pixeltable 0.3.0__py3-none-any.whl → 0.3.1__py3-none-any.whl

{pixeltable-0.3.0.dist-info → pixeltable-0.3.1.dist-info}/RECORD

@@ -1,5 +1,5 @@
 pixeltable/__init__.py,sha256=c-Z21TqJUbzGQKsO4CS4lVFwWlZnVhpIR94cL6RNmDo,1411
-pixeltable/__version__.py,sha256=zP_zQgwZByDmFisB1NLZeZLdsIy7Xh24k7e1Y59IAY4,112
+pixeltable/__version__.py,sha256=zVI6ltlHGwKITzBi4TSc2jGBnWUb4mnqjF_PYoGqTuw,112
 pixeltable/catalog/__init__.py,sha256=u8Ods4ncTY7DI5w0jyHVC0QDVDLHAy8Rqia3qsM5OH8,517
 pixeltable/catalog/catalog.py,sha256=tyDyI5wQw7vV6_FChrp9qgGCRClcjiSdW3eygYT0p9s,7849
 pixeltable/catalog/column.py,sha256=ezeKoGl6aBTzSZBihDA6vdETcvyCguAD4OsmrxWs73A,9595
@@ -11,45 +11,50 @@ pixeltable/catalog/path.py,sha256=QgccEi_QOfaKt8YsR2zLtd_z7z7QQkU_1kprJFi2SPQ,16
 pixeltable/catalog/path_dict.py,sha256=V7YQM0QhUWr4FbgUiDUIweGbmMJr687QCm2p614Xrs0,6473
 pixeltable/catalog/schema_object.py,sha256=NUE6Fx6Km0UUJ6WcE9mOgL2UQCLBMQjINQLp_mljzKA,2382
 pixeltable/catalog/table.py,sha256=Gqb1tT9l8oB5eeeFuj45fy0_1fdCqizmRWvyjFH-U9Y,62404
-pixeltable/catalog/table_version.py,sha256=hbnAtd-0EsVFCuWE1yrmm_CLar6NNQDRevgDjduh2CY,58581
+pixeltable/catalog/table_version.py,sha256=wuvFYbr6WrwhDWC202tUIgF8T61P69WOtduBUDq5EsI,58601
 pixeltable/catalog/table_version_path.py,sha256=CczGbcz5ESq663arreri_p4chMZHozgG6k3y-ajkJN4,5787
 pixeltable/catalog/view.py,sha256=qUjou_z_YwKTqTIRUY4Q3JxWUEiKvl2pDi7DNOnw4fw,10709
-pixeltable/dataframe.py,sha256=Xd1xJ0Drj2WBC1B9fJ2TCFMI23HeMV7L7RZJi7Wm2ew,47735
-pixeltable/env.py,sha256=jxnHxliUpWegRbsT6lRz83nNqDG29bWAy5wyAYImNGg,31338
+pixeltable/dataframe.py,sha256=KrAB-l_qdArlT3pnlSH69k0mqs0AK2wFhQ4V_7Wcjyw,48639
+pixeltable/env.py,sha256=VMS4Kj-RVEeFgmLzuQ2sFRQFtG4dLRbBZ9VzVzvO7hk,34993
 pixeltable/exceptions.py,sha256=NuFY2WtkQpLfLHT_J70kOw9Tr0kEDkkgo-u7As4Gaq4,410
-pixeltable/exec/__init__.py,sha256=bWbIA7U7y1hC40VOSv_p9doaMuOB0i3r1O2UFNw0VEs,494
-pixeltable/exec/aggregation_node.py,sha256=0LdoPp_CR_UbcS60XkDw66SqlrQnw6Dy6KDWqi4PJ6k,4005
-pixeltable/exec/cache_prefetch_node.py,sha256=hrJA5PxLSmFRuZdZO4Nc-vy6_PSb9YO4rnJc6MSTPHE,12017
-pixeltable/exec/component_iteration_node.py,sha256=ABuXGbDRQWLGuaBnfK7bvOxCrz81vMMiAvXHHI8SX4c,4930
-pixeltable/exec/data_row_batch.py,sha256=qPN0GtBGt72Jnap-UN53VlUPOpYEvqJhp1nuTeJ_8Co,2925
+pixeltable/exec/__init__.py,sha256=E_ub_ftAJVebEmKHZs47egeJqMfAbbx4SS1l7jR-Nd0,489
+pixeltable/exec/aggregation_node.py,sha256=iE3UHHW0IbkHsY7kYJAg6m1UKyKUu2_tJuzgL_rQUVE,4038
+pixeltable/exec/cache_prefetch_node.py,sha256=9jwCzWuAhymff6xwvBietw2S_eYno6y2n59QWbzowns,12220
+pixeltable/exec/component_iteration_node.py,sha256=4DkEzekeuaKok-VjI0Y3FIMF7BdoKHNmX2tAM10hsOY,4730
+pixeltable/exec/data_row_batch.py,sha256=5WhdS3ahtCP3wROw4NPJjtvaQUAWKwyN1rHKbfdT6TU,3308
 pixeltable/exec/exec_context.py,sha256=-FbfYLcGOL7mviru5dE1A7x4YxLbdKoXBHN3OSbqbcg,1084
-pixeltable/exec/exec_node.py,sha256=3t_Ri_FQfc-dV91YGSq17Av3vPJth5mHC-LAVc-1Whc,3222
-pixeltable/exec/expr_eval_node.py,sha256=8IzcxHvXCj69RKFfrt4XEol88SGAlLWMl0bjAijX7Ug,11887
-pixeltable/exec/in_memory_data_node.py,sha256=uK3puLZDzUSJUJwGwkTMYoPqNIvE67jtQodhE2sCz2M,3452
-pixeltable/exec/row_update_node.py,sha256=b8yuKtkmI2Q6U-7svKbkjdM3Z85ddZoJgJgUa17j-YE,2773
-pixeltable/exec/sql_node.py,sha256=TOeSI7TIUh8SG3esExj5zaWgLrjmpqMN74iJlPq16D0,19546
+pixeltable/exec/exec_node.py,sha256=qDp63PFlMQUVwav0JIjuP0r0BH9EwW7s_NgqFE_Zn20,3735
+pixeltable/exec/expr_eval/__init__.py,sha256=t0o2T7MxfH09TI7KlzxEjPUySejn2vkK06LC8t4FjsQ,41
+pixeltable/exec/expr_eval/evaluators.py,sha256=HtcrWivzoGWzpEAYRgyKvFRGVKq4rMO-QpK7fZq4fE8,11349
+pixeltable/exec/expr_eval/expr_eval_node.py,sha256=37c5eevFMAvvYmSbqzcO4Wjxouy3R5a5vSnLLjifeic,19786
+pixeltable/exec/expr_eval/globals.py,sha256=UAlOAOJNHw4sJ3CXXk2Gh0qhOxEm2cGtNlkvvvebx1g,3895
+pixeltable/exec/expr_eval/row_buffer.py,sha256=Wjc6ZklxSW4-sCAMEGIzESII1uuC6Urzmq6JeUO3ALA,2737
+pixeltable/exec/expr_eval/schedulers.py,sha256=MqO0r5TaQhywozpRPL87S8YwZmKEc8sdEerX2LHDaCY,10962
+pixeltable/exec/in_memory_data_node.py,sha256=CwuwkA2xdyApMfPfSQamlLdbb-IQn7R7zeBBJiI8I7g,3479
+pixeltable/exec/row_update_node.py,sha256=MEvI3Yzw6YcUDTufSNVNVtyG83Hj6zHzEI9fvKtAgRQ,2859
+pixeltable/exec/sql_node.py,sha256=vMHNHzhrlk2Ej58vMZzk7hYAeHFHlBtkLwYZsOVn3CQ,19573
 pixeltable/exprs/__init__.py,sha256=zx5OTxfUpFs_U36CAO-81v6RCX49wqRw3eHaLfytvkE,1044
 pixeltable/exprs/arithmetic_expr.py,sha256=I_OESFmUHlMSDmMy699-2JnFb7KbRjvh1GF06u8rBQ8,6322
 pixeltable/exprs/array_slice.py,sha256=eSqtQRxvweW_CwTRgkDD189i-BmTJ48dsW3znRPDebg,2180
 pixeltable/exprs/column_property_ref.py,sha256=ycHqZ7_LTFz_gag_53T5KRN6RfZckpw83DvvtxmcynM,3788
-pixeltable/exprs/column_ref.py,sha256=Z2YAiZ0nMv_ecbw91qTc9VN6DB2hZj-_RNkogZcu4ng,10784
+pixeltable/exprs/column_ref.py,sha256=win5w5oU0isvpa6bcB9c3I-ufI50RvZ2YOnNcSq-NX0,10886
 pixeltable/exprs/comparison.py,sha256=w1FuhnGsJRoCLYX3ztIHZDywf8nZeT2tKKwRjIQa6Lc,5101
 pixeltable/exprs/compound_predicate.py,sha256=6N_zN4A_hKGLordNY0si-8KHeFWG1UtyeiTHcs69IrM,3846
-pixeltable/exprs/data_row.py,sha256=rLtKxlst9mK6684A5y-nsjBcalyKEcKAWcYCtNpK10w,9852
-pixeltable/exprs/expr.py,sha256=E1V_fU1qA4nkHLHTqGkxr8exJE6I9-b9LwNVXxGNTXo,30463
+pixeltable/exprs/data_row.py,sha256=11WAmLmkhrjsKvKfPwj0dg8IkbeLwVcbFpK6eq3VUsM,10577
+pixeltable/exprs/expr.py,sha256=JYjdweVm6z5Fw37_WhefGDaVN_2jQtUV9BiVwt20oYE,32203
 pixeltable/exprs/expr_dict.py,sha256=xkvo_iVPOLMq3WkBZQ2FOoXqYoebuV6XGlidPJxdOkY,1588
 pixeltable/exprs/expr_set.py,sha256=GeUQXadzJbAqQOGdsO6Z5hPAp0A03YUr2hyIvfZDYEE,2576
-pixeltable/exprs/function_call.py,sha256=5lMTj4G6jOarabZ0b3xpgAMJe3rgDdCSXbYwnrZEjxs,26196
+pixeltable/exprs/function_call.py,sha256=DrBFDU2cKVBj7TiDsAJFuElFMClg1RO9W8klnLNz8l0,28033
 pixeltable/exprs/globals.py,sha256=5pwn5vdi-EEpYBpPty658YV45myY7W0iFIfTH7QIzak,2032
 pixeltable/exprs/in_predicate.py,sha256=VxSn9TA_3UDHTC6rqQ12No5HbZO7SuE4DglpwAoI58s,3783
-pixeltable/exprs/inline_expr.py,sha256=oALVAM5BVFskEQgkfgGIhKFm6V2FUlrPtocGtqtXM0w,7867
+pixeltable/exprs/inline_expr.py,sha256=PRiW4kvxBmvbJD27Rr-xQzcL7f5YWmnHd3GuhwOGocw,7707
 pixeltable/exprs/is_null.py,sha256=rnusy1j33o48Z-qjguj09G7AmsBHFJI5SWQXOAjgKQw,1092
 pixeltable/exprs/json_mapper.py,sha256=yhefrgvy76czfyTER_oRXnfxOVQb42Q4Us1KkXhSsLM,4545
 pixeltable/exprs/json_path.py,sha256=UTf3VUcyEmEBJs6yU4kONGcZ8fy1o6mIe9MlntnUd6M,6811
-pixeltable/exprs/literal.py,sha256=TTKb0gw6qck9D61SwVDuBrLrBrGwEhkCB-m0ILFpWFk,3764
+pixeltable/exprs/literal.py,sha256=Nhh-fXYMqGL4Tfm2YiFUd4ALMjzG7nJ2H4jHZe9XS1Y,4394
 pixeltable/exprs/method_ref.py,sha256=3O_uFMP6wWGiwJWri8on-47EVl-QD3AiAc7hXJMADxs,2615
 pixeltable/exprs/object_ref.py,sha256=GVg6uxZnKwFVTC0kouJq-wMFP-gUPb_ld_syHrsGMdE,1283
-pixeltable/exprs/row_builder.py,sha256=fJEb8XU7j8M4PIc67O35a5Pzhoh_fs-i77lYxJPrA6o,18444
+pixeltable/exprs/row_builder.py,sha256=hpvmQyBie5pRtOkMKQu0xCxB18EK8iw_aUbtxYlzWcs,19862
 pixeltable/exprs/rowid_ref.py,sha256=iNBZ0wIhBAGkGCrNP9UQ2FeK8OajVX-eI4dtDA_bfHU,4401
 pixeltable/exprs/similarity_expr.py,sha256=GhiopGxN3wu5MOEqfWN4PNbqgSszg2VIC1_bqVPpm3Q,4333
 pixeltable/exprs/sql_element_cache.py,sha256=8i9ZslKWKTd1lUN7JvwRtQ-PFD1QMi5LBYdwcGA2-p0,1364
@@ -60,21 +65,21 @@ pixeltable/ext/functions/__init__.py,sha256=hIjPEKC5E5uJOXlQqUyhP9yn9ZqWOCJAlj0k
 pixeltable/ext/functions/whisperx.py,sha256=jojjNhazcYiAh1scwUl-erhIDRr4kOTkcLrjy0xcp6g,2325
 pixeltable/ext/functions/yolox.py,sha256=k-pQTelv4Tea3AXvDB7Kc7YCIa1uexjVGqxETP0B_hc,5351
 pixeltable/func/__init__.py,sha256=Tmy5srLdxcWDwxwnXXj2mTNUMimdYY-MdU6fJ9SpePc,457
-pixeltable/func/aggregate_function.py,sha256=SWqU3stHBzrZbzoNCNHBhMHqQbT9rpVTelWY6otboSg,12649
-pixeltable/func/callable_function.py,sha256=AK2sO-nOlDqDSDJvJ5zOp2UNwDJY-2keJlfPycGzRDw,6500
-pixeltable/func/expr_template_function.py,sha256=xoi7yngicX4F_-qa64a38_dvX3Bs_4BDphodTEul06o,5381
-pixeltable/func/function.py,sha256=1bGC7e93QrRZhSij-DdMVL94Lz-A7OWwbncmPhJj1ps,16799
+pixeltable/func/aggregate_function.py,sha256=eRPKCX2gdpEokyn5tKWxgbrC2UneaC0o2zIUp39Gd-w,12717
+pixeltable/func/callable_function.py,sha256=KL69czs2EZ9Fv5EoDT4xaXILePhaV9owTBNBgtNU4RU,8941
+pixeltable/func/expr_template_function.py,sha256=prKOIfO5rO_M3y6lSOB2zTYqZYb2nycbJ5m5FlJQ27E,5454
+pixeltable/func/function.py,sha256=Ida2SOKQzryiuOqQDF5l9Q-8LmmGx1rNvzxHwuOrtsg,18853
 pixeltable/func/function_registry.py,sha256=Ps09iPDRJRKj-NQtancrI07zqFBIPpASGZ636mpBiI0,12333
 pixeltable/func/globals.py,sha256=sEwn6lGgHMp6VQORb_P5qRd_-Q2_bUSqvqM9-XPN_ec,1483
-pixeltable/func/query_template_function.py,sha256=p3zp5aNIxlr0QR1ROVJLISDnZXpfOhqAeMIwr8l2b5g,5170
+pixeltable/func/query_template_function.py,sha256=zaRpthtFDsN0RGsZMebXiCSukW7oBHY4oskUx_93H8Y,5316
 pixeltable/func/signature.py,sha256=wWf07OawOwDmqZOjUPgHJHAhAQ-fd61PGBx82aWLBgA,11212
 pixeltable/func/tools.py,sha256=7OAHVb9KbmcPtfMWsbKAoRu3kx-_japSTE-hgRQJefM,5988
-pixeltable/func/udf.py,sha256=a8FqOcDqPvUICvELpPOQVsrvr0sgZBQuV4aVHXuexac,8603
-pixeltable/functions/__init__.py,sha256=2vdpJ3S1pUle8dBzCVM70o8Lj1c0lgccZPJgPYms_CQ,403
-pixeltable/functions/anthropic.py,sha256=FzMK4zhz4HvPqJmOM4CBIlpVa8JLiRShi8jEd2abHsg,4710
+pixeltable/func/udf.py,sha256=p7imgwk7S03q3lHZxqAZUKr63XSeSsrD8kWTbwOLQWo,8882
+pixeltable/functions/__init__.py,sha256=C7Okwst3JdhzwIFWXhyJ3F0EHKGKor2R8dRMvQ3y6Us,428
+pixeltable/functions/anthropic.py,sha256=rSgMWmR2YVRfrdbbAPY5OvKkbyV1DHBJIHFGdDwVnP4,8800
 pixeltable/functions/audio.py,sha256=7213nTnqKJ6vM9kalaoJ283OwX5SGEJN10vDhaRNZ6E,644
 pixeltable/functions/fireworks.py,sha256=qwFC_eIaDs-glxyJ_IVXaNGkpgPzeRsQ_SdpzueBxq0,2605
-pixeltable/functions/gemini.py,sha256=RQ3I25nXSXUXdF0rmhnv5XOgZXb_6SIgMc_hEyF83SI,2783
+pixeltable/functions/gemini.py,sha256=0aqsDpxI0AgeWu6zXQxtFbp3nhhuwXqzulZyot6UKdg,2613
 pixeltable/functions/globals.py,sha256=Q5S84Vk_ovCADLivXkn4vfUnX8oSgW8v1dZ0aPoz7C4,4844
 pixeltable/functions/huggingface.py,sha256=fXcj6NC5Hz_nYE-0CjvSJ1sD3Jq726hQXzueqnQ6Kr4,20576
 pixeltable/functions/image.py,sha256=3Qm4ybAT_o4YUl3bzhEXy8dKOwgZ7RCUV-ky-dbL_jc,13836
@@ -83,7 +88,7 @@ pixeltable/functions/llama_cpp.py,sha256=1awALuAXVpQH64l7vQlM8gvxLDix4c1-6DV3nG5
 pixeltable/functions/math.py,sha256=WPoH9zD9_GdwvBs-FSC3Sqb70gOPNouhPcBZABsuLwI,1541
 pixeltable/functions/mistralai.py,sha256=GpxtT-a8ltx1kQC8XTJRflxkh-17VhzzFTkxqIUsba8,5494
 pixeltable/functions/ollama.py,sha256=z-g0cCJ-WMf6RI3SXIVVbbOlrWNR8nxGTyzET1rratU,4320
-pixeltable/functions/openai.py,sha256=Jikl9oYPiKy61r5Ohe1xqjvb6SLCbNTYFr4yAHFkYME,17101
+pixeltable/functions/openai.py,sha256=Sh2zqwW72CYiMoyEELLpY1fwTmxuLGnAQbtEkmTac4c,24030
 pixeltable/functions/replicate.py,sha256=j8ZedScOMInmHWmriQSUOviw6tp8gQr-W6n21PNNL2g,2188
 pixeltable/functions/string.py,sha256=VqzhVildxTt_XblW89Kl5Zd6MVOU71eaX2LTMY5jkUg,20366
 pixeltable/functions/timestamp.py,sha256=KOm5eVF51PqOCTqwq8cKCAEJNWMgu2xN_yTtGf7yixU,9095
@@ -92,7 +97,7 @@ pixeltable/functions/util.py,sha256=GgKTzCjvzUQNjWtSObTkfxkvJ9GVUOzKimY45WhE25M,
 pixeltable/functions/video.py,sha256=12jnOdU0G-Hk42rJx-S5QC77-bDkUflkxfkjej8n5pM,6702
 pixeltable/functions/vision.py,sha256=MEgkNp-4tUUeQS05VEJBrMgQFM48aNsNlksfAl9rH7w,15494
 pixeltable/functions/whisper.py,sha256=f2wqRd0n9jSBqRZN3W93UaetiAHtbsK0j9jXR2j2kkQ,2913
-pixeltable/globals.py,sha256=W5-4KEVqg9tZwfL6EffkjQYtRNXDLKXj_Bv32liZeHI,33895
+pixeltable/globals.py,sha256=9UfD7FOn71dbzPZjV-LxzhxXBhwPcpLt46J0_RVPynk,33899
 pixeltable/index/__init__.py,sha256=XBwetNQQwnz0fiKwonOKhyy_U32l_cjt77kNvEIdjWs,102
 pixeltable/index/base.py,sha256=zo0YvJI3oXiK6hZJztB36ZftKKhLfO75Zq3t-PeQA6M,1556
 pixeltable/index/btree.py,sha256=JFerLyyLoBaD0FSF_jJ6iJFBVa-z_et--KdNR02xjRg,2264
@@ -130,12 +135,12 @@ pixeltable/metadata/converters/convert_25.py,sha256=3sCXMbHiLJWMvbghj-XaeW4_1xSE
 pixeltable/metadata/converters/util.py,sha256=NgzoKralRIXwOqtguifVsBbpEN2M2X1v8lvOnsdGnPs,6100
 pixeltable/metadata/notes.py,sha256=RsyIuwl3KhFRstB6cLX8ySWfIx67LAyr05HYXovu7Yc,779
 pixeltable/metadata/schema.py,sha256=H9t51cbhXOhNu9Xog2VTjZlkTCx4cjQAlwOXi3HSd2E,9448
-pixeltable/plan.py,sha256=dzG3iD9ClggFyIYmDuIK7H3eBC8HV3ft__f4ZULQvow,40687
+pixeltable/plan.py,sha256=DgzP5kDmBooaUioZ0KBlSUcx_jpqy456CnbDRVzFBJM,41501
 pixeltable/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-pixeltable/store.py,sha256=m-OeilqOFai5wUXvtXTV9txzn-GhK2lfA7fkvzrfWFg,21974
-pixeltable/type_system.py,sha256=IbUXeaqsguRAneiXZc2WrAXskUiN1IPJ3O9xHBGGQhk,48154
+pixeltable/store.py,sha256=-A7oRGHDpvhUZzx68QSde0t2NvUjeM1rOw9KdwukQ9s,22497
+pixeltable/type_system.py,sha256=nR004iZjvA5BIBSXsBXaHbpUPbddtiUCUzB0To1Bakw,48115
 pixeltable/utils/__init__.py,sha256=UYlrf6TIWJT0g-Hac0b34-dEk478B5Qx8dGco34YlIk,439
-pixeltable/utils/arrow.py,sha256=4AyCAkkNfNChVTaPAVS5j1ZO7BtTxn11duonolfktV8,3931
+pixeltable/utils/arrow.py,sha256=iuZhueel1FT_3UDAMITFXxodLB7w8TrUpsMpie3IDXg,3991
 pixeltable/utils/coco.py,sha256=WCUyoj0dTyJFbPB7frEQUvY92SlEPxQ068r-3QAwROE,7285
 pixeltable/utils/code.py,sha256=AOw1u2r8_DQXpX-lxJhyHWARGrCRDXOJHFVgKOi54Uc,1231
 pixeltable/utils/description_helper.py,sha256=P-8EE2pRFP8s3coe73frgV68yt4Dp3saErCUehMxKUw,3759
@@ -148,8 +153,8 @@ pixeltable/utils/pytorch.py,sha256=6RvOCjy_QV4gc-aht-3d0zoASkuv-warfpl87vgmuKw,3
 pixeltable/utils/s3.py,sha256=huA5hxDGkPIu18zWet76o0FsO7Vbtp-iPmnOzCe-MvA,586
 pixeltable/utils/sql.py,sha256=j_tj0h4ffm-DhUIJbvGphxrVyBKlNTwDKqWGhRQ5_PY,795
 pixeltable/utils/transactional_directory.py,sha256=UGzCrGtLR3hEEf8sYGuWBzLVFAEQml3vdIavigWeTBM,1349
-pixeltable-0.3.0.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-pixeltable-0.3.0.dist-info/METADATA,sha256=GiNmcGREs3AUhpt4tKPImEGYzzD4XHMR1StiJSEkWhI,19304
-pixeltable-0.3.0.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
-pixeltable-0.3.0.dist-info/entry_points.txt,sha256=ToOd-pRgG7AitEBgYoBCRRB4-KVDQ0pj_9T4a1LgwA4,97
-pixeltable-0.3.0.dist-info/RECORD,,
+pixeltable-0.3.1.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+pixeltable-0.3.1.dist-info/METADATA,sha256=tRC6nn3plpwtLqC5TZsNDfDRB9SCj2NyNJIvzK6kWMA,19370
+pixeltable-0.3.1.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+pixeltable-0.3.1.dist-info/entry_points.txt,sha256=ToOd-pRgG7AitEBgYoBCRRB4-KVDQ0pj_9T4a1LgwA4,97
+pixeltable-0.3.1.dist-info/RECORD,,

pixeltable/exec/expr_eval_node.py

@@ -1,232 +0,0 @@
-import logging
-import sys
-import time
-import warnings
-from dataclasses import dataclass
-from typing import Iterable, Optional
-
-from tqdm import TqdmWarning, tqdm
-
-from pixeltable import exprs
-from pixeltable.func import CallableFunction
-
-from .data_row_batch import DataRowBatch
-from .exec_node import ExecNode
-
-_logger = logging.getLogger('pixeltable')
-
-
-class ExprEvalNode(ExecNode):
-    """Materializes expressions
-    """
-    @dataclass
-    class Cohort:
-        """List of exprs that form an evaluation context and contain calls to at most one external function"""
-        exprs_: list[exprs.Expr]
-        batched_fn: Optional[CallableFunction]
-        segment_ctxs: list['exprs.RowBuilder.EvalCtx']
-        target_slot_idxs: list[int]
-        batch_size: int = 8
-
-    def __init__(
-            self, row_builder: exprs.RowBuilder, output_exprs: Iterable[exprs.Expr], input_exprs: Iterable[exprs.Expr],
-            input: ExecNode
-    ):
-        super().__init__(row_builder, output_exprs, input_exprs, input)
-        self.input_exprs = input_exprs
-        input_slot_idxs = {e.slot_idx for e in input_exprs}
-        # we're only materializing exprs that are not already in the input
-        self.target_exprs = [e for e in output_exprs if e.slot_idx not in input_slot_idxs]
-        self.pbar: Optional[tqdm] = None
-        self.cohorts: list[ExprEvalNode.Cohort] = []
-        self._create_cohorts()
-
-    def __next__(self) -> DataRowBatch:
-        input_batch = next(self.input)
-        # compute target exprs
-        for cohort in self.cohorts:
-            self._exec_cohort(cohort, input_batch)
-        _logger.debug(f'ExprEvalNode: returning {len(input_batch)} rows')
-        return input_batch
-
-    def _open(self) -> None:
-        warnings.simplefilter("ignore", category=TqdmWarning)
-        # This is a temporary hack. When B-tree indices on string columns were implemented (via computed columns
-        # that invoke the `BtreeIndex.str_filter` udf), it resulted in frivolous progress bars appearing on every
-        # insertion. This special-cases the `str_filter` call to suppress the corresponding progress bar.
-        # TODO(aaron-siegel) Remove this hack once we clean up progress bars more generally.
-        is_str_filter_node = all(
-            isinstance(expr, exprs.FunctionCall) and expr.fn.name == 'str_filter' for expr in self.output_exprs
-        )
-        if self.ctx.show_pbar and not is_str_filter_node:
-            self.pbar = tqdm(
-                total=len(self.target_exprs) * self.ctx.num_rows,
-                desc='Computing cells',
-                unit=' cells',
-                ncols=100,
-                file=sys.stdout
-            )
-
-    def _close(self) -> None:
-        if self.pbar is not None:
-            self.pbar.close()
-
-    def _get_batched_fn(self, expr: exprs.Expr) -> Optional[CallableFunction]:
-        if isinstance(expr, exprs.FunctionCall) and isinstance(expr.fn, CallableFunction) and expr.fn.is_batched:
-            return expr.fn
-        return None
-
-    def _is_batched_fn_call(self, expr: exprs.Expr) -> bool:
-        return self._get_batched_fn(expr) is not None
-
-    def _create_cohorts(self) -> None:
-        all_exprs = self.row_builder.get_dependencies(self.target_exprs)
-        # break up all_exprs into cohorts such that each cohort contains calls to at most one external function;
-        # seed the cohorts with only the ext fn calls
-        cohorts: list[list[exprs.Expr]] = []
-        current_batched_fn: Optional[CallableFunction] = None
-        for e in all_exprs:
-            if not self._is_batched_fn_call(e):
-                continue
-            assert isinstance(e, exprs.FunctionCall)
-            assert isinstance(e.fn, CallableFunction)
-            if current_batched_fn is None or current_batched_fn != e.fn:
-                # create a new cohort
-                cohorts.append([])
-                current_batched_fn = e.fn
-            cohorts[-1].append(e)
-
-        # expand the cohorts to include all exprs that are in the same evaluation context as the external calls;
-        # cohorts are evaluated in order, so we can exclude the target slots from preceding cohorts and input slots
-        exclude = set(e.slot_idx for e in self.input_exprs)
-        all_target_slot_idxs = set(e.slot_idx for e in self.target_exprs)
-        target_slot_idxs: list[list[int]] = []  # the ones materialized by each cohort
-        for i in range(len(cohorts)):
-            cohorts[i] = self.row_builder.get_dependencies(
-                cohorts[i], exclude=[self.row_builder.unique_exprs[slot_idx] for slot_idx in exclude])
-            target_slot_idxs.append(
-                [e.slot_idx for e in cohorts[i] if e.slot_idx in all_target_slot_idxs])
-            exclude.update(target_slot_idxs[-1])
-
-        all_cohort_slot_idxs = set(e.slot_idx for cohort in cohorts for e in cohort)
-        remaining_slot_idxs = set(all_target_slot_idxs) - all_cohort_slot_idxs
-        if len(remaining_slot_idxs) > 0:
-            cohorts.append(self.row_builder.get_dependencies(
-                [self.row_builder.unique_exprs[slot_idx] for slot_idx in remaining_slot_idxs],
-                exclude=[self.row_builder.unique_exprs[slot_idx] for slot_idx in exclude]))
-            target_slot_idxs.append(list(remaining_slot_idxs))
-        # we need to have captured all target slots at this point
-        assert all_target_slot_idxs == set().union(*target_slot_idxs)
-
-        for i in range(len(cohorts)):
-            cohort = cohorts[i]
-            # segment the cohort into sublists that contain either a single ext. function call or no ext. function calls
-            # (i.e., only computed cols)
-            assert len(cohort) > 0
-            # create the first segment here, so we can avoid checking for an empty list in the loop
-            segments = [[cohort[0]]]
-            is_batched_segment = self._is_batched_fn_call(cohort[0])
-            batched_fn: Optional[CallableFunction] = self._get_batched_fn(cohort[0])
-            for e in cohort[1:]:
-                if self._is_batched_fn_call(e):
-                    segments.append([e])
-                    is_batched_segment = True
-                    batched_fn = self._get_batched_fn(e)
-                else:
-                    if is_batched_segment:
-                        # start a new segment
-                        segments.append([])
-                        is_batched_segment = False
-                    segments[-1].append(e)
-
-            # we create the EvalCtxs manually because create_eval_ctx() would repeat the dependencies of each segment
-            segment_ctxs = [
-                exprs.RowBuilder.EvalCtx(
-                    slot_idxs=[e.slot_idx for e in s], exprs=s, target_slot_idxs=[], target_exprs=[])
-                for s in segments
-            ]
-            cohort_info = self.Cohort(cohort, batched_fn, segment_ctxs, target_slot_idxs[i])
-            self.cohorts.append(cohort_info)
-
-    def _exec_cohort(self, cohort: Cohort, rows: DataRowBatch) -> None:
-        """Compute the cohort for the entire input batch by dividing it up into sub-batches"""
-        batch_start_idx = 0  # start row of the current sub-batch
-        # for multi-resolution models, we re-assess the correct ext fn batch size for each input batch
-        ext_batch_size = cohort.batched_fn.get_batch_size() if cohort.batched_fn is not None else None
-        if ext_batch_size is not None:
-            cohort.batch_size = ext_batch_size
-
-        while batch_start_idx < len(rows):
-            num_batch_rows = min(cohort.batch_size, len(rows) - batch_start_idx)
-            for segment_ctx in cohort.segment_ctxs:
-                if not self._is_batched_fn_call(segment_ctx.exprs[0]):
-                    # compute batch row-wise
-                    for row_idx in range(batch_start_idx, batch_start_idx + num_batch_rows):
-                        self.row_builder.eval(
-                            rows[row_idx], segment_ctx, self.ctx.profile, ignore_errors=self.ctx.ignore_errors)
-                else:
-                    fn_call = segment_ctx.exprs[0]
-                    assert isinstance(fn_call, exprs.FunctionCall)
-                    # make a batched external function call
-                    arg_batches: list[list[exprs.Expr]] = [[] for _ in range(len(fn_call.args))]
-                    kwarg_batches: dict[str, list[exprs.Expr]] = {k: [] for k in fn_call.kwargs.keys()}
-
-                    valid_batch_idxs: list[int] = []  # rows with exceptions are not valid
-                    for row_idx in range(batch_start_idx, batch_start_idx + num_batch_rows):
-                        row = rows[row_idx]
-                        if row.has_exc(fn_call.slot_idx):
-                            # one of our inputs had an exception, skip this row
-                            continue
-                        valid_batch_idxs.append(row_idx)
-                        args, kwargs = fn_call._make_args(row)
-                        for i in range(len(args)):
-                            arg_batches[i].append(args[i])
-                        for k in kwargs.keys():
-                            kwarg_batches[k].append(kwargs[k])
-                    num_valid_batch_rows = len(valid_batch_idxs)
-
-                    if ext_batch_size is None:
-                        # we need to choose a batch size based on the args
-                        assert isinstance(fn_call.fn, CallableFunction)
-                        sample_args = [arg_batches[i][0] for i in range(len(arg_batches))]
-                        ext_batch_size = fn_call.fn.get_batch_size(*sample_args)
-
-                    num_remaining_batch_rows = num_valid_batch_rows
-                    while num_remaining_batch_rows > 0:
-                        # we make ext. fn calls in batches of ext_batch_size
-                        if ext_batch_size is None:
-                            pass
-                        num_ext_batch_rows = min(ext_batch_size, num_remaining_batch_rows)
-                        ext_batch_offset = num_valid_batch_rows - num_remaining_batch_rows  # offset into args, not rows
-                        call_args = [
-                            arg_batches[i][ext_batch_offset:ext_batch_offset + num_ext_batch_rows]
-                            for i in range(len(arg_batches))
-                        ]
-                        call_kwargs = {
-                            k: kwarg_batches[k][ext_batch_offset:ext_batch_offset + num_ext_batch_rows]
-                            for k in kwarg_batches.keys()
-                        }
-                        start_ts = time.perf_counter()
-                        assert isinstance(fn_call.fn, CallableFunction)
-                        result_batch = fn_call.fn.exec_batch(call_args, call_kwargs)
-                        self.ctx.profile.eval_time[fn_call.slot_idx] += time.perf_counter() - start_ts
-                        self.ctx.profile.eval_count[fn_call.slot_idx] += num_ext_batch_rows
-
-                        # move the result into the row batch
-                        for result_idx in range(len(result_batch)):
-                            row_idx = valid_batch_idxs[ext_batch_offset + result_idx]
-                            row = rows[row_idx]
-                            row[fn_call.slot_idx] = result_batch[result_idx]
-
-                        num_remaining_batch_rows -= num_ext_batch_rows
-
-                    # switch to the ext fn batch size
-                    cohort.batch_size = ext_batch_size
-
-            # make sure images for stored cols have been saved to files before moving on to the next batch
-            rows.flush_imgs(
-                slice(batch_start_idx, batch_start_idx + num_batch_rows), self.stored_img_cols, self.flushed_img_slots)
-            if self.pbar is not None:
-                self.pbar.update(num_batch_rows * len(cohort.target_slot_idxs))
-            batch_start_idx += num_batch_rows
-