0.6/hypergan/trainers/needs_pytorch/competitive_trainer.py +0 -305
  780. hypergan-1.0.6/hypergan/trainers/needs_pytorch/consensus_trainer.py +0 -127
  781. hypergan-1.0.6/hypergan/trainers/needs_pytorch/curriculum_trainer.py +0 -92
  782. hypergan-1.0.6/hypergan/trainers/needs_pytorch/depth_trainer.py +0 -106
  783. hypergan-1.0.6/hypergan/trainers/needs_pytorch/evolution_trainer.py +0 -100
  784. hypergan-1.0.6/hypergan/trainers/needs_pytorch/gang_trainer.py +0 -574
  785. hypergan-1.0.6/hypergan/trainers/needs_pytorch/incremental_trainer.py +0 -29
  786. hypergan-1.0.6/hypergan/trainers/needs_pytorch/kbeam_trainer.py +0 -77
  787. hypergan-1.0.6/hypergan/trainers/needs_pytorch/line_search_trainer.py +0 -125
  788. hypergan-1.0.6/hypergan/trainers/needs_pytorch/multi_marginal_trainer.py +0 -85
  789. hypergan-1.0.6/hypergan/trainers/needs_pytorch/multi_step_trainer.py +0 -60
  790. hypergan-1.0.6/hypergan/trainers/needs_pytorch/multi_trainer_trainer.py +0 -33
  791. hypergan-1.0.6/hypergan/trainers/needs_pytorch/proportional_control_trainer.py +0 -56
  792. hypergan-1.0.6/hypergan/trainers/needs_pytorch/qualified_step_trainer.py +0 -127
  793. hypergan-1.0.6/hypergan/trainers/simultaneous_trainer.py +0 -69
  794. hypergan-1.0.6/hypergan/viewer.py +0 -12
  795. hypergan-1.0.6/hypergan.egg-info/PKG-INFO +0 -29
  796. hypergan-1.0.6/hypergan.egg-info/SOURCES.txt +0 -236
  797. hypergan-1.0.6/setup.cfg +0 -7
  798. hypergan-1.0.6/setup.py +0 -41
  799. {hypergan-1.0.6 → hypergan-2.0.0a2/src}/hypergan.egg-info/dependency_links.txt +0 -0
  800. {hypergan-1.0.6 → hypergan-2.0.0a2/src}/hypergan.egg-info/top_level.txt +0 -0
@@ -0,0 +1,21 @@
1
+ The MIT License (MIT)
2
+
3
+ Copyright (c) 2016 255BITS
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
@@ -0,0 +1,34 @@
1
+ prune hypergan
2
+ include setup.py
3
+ include src/hypergan/py.typed
4
+ include src/hypergan/_build_provenance.json
5
+ prune docs
6
+ prune reports
7
+ prune .agents
8
+ prune .claude
9
+ prune examples
10
+ recursive-include examples *.toml *.hndl *.py *.md *.json
11
+ prune worktrees
12
+ prune tests
13
+ include tests/__init__.py
14
+ include tests/hndl_fixtures.py
15
+ include tests/dinov3_fixtures.py
16
+ include tests/distributed_fixtures.py
17
+ include tests/config_fixtures.py
18
+ recursive-include tests/fixtures *.toml
19
+ recursive-include tests/foundation *.py
20
+ recursive-include tests/reference *.py
21
+ recursive-include tests/cuda *.py
22
+ recursive-include tests/reducers *.py
23
+ recursive-include tests/browser *.py
24
+ recursive-include tests/web *.py
25
+ recursive-include tests/unit *.py
26
+ recursive-include tests/playbooks *.py
27
+ recursive-include tests/e2e *.py
28
+ recursive-include tests/e2e_integration *.py
29
+ recursive-include scripts *.py *.sh
30
+ recursive-include src/hypergan/web_assets *.html *.css *.js *.txt *.json *.png
31
+ recursive-include src/hypergan/networks *.hndl
32
+ recursive-include src/hypergan/models *.toml
33
+ recursive-include frontend *.json *.js *.mjs *.md
34
+ prune frontend/node_modules
@@ -0,0 +1,390 @@
1
+ Metadata-Version: 2.4
2
+ Name: hypergan
3
+ Version: 2.0.0a2
4
+ Summary: Train and deploy proven GAN recipes with reproducible workflows.
5
+ License-Expression: MIT
6
+ Project-URL: Homepage, https://github.com/HyperGAN/HyperGAN
7
+ Project-URL: Documentation, https://github.com/HyperGAN/HyperGAN/tree/master/docs
8
+ Project-URL: Repository, https://github.com/HyperGAN/HyperGAN
9
+ Project-URL: Changelog, https://github.com/HyperGAN/HyperGAN/blob/master/changelog.md
10
+ Project-URL: Issues, https://github.com/HyperGAN/HyperGAN/issues
11
+ Classifier: Development Status :: 3 - Alpha
12
+ Classifier: Programming Language :: Python :: 3
13
+ Classifier: Programming Language :: Python :: 3 :: Only
14
+ Classifier: Programming Language :: Python :: 3.11
15
+ Classifier: Programming Language :: Python :: 3.12
16
+ Classifier: Programming Language :: Python :: 3.13
17
+ Classifier: Programming Language :: Python :: 3.14
18
+ Classifier: Operating System :: OS Independent
19
+ Classifier: Intended Audience :: Science/Research
20
+ Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
21
+ Requires-Python: <3.15,>=3.11
22
+ Description-Content-Type: text/markdown
23
+ License-File: LICENSE
24
+ Requires-Dist: foldlitedb<0.2,>=0.1.2
25
+ Provides-Extra: train
26
+ Requires-Dist: hndl<1,>=0.11.0; extra == "train"
27
+ Requires-Dist: particlegan<0.10,>=0.9.5; extra == "train"
28
+ Requires-Dist: torch<3,>=2.6; extra == "train"
29
+ Requires-Dist: numpy<3,>=1.26; extra == "train"
30
+ Requires-Dist: safetensors<1,>=0.8; extra == "train"
31
+ Provides-Extra: image
32
+ Requires-Dist: Pillow<14,>=13.0.0; extra == "image"
33
+ Provides-Extra: cifar
34
+ Requires-Dist: torchvision<1,>=0.21; extra == "cifar"
35
+ Requires-Dist: Pillow<14,>=13.0.0; extra == "cifar"
36
+ Provides-Extra: fid
37
+ Requires-Dist: torch-fidelity<0.5,>=0.4.0; extra == "fid"
38
+ Requires-Dist: torchvision<1,>=0.21; extra == "fid"
39
+ Requires-Dist: Pillow<14,>=13.0.0; extra == "fid"
40
+ Requires-Dist: scipy<2.0,>=1.10; extra == "fid"
41
+ Provides-Extra: reducers
42
+ Requires-Dist: foldlitedb[reducers]<0.2,>=0.1.2; extra == "reducers"
43
+ Provides-Extra: web
44
+ Requires-Dist: foldlitedb[reducers]<0.2,>=0.1.2; extra == "web"
45
+ Requires-Dist: starlette<2.0,>=1.6; extra == "web"
46
+ Requires-Dist: uvicorn<1.0,>=0.53; extra == "web"
47
+ Provides-Extra: safetensors
48
+ Requires-Dist: safetensors<1,>=0.8; extra == "safetensors"
49
+ Provides-Extra: onnx
50
+ Requires-Dist: onnx<2.0,>=1.16; extra == "onnx"
51
+ Provides-Extra: all
52
+ Requires-Dist: foldlitedb[reducers]<0.2,>=0.1.2; extra == "all"
53
+ Requires-Dist: hndl<1,>=0.11.0; extra == "all"
54
+ Requires-Dist: numpy<3,>=1.26; extra == "all"
55
+ Requires-Dist: onnx<2.0,>=1.16; extra == "all"
56
+ Requires-Dist: particlegan<0.10,>=0.9.5; extra == "all"
57
+ Requires-Dist: Pillow<14,>=13.0.0; extra == "all"
58
+ Requires-Dist: safetensors<1,>=0.8; extra == "all"
59
+ Requires-Dist: scipy<2.0,>=1.10; extra == "all"
60
+ Requires-Dist: starlette<2.0,>=1.6; extra == "all"
61
+ Requires-Dist: torch<3,>=2.6; extra == "all"
62
+ Requires-Dist: torch-fidelity<0.5,>=0.4.0; extra == "all"
63
+ Requires-Dist: torchvision<1,>=0.21; extra == "all"
64
+ Requires-Dist: uvicorn<1.0,>=0.53; extra == "all"
65
+ Provides-Extra: test
66
+ Requires-Dist: pytest<10,>=8; extra == "test"
67
+ Provides-Extra: dev
68
+ Requires-Dist: build<2,>=1; extra == "dev"
69
+ Requires-Dist: pytest<10,>=8; extra == "dev"
70
+ Dynamic: license-file
71
+
72
+ # <img src="https://raw.githubusercontent.com/HyperGAN/HyperGAN/master/docs/assets/hypergan-logo.png" width="48" height="48" align="absmiddle" alt="HyperGAN logo" /> HyperGAN
73
+
74
+ ![HyperGAN 128×128 Logo Colorizer: Grayscale HG to 4×4 Colorized Grid](https://raw.githubusercontent.com/HyperGAN/HyperGAN/master/docs/assets/hypergan-colorizer-banner.png)
75
+
76
+ _HyperGAN 2.0 logo ([single badge](https://raw.githubusercontent.com/HyperGAN/HyperGAN/master/docs/assets/hypergan-logo.png) · [4×4 colorized grid](https://raw.githubusercontent.com/HyperGAN/HyperGAN/master/docs/assets/hypergan-logo-grid.png) · [grayscale HG source](https://raw.githubusercontent.com/HyperGAN/HyperGAN/master/docs/assets/hg-bw-128.png)) generated by conditioning the [128×128 HNDL particle colorizer](docs/colorization.md) ([`examples/archive/logos-colorization-128.toml`](examples/archive/logos-colorization-128.toml)) on a single grayscale **HG** monogram across 16 particle draws._
77
+
78
+ HyperGAN 2.0.0a2 is the second alpha release of HG 2: declarative `hypergan-model/1` TOML model files, HNDL neural graphs, configurable ParticleGAN primitives, distributed multi-GPU & multi-node training, a typed Python API (`import hypergan as hg`), a real-time Web Observatory, and standalone privacy-preserving model export.
79
+
80
+ The engine provides native CUDA acceleration, multi-GPU scaling (2.20× on dual RTX A6000), complete training checkpoints, a Docker-style CLI for managing runs, and packaged production recipes including [CIFAR-10 TransGAN](docs/cifar-recipe.md) and [128×128 / 256×256 Logo Colorization](docs/colorization.md). Lightweight CLI commands (`new`, `validate`, `ps`, `status`, `logs`, `inspect`, `tag`, `runs`, `queue`, `prune`, `rm`) do not import PyTorch.
81
+
82
+ ---
83
+
84
+ ## Install
85
+
86
+ Use Python 3.12+ for the tested training runtime. Lightweight package checks cover Python 3.11–3.14 on Linux, macOS and Windows; the CPU reference profile is tested on Linux.
87
+
88
+ Install the HyperGAN 2.0.0a2 alpha from PyPI:
89
+
90
+ ```sh
91
+ python -m venv .venv
92
+ . .venv/bin/activate
93
+ python -m pip install "hypergan[all]==2.0.0a2"
94
+ hypergan --help
95
+ python -m hypergan --version
96
+ ```
97
+
98
+ For NVIDIA GPU acceleration:
99
+
100
+ ```sh
101
+ python -m pip install torch torchvision --index-url https://download.pytorch.org/whl/cu124
102
+ python -m pip install "hypergan[all]==2.0.0a2"
103
+ ```
104
+
105
+ Or install from a source checkout:
106
+
107
+ ```sh
108
+ git clone https://github.com/HyperGAN/HyperGAN.git
109
+ cd HyperGAN
110
+ python -m venv .venv
111
+ . .venv/bin/activate
112
+ python -m pip install -e .
113
+ hypergan --help
114
+ python -m hypergan --version
115
+ ```
116
+
117
+ For local NVIDIA GPU training from source:
118
+
119
+ ```sh
120
+ python -m pip install torch torchvision --index-url https://download.pytorch.org/whl/cu124
121
+ python -m pip install -e ".[all]"
122
+ ```
123
+
124
+ For a CPU correctness fixture, install the CPU PyTorch wheel instead (`--index-url https://download.pytorch.org/whl/cpu`) and pass `--device cpu` (or `-d cpu`). Lightweight CLI commands (`new`, `validate`, `ps`, `status`, `logs`, `inspect`, `tag`, `runs`, `queue`, `prune`, `rm`) do not import PyTorch or ParticleGAN.
125
+
126
+ ---
127
+
128
+ ## Quickstart: Project Scaffolding & Starter Templates
129
+
130
+ Explore starter architectures (`dcgan`, `tiny-transformer`, `resnet`, `gaussian-grid`) or packaged recipes:
131
+
132
+ ```sh
133
+ hypergan new --list
134
+ hypergan recipes
135
+ ```
136
+
137
+ Scaffold a complete project workspace (`hypergan.toml`, `runs/`, and `configs/<template>.toml`). When `--dataset` points at an image folder, `hypergan new` inspects image headers without loading pixels into RAM, auto-detects the resolution (`32`, `64`, `128`, or `256`) and color mode (`RGB` or `L`), and sizes the generator/critic HNDL architectures automatically:
138
+
139
+ ```sh
140
+ # Image GAN project sized automatically from a folder of images
141
+ hypergan new --project my_gan --dataset ./images
142
+
143
+ # Or specify an explicit template and resolution (or 2D toy fixture on CPU)
144
+ hypergan new --project my_transformer --template tiny-transformer --resolution 64
145
+ hypergan new --project my_toy --template gaussian-grid --device cpu
146
+ ```
147
+
148
+ Validate the generated `format = "hypergan-model/1"` configuration (or the project directory itself):
149
+
150
+ ```sh
151
+ # Fast structural validation (no Torch imports)
152
+ hypergan validate my_gan/configs/dcgan.toml
153
+ hypergan validate my_gan
154
+
155
+ # Validate with inline overrides and emit resolved JSON
156
+ hypergan validate my_gan/configs/dcgan.toml -s train.batch_size=32 -s optimizer.lr=0.0002 --json
157
+
158
+ # Verify image dataset inventory/hashes (--data) and construct networks in a worker (--runtime)
159
+ hypergan validate my_gan/configs/dcgan.toml --data
160
+ hypergan validate my_gan/configs/dcgan.toml --runtime
161
+ ```
162
+
163
+ ---
164
+
165
+ ## Training & Docker-Style Live Run Management
166
+
167
+ Start training from a model `.toml` file. You can customize the run name (`-n`), override any config setting (`-s KEY=VALUE`), or select the execution device (`-d`):
168
+
169
+ ```sh
170
+ hypergan train my_gan/configs/dcgan.toml --steps 1000
171
+ hypergan train my_gan/configs/dcgan.toml -n exp-lr2e4 -s train.lr=0.0002 -s train.batch_size=32
172
+ hypergan train my_toy/configs/gaussian-grid.toml --steps 20 -d cpu --no-server
173
+ ```
174
+
175
+ Every run is registered in a daemonless global registry with a unique **Run ID** (`<short_id>`, e.g. `0608d849`), a memorable Docker-style **Slug** (`<adj>-<noun>`, e.g. `fleet-hawk`), and an experiment **Name** / tag. You can inspect and control runs from **any working directory** using zero arguments (which automatically targets the active or most recent run) or by passing `<slug>`, `<short_id>`, `<tag>`, or `<run_dir>`:
176
+
177
+ ```sh
178
+ # List active runs, live GPU VRAM usage, and queued jobs
179
+ hypergan ps
180
+
181
+ # List all historical runs (active, completed, stopped, failed) across the host/workspace
182
+ hypergan ps -a
183
+ hypergan ps -a --json
184
+
185
+ # Concise single-screen status summary (progress, speed, ETA, latest losses, best FID, checkpoints)
186
+ hypergan status
187
+ hypergan status fleet-hawk
188
+
189
+ # View or follow (-f) formatted training logs (Ctrl-C detaches without stopping training)
190
+ hypergan logs -n 50
191
+ hypergan logs -f fleet-hawk
192
+
193
+ # Request an immediate durable checkpoint at the next update boundary
194
+ hypergan checkpoint
195
+ hypergan checkpoint fleet-hawk
196
+
197
+ # Gracefully stop an active run (saves a complete checkpoint before exiting)
198
+ hypergan stop
199
+ hypergan stop fleet-hawk
200
+ hypergan stop --all
201
+ ```
202
+
203
+ ---
204
+
205
+ ## Post-Run Inspection, Tagging, Resuming, Sampling, Evaluation, Export & Cleanup
206
+
207
+ ### Tagging & Inspecting Runs
208
+ ```sh
209
+ # Rename or tag a run in its manifest, summary, and global registry
210
+ hypergan tag fleet-hawk baseline-dcgan32
211
+
212
+ # Inspect full run manifest JSON or model architecture/loss/optimizer summary
213
+ hypergan inspect baseline-dcgan32
214
+ hypergan model baseline-dcgan32
215
+ hypergan model baseline-dcgan32 --write # build on meta device and save RUN/model.json
216
+ ```
217
+
218
+ ### Resuming Training
219
+ Continue a stopped or completed run from its latest checkpoint (or a specific `--checkpoint`). When `lr_floor = 1.0` (constant learning rate), `--steps` can extend the total step budget seamlessly:
220
+
221
+ ```sh
222
+ hypergan resume baseline-dcgan32 --steps 2000
223
+ ```
224
+
225
+ ### Sampling, Evaluation & Standalone Export
226
+ ```sh
227
+ # Draw fresh samples from the EMA generator (writes PNG grid for image models or JSON)
228
+ hypergan sample baseline-dcgan32 --count 16 --seed 42
229
+
230
+ # Condition a trained colorizer on a single grayscale input image across 16 particles
231
+ hypergan sample logos-colorization-128 --input docs/assets/hg-bw-128.png --count 16 --output hg-grid.png
232
+
233
+ # Run configured snapshot evaluations (e.g. FID, color moments, sample diversity)
234
+ hypergan eval baseline-dcgan32
235
+ hypergan evaluate baseline-dcgan32 --metric sample_diversity
236
+
237
+ # Export a standalone, hash-verified EMA inference bundle (model.pt + model.sha256)
238
+ hypergan export baseline-dcgan32 --checkpoint latest -o exported_model.pt
239
+ ```
240
+
241
+ ### Comparing Runs & Multi-GPU Job Queue
242
+ ```sh
243
+ # List and filter runs, or compare metrics across step milestones side-by-side
244
+ hypergan runs ls --sort fid
245
+ hypergan runs compare baseline-dcgan32 exp-lr2e4
246
+ hypergan runs compare baseline-dcgan32 exp-lr2e4 --metric loss_g --at 500,1000,best,latest --json
247
+
248
+ # Schedule and run experiments sequentially across GPUs
249
+ hypergan queue status
250
+ hypergan queue add my_gan/configs/dcgan.toml -n job1 --steps 5000 -s train.lr=0.0005 -d gpu0
251
+ hypergan queue ls --json
252
+ hypergan queue run --once
253
+ hypergan queue stop --all
254
+ ```
255
+
256
+ ### Garbage Collection & Run Removal
257
+ ```sh
258
+ # Reclaim disk space by cleaning rebuildable views and compacting routine train events
259
+ hypergan prune baseline-dcgan32 --dry-run
260
+ hypergan gc baseline-dcgan32
261
+ hypergan checkpoints prune baseline-dcgan32 --dry-run
262
+
263
+ # Delete a run directory and unregister it from the global registry (-f stops if active)
264
+ hypergan rm exp-lr2e4
265
+ ```
266
+
267
+ ---
268
+
269
+ ## Web Observatory (`hypergan view`)
270
+
271
+ With `.[web]` installed, interactive training runs launch the local Web Observatory on `127.0.0.1:8765` by default. To bind across local network interfaces, pass `-b 0.0.0.0` (or `--bind 0.0.0.0`):
272
+
273
+ ```sh
274
+ # Start the global runs dashboard and per-run observatory
275
+ hypergan view
276
+ hypergan view my_gan/runs --port 8765 --open
277
+ hypergan view baseline-dcgan32 --port 8765
278
+ hypergan view -b 0.0.0.0
279
+
280
+ # Check status or stop a background dashboard
281
+ hypergan view --status
282
+ hypergan view --stop
283
+ ```
284
+
285
+ In non-interactive environments (CI, background jobs, AI coding agent sandboxes), viewer autostart is automatically suppressed (can also be disabled via `HYPERGAN_NO_SERVER=1` or `--no-server`).
286
+
287
+ The Web Observatory provides:
288
+ - **Global Runs Dashboard (`GET /` and `GET /api/runs`)**: Live-updating table of all runs with search/status filtering, sortable columns, progress bars, active throughput (`steps/s` and `samples/s`), latest/best FID, hardware topology, batch size, and disk usage. Click a row to open its Observatory and use the per-run controls.
289
+ - **Per-Run Observatory (`GET /runs/<id_or_slug_or_tag>`)**: Real-time learning curves backed by a WebAssembly streaming reducer, retained EMA sample history sliders, snapshot evaluation charts, and an inline **Model** section showing the ParticleGAN integration, loss metrics, and per-layer HNDL network tables.
290
+
291
+ See [Local browser and streaming API](docs/local-web.md) for authentication (`--auth token`), TLS proxy setup (`--public-origin`), and the full `/api/runs/*` and `/api/v1/runs/*` REST/SSE contract.
292
+
293
+ ---
294
+
295
+ ## Distributed Multi-GPU Scaling
296
+
297
+ HyperGAN features first-class distributed multi-GPU and multi-node training:
298
+ - Zero-TOML CLI flags: `--devices 0,1`, `--nodes 2`, `--strategy gradient_allreduce`.
299
+ - Memory-contiguous gradient bucketing and barrier elimination.
300
+ - Documented **2.20× scaling speedup (110.2% efficiency)** on dual NVIDIA RTX A6000 at `batch_size >= 256`, cutting per-GPU VRAM in half (see [Scaling Benchmark](docs/scaling-benchmark.md)).
301
+
302
+ ---
303
+
304
+ ## Python Developer API (`import hypergan as hg`) & Environment Variables
305
+
306
+ Everything available in the CLI is exposed as a clean, functional Python API at the top level of `hypergan` (lazy-loaded so importing `hypergan` never imports PyTorch until training or sampling runs):
307
+
308
+ ```python
309
+ import hypergan as hg
310
+
311
+
312
+ def main():
313
+ # 1. Scaffold or load a model configuration
314
+ info = hg.new(project="my_gan", template="dcgan", data_dir="./images", device="cpu")
315
+ model = hg.load(info["recipe_path"])
316
+ model = hg.override(model, {"train.steps": 100, "train.batch_size": 16})
317
+ warnings = hg.validate(model)
318
+
319
+ # 2. Train and inspect runs by slug, short ID, custom tag, or path
320
+ run = hg.train(model, "my_gan/runs/exp1", steps=100)
321
+ active_and_all = hg.ps(all_runs=True)
322
+ all_runs = hg.list_runs()
323
+
324
+ st = hg.status("exp1") # typed RunStatus with step, status, latest_metrics, best_fid, etc.
325
+ log_lines = hg.logs("exp1", tail=20) # list[str] of formatted log lines (silent on stdout)
326
+ manifest = hg.inspect("exp1") # dict of the run's manifest.json
327
+ hg.tag("exp1", "my-best-gan")
328
+
329
+ # 3. Sample, export, and reload standalone generator
330
+ sample = hg.sample("my-best-gan", count=8, seed=42)
331
+ bundle_path = hg.export("my-best-gan", checkpoint="latest")
332
+ generator = hg.load_bundle(bundle_path, device="cuda") # nn.Module LoadedGenerator
333
+ samples = generator.sample(count=16, seed=42)
334
+ walk = generator.interpolate(seed_a=100, seed_b=200, steps=10, mode="slerp")
335
+
336
+ # 4. Clean up
337
+ hg.prune("my-best-gan", dry_run=True)
338
+ hg.rm("my-best-gan")
339
+
340
+
341
+ # Training starts worker processes that re-import this script, so keep the
342
+ # work under the main guard.
343
+ if __name__ == "__main__":
344
+ main()
345
+ ```
346
+
347
+ `hg.train(model)` without a run directory starts a fresh `<runs_root>/<stem>-<adj>-<noun>` run on every call, as `hypergan train` does (an overridden model still uses its file's workspace and stem). Pass `name=` to name the directory instead; if that directory already holds a different configuration, the run starts in a sibling `<name>-<adj>-<noun>` directory, as the CLI does.
348
+
349
+ ### Environment Variables
350
+
351
+ | Variable | Purpose |
352
+ | --- | --- |
353
+ | `HYPERGAN_HOME` / `HYPERGAN_STATE_DIR` | Override the directory used for the global run registry (`runs.json` and `registry.lock`). Defaults to `~/.local/share/hypergan` (or `$XDG_DATA_HOME/hypergan`). Setting either variable isolates run discovery to that state directory. |
354
+ | `HYPERGAN_WORKSPACE` | Explicitly point to a workspace `hypergan.toml` file (or its parent directory) when running outside the workspace directory tree. |
355
+ | `HYPERGAN_QUEUE_FILE` | Override the path to the multi-GPU job queue state file (defaults to `<workspace>/.hypergan/queue.json` or `./.hypergan/queue.json`). |
356
+ | `HYPERGAN_DEBUG` | Set to `1` (`true`/`yes`/`on`) to print the full traceback when a CLI command fails (same as `hypergan --debug`). |
357
+ | `HYPERGAN_VIEWER_DEV` | Set to `1` (`true`/`yes`/`on`) to serve Web Observatory frontend assets fresh from disk on every request with auto-reload. |
358
+
359
+ ---
360
+
361
+ ## Configure the recipe
362
+
363
+ `hypergan new` generates `format = "hypergan-model/1"` TOML model files (`[data]`, `[prior]`, `[networks.generator]`, `[networks.discriminator]`, `[[losses]]`, `[train]`, `[metrics]`, `[evaluations.<id>]`, `[checkpoints]`, `[sampling]`). During validation and training, HyperGAN lowers the model file into the engine's resolved numerical recipe.
364
+
365
+ ParticleGAN supplies the released GAN formulation and its selected defaults for loss, learning rates, hyperparameters, normalization and optimizers. HyperGAN consumes its public recipe and owns networks, data, execution, checkpoints and metrics. See the [integration boundary](docs/particlegan.md). Networks are declared in HNDL (`source = "..."` or `hndl = "file.hndl"`) or importable Python modules (`module = "pkg.mod:Class"`), with full source recorded in every run.
366
+
367
+ See:
368
+ - [Configuration and component contracts](docs/configuration.md) and the [paired synthetic example](examples/paired-linear.toml)
369
+ - [Image data and preprocessing](docs/image-data.md)
370
+ - [Recovery and compatibility](docs/recovery.md) and [Run observation contract](docs/observation.md)
371
+ - [Public execution profiles](docs/execution.md), [Execution profiles and preflight](docs/execution-profiles.md), [Replicated run service](docs/replicated-run-service.md), [CUDA/NCCL workers](docs/replicated-cuda.md), and [Bounded previews and progress](docs/replicated-observation.md)
372
+
373
+ ## Development and migration
374
+
375
+ The map/reduce storage and shared Rust reducers now live in **Foldlite DB**.
376
+ This branch requires `foldlitedb>=0.1.2,<0.2`:
377
+
378
+ ```sh
379
+ python -m pip install "foldlitedb>=0.1.2,<0.2"
380
+ python -m pip install -e '.[dev,web]'
381
+ ```
382
+
383
+ CI accepts the repository variable `FOLDLITEDB_SOURCE` as a wheel URL or a pip
384
+ source reference pinned to an exact revision. Without that variable it uses the
385
+ configured package index, which requires Foldlite DB to have been released.
386
+ See the [Foldlite integration and cache contract](docs/foldlitedb.md).
387
+
388
+ Install `.[dev,train,image,web]` in the tested runtime environment, then run `python -m pytest` for CPU and lightweight correctness tests. That default selection deselects the `heavy` marker, which gates the tests that start real subprocesses, multi-rank jobs and worker services; run those deliberately with `python -m pytest -m heavy`. The default run covers `tests/foundation`, `tests/reference`, `tests/unit`, `tests/playbooks`, `tests/e2e` and `tests/e2e_integration`; the viewer, browser, reducer and CUDA tiers need extra dependencies or hardware and are run by path. [tests/README.md](tests/README.md) lists every tier and how to run it; see also [CUDA execution and validation](docs/cuda.md) and [CI and PyPI releases](docs/releasing.md). All guides are indexed in [docs/README.md](docs/README.md), and [CONTRIBUTING.md](CONTRIBUTING.md) covers the pull-request workflow.
389
+
390
+ Historical HyperGAN code and experiments are preserved in archive tags. Legacy configurations and checkpoints require their archived runtime; see [migration notes](docs/migration.md), [preservation evidence](reports/archive/resurrection-preservation-2026-09-18.md), [the September 2026 execution ledger](reports/archive/resurrection-status.md), and [resurrection plan](reports/archive/resurrecting-hypergan-plan-2026-09-18.md).