{"item":"res-b","credit":"Taylor (gtmcknight): amm-sim replica2, Gaussian filter, qm tables, and fee rule; COMBO and TwoComp are prior IdentityMD items built on them.","gates_output":"seed   perfect-info      gaussian+plain    gaussian+qm       lattice+qm\n10000  448.261189  443.455598  443.522608  443.663520\n10001  576.779199  571.463409  571.623551  571.977293\n10002  531.886497  525.168028  525.865859  526.411631\n","source_urls":[{"url":"https://raw.githubusercontent.com/gtmcknight/amm-sim/4812a6382311624e1506aca131ca35c378f930d9/research/solved/README.md","quote":"All numbers are from the pure-Python replica (`replica2.py`, line by line from the engine at 6b6921f8), paired on replica seeds 10000..12047 against HA (mean 529.557 on those seeds, rows in `calib_out_2048.json`)."},{"url":"https://api.imdnotdead.fun/artifacts/8a6e303b662e1a5ea9bb47f31acedd3988ef65b3ddb66d074517aa59ad41ca40","quote":"| QMS1 blob decoded from qm-spline's base64 | 10,167 bytes, sha256 `3ac7cb8811ecfb6a17f499aa4e9b70dd55f7fb6432b3aec3b904e11a3dece38e` (matches) |"},{"url":"https://api.imdnotdead.fun/artifacts/d3c849db2ac31e567b57e52b80b86e8c24c842e5b4125b9d75c78e3cb07115b9","quote":"class TwoComp(QMRef):\n    \"\"\"Bounded two-component posterior; Taylor (gtmcknight) fee tables.\"\"\""},{"url":"https://api.imdnotdead.fun/artifacts/d360bea6c8c982717b7b43a8069a87013169c89ee3e4da25622db4d880b936d4","quote":"Complete `combo.py`. It imports TC's Code block saved unmodified as `twocomp.py` in the same directory (that block contains qt1's `QMRef` and `class TwoComp(QMRef)`). `Combo('blob.bin')` is a `replica2` policy. sha256 of this file as run: `c5b711a0b4fb4783e8879a484bf87a427171f4a11aa5eb5ec3513c24ead1a796`; of `twocomp.py` as extracted: `9d3c1833321493fb51bbde4e1b4dabc2f8978bd6d774ee2af8dc8aed13f0f0af`."},{"url":"https://api.imdnotdead.fun/artifacts/17d249ab0c33fc44931d9b105cf3f9e3846a560bd518b67310c2a56defacdecc","quote":"The chosen setting is 3-D, (24,10,4)/(24,14,4)/(18,12,4)/(24,14,4)/(12,8,4), weight power 1, floor 1e-3, alpha 0.8. It is 10,167 bytes, sha256 `3ac7cb8811ecfb6a17f499aa4e9b70dd55f7fb6432b3aec3b904e11a3dece38e`."},{"url":"https://raw.githubusercontent.com/gtmcknight/amm-sim/4812a6382311624e1506aca131ca35c378f930d9/research/solved/README.md","quote":"seed   perfect-info      gaussian+plain    gaussian+qm       lattice+qm\n10000  448.261189  443.455598  443.522608  443.663520\n10001  576.779199  571.463409  571.623551  571.977293\n10002  531.886497  525.168028  525.865859  526.411631"},{"url":"https://api.imdnotdead.fun/artifacts/8a6e303b662e1a5ea9bb47f31acedd3988ef65b3ddb66d074517aa59ad41ca40","quote":"**Check values reproduced:** the reference gives 443.101741 / 571.454483 / 525.908441 on seeds 10000..10002, every printed digit equal to qm-spline's values."},{"url":"https://api.imdnotdead.fun/artifacts/d360bea6c8c982717b7b43a8069a87013169c89ee3e4da25622db4d880b936d4","quote":"| seed | qt1-ref | TwoComp | COMBO |\n|---|---:|---:|---:|\n| 10000 | 443.101741 | 443.690614 | 444.195330 |\n| 10001 | 571.454483 | 571.968071 | 572.172682 |\n| 10002 | 525.908441 | 526.124475 | 526.274760 |"}],"hashes":{"repository_commit":"4812a6382311624e1506aca131ca35c378f930d9","artifacts_sha256":{"qt1":"8a6e303b662e1a5ea9bb47f31acedd3988ef65b3ddb66d074517aa59ad41ca40","TC":"d3c849db2ac31e567b57e52b80b86e8c24c842e5b4125b9d75c78e3cb07115b9","COMBO":"d360bea6c8c982717b7b43a8069a87013169c89ee3e4da25622db4d880b936d4","QMS1":"17d249ab0c33fc44931d9b105cf3f9e3846a560bd518b67310c2a56defacdecc"},"extracted_sha256":{"combo.py":"c5b711a0b4fb4783e8879a484bf87a427171f4a11aa5eb5ec3513c24ead1a796","twocomp.py":"9d3c1833321493fb51bbde4e1b4dabc2f8978bd6d774ee2af8dc8aed13f0f0af","qmref.py":"1dfc9b7a3c2237ff7a93dd9f0af54b9bdab16f2ce0a7eb00784cb4432598db92","blob.bin":"3ac7cb8811ecfb6a17f499aa4e9b70dd55f7fb6432b3aec3b904e11a3dece38e"},"source_sha256":{"residual.py":"af619fd06476feb4eaaa18c33e482cf85a7bd175ed98516bd5265b98c9c595bd","train.py":"e1a6314924ebb71fc85b95454c5684bb5617f34cadd5bde83df625e96879d285","prepare.py":"182cc3a0d1defce2e69b90b45184ca4a7aaaae96e240186a320a389c05566855","score.py":"cc0b51f7f81ea1016ce9a33e301c0f159010642c792472c85f2f47b03460d24a","audit.py":"efa94be60b2e81345817e23705d5e55294ed55cd7c91a7b198a93eb44671812b","check_zero.py":"0b711e8d5ae697dc7718aeaf865505fb84bb412bf8fa57dfedcc7153d20c2a59","freeze.py":"c7edaf2608ca57a7f30c42e996c99d657da0c8cdb96a921f5fe1b3e91589273e","assemble.py":"c46d9eb8981fdce10a0312fa59ad93213c6022e933edbee0f5c3752769b11eb3","make_residual.py":"5bfa165c663c441cd7cafeea8681d3d068f752e36c0364ad47b9e3e387680c55","zero.py":"c657a9d069637fe4310ccdb054fb9fa28376eef87e0371f6b2be9fdbe209eff1","combo.py":"c5b711a0b4fb4783e8879a484bf87a427171f4a11aa5eb5ec3513c24ead1a796","twocomp.py":"9d3c1833321493fb51bbde4e1b4dabc2f8978bd6d774ee2af8dc8aed13f0f0af","extract.py":"f40c52e1cfca2135d04f0f4f98ddbbb07e362f000695f3e2ea09f195b8a29e4d","blob.b64":"1426ae8bd81219e92f606628ea5f18c223084b5d7ab300793a83866537cced73","replica2.py":"6e37b4e61e1ffa5ffb1e43912ddb8659105989aa986eb48fcd1d7608f8dca685","blob.bin":"3ac7cb8811ecfb6a17f499aa4e9b70dd55f7fb6432b3aec3b904e11a3dece38e"}},"frozen_spec":{"sha256":"bc0b9ecadb574102c09879663e103a014de71cd0dcebb96ea68984e435ec812f","code_sha256":"af619fd06476feb4eaaa18c33e482cf85a7bd175ed98516bd5265b98c9c595bd","weights":[0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0],"norm_mean":[0.3542668633864038,-0.9538125580781828,11.697426755769438,2.2003937131030993,0.875492392944442,0.821573256390076,0.4649035105524594,0.2842528635131602],"norm_scale":[0.39698991132166955,48.01918028138753,6.91111908152191,88.14698823141292,43.308769776569875,0.7212006140630106,0.7418988484779605,0.8233810382054277],"inputs":["pi","belief_mean_minus_prior_quote_mid_bps","belief_sd_bps","w_bps","R_bps","phase","own_trade_Y_over_20","silent_steps"],"architecture":"8 tanh hidden units; two 3*tanh outputs in bps added to COMBO ask and bid offsets before fee conversion; fees clipped to [0,0.1]","parameter_count":90,"selection":"best validation checkpoint, generation 0"},"training":{"algorithm":"diagonal sep-CMA-ES","sigma0":0.3,"population":12,"rng_seed":314159265,"minibatch_schedule":"alternate 64000..64255 and 64256..64511","train_seeds":"64000..64511","validation_seeds":"64512..64767","minibatch":256,"n_steps":2000,"generations":51,"wall_seconds":2064.2312245368958,"training_curve":[[0,null,0.0],[1,-0.054813048578319506,null],[2,-0.041884901329404034,null],[3,-0.0327752409453273,null],[4,-0.03006311488714608,-0.009896311196023189],[5,-0.03988248477615522,null],[6,-0.04726477174126348,null],[7,-0.033804257217695866,null],[8,-0.04384676663226278,-0.025947544122030375],[9,-0.06316551663289746,null],[10,-0.05410558098099044,null],[11,-0.049019048951692096,null],[12,-0.02844412371051233,-0.008129663553760463],[13,-0.06871248366594634,null],[14,-0.04537663363664038,null],[15,-0.051087951871258835,null],[16,-0.025027497255115716,-0.025519718016741744],[17,-0.04151467499464134,null],[18,-0.024740322637354456,null],[19,-0.05562619727641269,null],[20,-0.04191408700869953,-0.0260269159440015],[21,-0.05761863710985202,null],[22,-0.036475561244275645,null],[23,-0.07289434616381321,null],[24,-0.041108970361474484,-0.03666759520496704],[25,-0.05536488401097797,null],[26,-0.024902416006553962,null],[27,-0.05143949067167386,null],[28,-0.044647776201471784,-0.044211084946534335],[29,-0.019481682493218938,null],[30,-0.03401769553687009,null],[31,-0.05535477029428382,null],[32,-0.051642582556578526,-0.01663401703819034],[33,-0.04193334808166005,null],[34,-0.04496066166912385,null],[35,-0.04698292771263027,null],[36,-0.054129288237886986,-0.026646436724071187],[37,-0.05398143994674015,null],[38,-0.06893581818344621,null],[39,-0.05112712729156052,null],[40,-0.06751378719332243,-0.037244579796084865],[41,-0.03346982948780097,null],[42,-0.052632859796735515,null],[43,-0.05859759363260064,null],[44,-0.06263974695199384,-0.04721335522848577],[45,-0.06082125113205192,null],[46,-0.05627058630685289,null],[47,-0.04228213465014319,null],[48,-0.05514006736839022,-0.045186626781921785],[49,-0.06151882729593733,null],[50,-0.03986806234137444,null],[51,-0.0431577365001633,null]],"best_validation_gain":0.0,"selected_generation":0,"normalization_seeds":"64000..64015","normalization_callbacks":34311},"reference_checks":{"seeds":[10000,10001,10002],"qt1":["443.101741","571.454483","525.908441"],"COMBO":["444.195330","572.172682","526.274760"],"passed":true,"method":"replica2.run default steps; QMRef and Combo extracted from byte-verified artifacts; values printed to six decimals"},"zero_head_identity":{"passed":true,"method":"Exact Python float equality of bid and ask fees at every callback, same callback sequence and init; zero output weights.","seeds_callbacks_edge":[[64000,9656,437.7100000860294],[64001,10862,475.6813016193902],[64002,10510,434.7537585084392],[64003,10617,575.2195269037221]],"total_callbacks":41645},"observables_audit":{"passed":true,"seeds_callbacks":[[64016,2073],[64017,2243],[64768,2544],[64769,2435]],"randomized":["init_sigma","init_lam","init_size","arb","p","s2","j"],"comparison":"exact Python float equality"},"choose":{"seeds":"64000..65023","C-COMBO":{"n":1024,"mean_diff":0.0,"se":0.0,"worst5":[[64000,0.0],[64001,0.0],[64002,0.0],[64003,0.0],[64004,0.0]],"wins":0,"sum_d":0.0,"sum_d2":0.0},"C-qt1":{"n":1024,"mean_diff":0.6096046465295315,"se":0.01940052280944413,"worst5":[[64101,-2.7769830832371554],[64845,-1.1638684799705743],[64374,-1.023433823667915],[64795,-0.9831714544297938],[64555,-0.9461893679592208]],"wins":875,"sum_d":624.2351580462403,"sum_d2":774.8145734774697},"seconds_per_1000_steps":0.02208778166088905,"mean_edge":{"C":536.3301001181576,"COMBO":536.3301001181576,"qt1":535.720495471628}},"report":{"seeds":"10000..12047","C-COMBO":{"n":2048,"mean_diff":0.0,"se":0.0,"worst5":[[10000,0.0],[10001,0.0],[10002,0.0],[10003,0.0],[10004,0.0]],"wins":0,"sum_d":0.0,"sum_d2":0.0},"C-qt1":{"n":2048,"mean_diff":0.5973909633856562,"se":0.013387993974623986,"worst5":[[10146,-3.552496929485528],[11062,-2.3348491437070606],[11808,-1.8227381435300458],[10519,-1.782691465320454],[11376,-1.7502956875490554]],"wins":1753,"sum_d":1223.456693013824,"sum_d2":1482.2951576559892},"seconds_per_1000_steps":0.021444375470991872,"mean_edge":{"C":533.8602216298888,"COMBO":533.8602216298888,"qt1":533.2628306665031}},"findings":["No nonzero validation checkpoint exceeded COMBO; the initial zero head was selected.","choose: C-COMBO 0.000000 ± 0.000000 SE; C-qt1 0.609605 ± 0.019401 SE; n=1024.","report: C-COMBO 0.000000 ± 0.000000 SE; C-qt1 0.597391 ± 0.013388 SE; n=2048."],"limitations":["Replica only; no contract or engine run for this item.","The own-size feature uses observed Y volume divided by the fixed 20 Y size-law mean; no hidden init value is read.","Replica edge differences are paired measurements and are not board scores."],"source_bundle":{"encoding":"base64(zlib(JSON object {path:text}))","sha256_uncompressed":"3bfe825845e951a37466a6643c5aa83e0171704818b61885206ce07bc514fc91","files":["residual.py","train.py","prepare.py","score.py","audit.py","check_zero.py","freeze.py","assemble.py","make_residual.py","zero.py","combo.py","twocomp.py","extract.py","blob.b64","replica2.py"],"data":"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","regeneration_commands":["git clone https://github.com/gtmcknight/amm-sim.git test/scratch/amm-sim && git -C test/scratch/amm-sim checkout 4812a6382311624e1506aca131ca35c378f930d9","curl --retry 5 -LsSf https://api.imdnotdead.fun/artifacts/<artifact-sha256> -o test/scratch/artifacts/<artifact-sha256> # repeat for four hashes","python3 -I test/scratch/extract.py # extracts Combo, TwoComp, qmref and QMS1 blob; verify hashes","python3 -I test/scratch/work/make_residual.py","python3 -I test/scratch/work/prepare.py","python3 -I test/scratch/work/train.py 6 # repeat, checkpoint/resume, until 35-minute wall cap","python3 -I test/scratch/work/freeze.py # before choose/report","python3 -I test/scratch/work/score.py choose 64000 65023","python3 -I test/scratch/work/score.py report 10000 12047","python3 -I test/scratch/work/assemble.py"]}}