{"metadata":{"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.11.13"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":96164,"databundleVersionId":12993472,"sourceType":"competition"},{"sourceId":12462463,"sourceType":"datasetVersion","datasetId":7861436},{"sourceId":12503273,"sourceType":"datasetVersion","datasetId":7870836},{"sourceId":12512096,"sourceType":"datasetVersion","datasetId":7897419},{"sourceId":12522757,"sourceType":"datasetVersion","datasetId":7904568},{"sourceId":12557942,"sourceType":"datasetVersion","datasetId":7909478}],"dockerImageVersionId":31089,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false},"papermill":{"default_parameters":{},"duration":374.271766,"end_time":"2025-07-16T09:01:49.809559","environment_variables":{},"exception":null,"input_path":"__notebook__.ipynb","output_path":"__notebook__.ipynb","parameters":{},"start_time":"2025-07-16T08:55:35.537793","version":"2.6.0"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"#### Hello everyone! The competition is ending..\n\nWe want to try another technology related to FE.\nWe probably won't take part in competitions using dynamic blend, and most likely, we won't. And even if we did, we'd look at the following submissions: [**0.94857, 0.95002, 0.95109, 0.95129, 0.95163, 0.95164, 0.95165**]((https://) and in order to remove the \"extra clutter\" we'd use five of these seven - we'd choose those submissions that differ from each other, that is, we'd remove the two that are most similar to the participants in the main group. ([**You can easily and simply look at this in the dataset section**](https://www.kaggle.com/datasets/nina2025/21-juli-2025-drw))\nNow bye everyone, and good luck, we'll meet in other competitions.\n\nps\n\nBy and large, one horizontal blend is enough here. Yes, the vertical blend has its place, and what a place. It emphasized itself, leading to the next level of refinements. But how to determine these levels of vertical depth - where can different parameters be given depending on the level? Here we need a principled approach from the ML space. Powerful and precise ML methods are able to \"detect\" and determine these levels - at which of them certain parameters or assemblies will behave better.\n\npsps\n\nConclusions based on participation in this competition using a dynamic blend:\n\nIt has proven itself, but it is not ML. It is intended to show at some level of model building - that perhaps it is necessary to move in a slightly different direction - either by refining the FE or refining the architecture, or perhaps by rebuilding it.\n\nThis type of blend, and most likely all the others known at the moment, this includes ensemble systems (even those described in scientific libraries and well-functioning ones), operates only in the space of public solutions with an initially prescribed architecture and a system response that shows the result of their friendly \"crossbreeding\" after evolution. Yes, all these things are capable of clarifying various solutions. But they will not be able to \"get through\" further than they are in the architectures of solutions.\n\nThinking out loud, I would like to draw attention to two fundamentally different things - for example, one of the masters on one of the discussion channels, I will not pronounce his name, I will only say that he is Norwegian - stated directly and openly his opinion about blends - \"he says that they do not carry learning ability, and it would be nice not to consider them at all in case of filing an application.\" This opinion was supported by the grandmasters. Ask them a question - why, on what basis. 95% of unclear answers will follow! And the main line there will be - \"because we do not like it.\" Following the protection of their names and so on and so forth - they can ask a question - \"don't you think that this can be done using **oof** and, for example, logistic regression\" and for the sake of a catchphrase they will add - \"we are showing how to do it.\" To this question you can give a bold answer: You are showing this based on your architecture and your conclusions. And there are other solutions and architectures. Which achieve almost the same ratings. And in order to demonstrate your skills regarding a simple and unsophisticated ensemble based on **OOF** and logistic regression - you simply do not have enough time. The legal answer is very correct and very legal. And those who do not particularly agree with it can be asked a question to which they will not give an answer at all - look - you act with all known - you have an experienced training sample, there are ML methods. And if it is not there or its relevance due to the time threshold is reduced to zero. What will you do? Yes, one person has answered this question so far - this is Grigory Perelman! And we are just following his example trying to solve this or that problem. For example, without one unknown - without an experienced training sample. That's all this conversation will be closed! This does not apply to honest grandmasters such as, for example, the American Chris Deotte and there are other people like him here - this applies more to lovers of free pearls and voices and the attention of their colleagues. And they know this very well.\n\nTherefore, my wishes to all those who share my point of view - architecture and FE will always be above empty words. And at the initial stage - either training or designing blends - or as these wizards like to say - \"blind blend\", is a necessary tool on the way, especially in the continuation of this path!\n\n \n\nOur plan for the last week of competition:\n\n- 5-6 days before the end of the competition - an attempt to artificially increase/decrease the signal\n- 3-4 days before the end of the competition - an attempt to slightly change/add to the FE in the public solutions used\n- 1-2 days before the end of the competition - preparation and sending for evolution of the system not blends but active solution\n\n \n\n#### Public solutions:\n\n \n\n- 0.90038  v.03 - [XGB + Deep Learning Ensemble](https://www.kaggle.com/code/taylorsamarel/xgb-deep-learning-ensemble) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.89178  v.13 - [REBOOT: SGD CONVERGENCE ++](https://www.kaggle.com/code/taylorsamarel/reboot-sgd-convergence/notebook) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.88377  v.09 - [REBOOT: Use the lead values - IMPROVED](https://www.kaggle.com/code/taylorsamarel/reboot-use-the-lead-values-improved) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.86767  v.01 - [DRW-Ensembling-0.86767](https://www.kaggle.com/code/vishalpainjane/drw-ensemble-0-72837) - [Vishal Painjane](https://www.kaggle.com/vishalpainjane)\n- 0.83975  v.07 - [REBOOT: Use the lead values](https://www.kaggle.com/code/gromml/reboot-use-the-lead-values/notebook) - [gromml](https://www.kaggle.com/gromml)\n\n\n17-july-2025 - LB = [0.94915](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n\n \n\n- 0.90038  v.03 - [XGB + Deep Learning Ensemble](https://www.kaggle.com/code/taylorsamarel/xgb-deep-learning-ensemble) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.89178  v.13 - [REBOOT: SGD CONVERGENCE ++](https://www.kaggle.com/code/taylorsamarel/reboot-sgd-convergence/notebook) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.88377  v.09 - [REBOOT: Use the lead values - IMPROVED](https://www.kaggle.com/code/taylorsamarel/reboot-use-the-lead-values-improved) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.86767  v.01 - [DRW-Ensembling-0.86767](https://www.kaggle.com/code/vishalpainjane/drw-ensemble-0-72837) - [Vishal Painjane](https://www.kaggle.com/vishalpainjane)\n\n\n18-july-2025 - LB = [0.94608](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n\n \n\n- 0.89178  v.13 - [REBOOT: SGD CONVERGENCE ++](https://www.kaggle.com/code/taylorsamarel/reboot-sgd-convergence/notebook) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.88377  v.09 - [REBOOT: Use the lead values - IMPROVED](https://www.kaggle.com/code/taylorsamarel/reboot-use-the-lead-values-improved) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.86767  v.01 - [DRW-Ensembling-0.86767](https://www.kaggle.com/code/vishalpainjane/drw-ensemble-0-72837) - [Vishal Painjane](https://www.kaggle.com/vishalpainjane)\n- 0.83975  v.07 - [REBOOT: Use the lead values](https://www.kaggle.com/code/gromml/reboot-use-the-lead-values/notebook) - [gromml](https://www.kaggle.com/gromml)\n\n\n18-july-2025 - LB = [0.94807](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n\n \n\n- 0.95002  v.16 - [DRW | blend.Horizontal + blend.Vertical REMIX](https://www.kaggle.com/code/taylorsamarel/drw-blend-horizontal-blend-vertical-remix) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.89178  v.13 - [REBOOT: SGD CONVERGENCE ++](https://www.kaggle.com/code/taylorsamarel/reboot-sgd-convergence/notebook) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.86767  v.01 - [DRW-Ensembling-0.86767](https://www.kaggle.com/code/vishalpainjane/drw-ensemble-0-72837) - [Vishal Painjane](https://www.kaggle.com/vishalpainjane)\n- 0.83975  v.07 - [REBOOT: Use the lead values](https://www.kaggle.com/code/gromml/reboot-use-the-lead-values/notebook) - [gromml](https://www.kaggle.com/gromml)\n\n\n18-july-2025 - LB = [0.94812](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n\n \n\n- 0.95004  v.03 - [DRW | blend.H&V REMIX | higher changepoint](https://www.kaggle.com/code/ducknew/drw-blend-h-v-remix-higher-changepoint) - [ducknew](https://www.kaggle.com/ducknew)\n- 0.95002  v.16 - [DRW | blend.Horizontal + blend.Vertical REMIX](https://www.kaggle.com/code/taylorsamarel/drw-blend-horizontal-blend-vertical-remix) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.95001  v.33 - [DRW Crypto Market Submission](https://www.kaggle.com/code/rosswade/drw-crypto-market-submission) - [Ross Wade](https://www.kaggle.com/rosswade)\n\n\n19-july-2025 - LB = [0.95004](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n \n\n- 0.95129  v.03 - [DRW | blend.H&V REMIX | higher changepoint](https://www.kaggle.com/code/ducknew/drw-blend-h-v-remix-higher-changepoint) - [ducknew](https://www.kaggle.com/ducknew)\n- 0.95109  v.02 - [Ensemble Public LB 0.95109](https://www.kaggle.com/code/johndoe2011/ensemble-public-lb-0-95109) - [ZULQAR.](https://www.kaggle.com/johndoe2011)\n- nodata1  v.17 - [DRW | blend.Horizontal + blend.Vertical](https://www.kaggle.com/code/nina2025/drw-blend-horizontal-blend-vertical) - [Nina](https://www.kaggle.com/nina2025)\n- nodata3  v.17 - [DRW | blend.Horizontal + blend.Vertical](https://www.kaggle.com/code/nina2025/drw-blend-horizontal-blend-vertical) - [Nina](https://www.kaggle.com/nina2025)\n- nodata7  v.17 - [DRW | blend.Horizontal + blend.Vertical](https://www.kaggle.com/code/nina2025/drw-blend-horizontal-blend-vertical) - [Nina](https://www.kaggle.com/nina2025)\n\n\n20-july-2025 - LB = [0.95103](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n\n \n\n100% copy of the assembly - [🔥Top 2% |🏆0.95129 | Ensemble](https://www.kaggle.com/code/ducknew/top-2-0-95129-ensemble) by [ducknew](https://www.kaggle.com/ducknew)\n\n- 0.95109  v.02 - [Ensemble Public LB 0.95109](https://www.kaggle.com/code/johndoe2011/ensemble-public-lb-0-95109) - [ZULQAR.](https://www.kaggle.com/johndoe2011)\n- 0.95004  v.03 - [DRW | blend.H&V REMIX | higher changepoint](https://www.kaggle.com/code/ducknew/drw-blend-h-v-remix-higher-changepoint) - [ducknew](https://www.kaggle.com/ducknew)\n- 0.95002  v.16 - [DRW | blend.Horizontal + blend.Vertical REMIX](https://www.kaggle.com/code/taylorsamarel/drw-blend-horizontal-blend-vertical-remix) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.94857  v.11 - [DRW - Crypto | Ensemble](https://www.kaggle.com/code/guanyuzhen/drw-crypto-ensemble) - [Alex GUAN](https://www.kaggle.com/guanyuzhen)\n\n21-july-2025 - LB = [0.95190](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n\n \n\n80% copy of the prev. assembly + new elem [ 0.89954 ](https://www.kaggle.com/code/nina2025/xgb-deep-learning-ensemble)\n\n- 0.95109  v.02 - [Ensemble Public LB 0.95109](https://www.kaggle.com/code/johndoe2011/ensemble-public-lb-0-95109) - [ZULQAR.](https://www.kaggle.com/johndoe2011)\n- 0.95004  v.03 - [DRW | blend.H&V REMIX | higher changepoint](https://www.kaggle.com/code/ducknew/drw-blend-h-v-remix-higher-changepoint) - [ducknew](https://www.kaggle.com/ducknew)\n- 0.95002  v.16 - [DRW | blend.Horizontal + blend.Vertical REMIX](https://www.kaggle.com/code/taylorsamarel/drw-blend-horizontal-blend-vertical-remix) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.94857  v.11 - [DRW - Crypto | Ensemble](https://www.kaggle.com/code/guanyuzhen/drw-crypto-ensemble) - [Alex GUAN](https://www.kaggle.com/guanyuzhen)\n- 0.89954  v.01 - [XGB + Deep Learning Ensemble](https://www.kaggle.com/code/nina2025/xgb-deep-learning-ensemble) - [99.8% copy (0.90038)](https://www.kaggle.com/code/taylorsamarel/xgb-deep-learning-ensemble)\n\n22-july-2025 - LB = [0.95155](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n \n\n80% copy of the prev. assembly + new elem [ 0.90222 ](https://www.kaggle.com/code/nina2025/reboot-sgd-convergence-hs-feats)\n\n- 0.95109  v.02 - [Ensemble Public LB 0.95109](https://www.kaggle.com/code/johndoe2011/ensemble-public-lb-0-95109) - [ZULQAR.](https://www.kaggle.com/johndoe2011)\n- 0.95004  v.03 - [DRW | blend.H&V REMIX | higher changepoint](https://www.kaggle.com/code/ducknew/drw-blend-h-v-remix-higher-changepoint) - [ducknew](https://www.kaggle.com/ducknew)\n- 0.95002  v.16 - [DRW | blend.Horizontal + blend.Vertical REMIX](https://www.kaggle.com/code/taylorsamarel/drw-blend-horizontal-blend-vertical-remix) - [Taylor S. Amarel](https://www.kaggle.com/taylorsamare)\n- 0.94857  v.11 - [DRW - Crypto | Ensemble](https://www.kaggle.com/code/guanyuzhen/drw-crypto-ensemble) - [Alex GUAN](https://www.kaggle.com/guanyuzhen)\n- 0.90222  v.01 - [REBOOT: SGD CONVERGENCE ++ (NOT BLEND SOLUTION)](https://www.kaggle.com/code/nina2025/xgb-deep-learning-ensemble) - [99.7% copy (0.90278)](https://www.kaggle.com/code/taylorsamarel/reboot-sgd-convergence-not-blend-solution)\n\n23-july-2025 - LB = [0.95164](https://www.kaggle.com/code/nina2025/drw-crypto-market-prediction-iblend)\n\n ","metadata":{"_kg_hide-input":false,"_kg_hide-output":false,"papermill":{"duration":0.002055,"end_time":"2025-07-16T08:55:40.398958","exception":false,"start_time":"2025-07-16T08:55:40.396903","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"files of the 'nodata_1-7' series were obtained in the following way: 5 files were taken for the experiment (all different, from our numerous lists), for example A, B, C, D, E. Then 'C' 'approaches' the group 'A, B' - and they ask 'C' a question - our average opinion-value is 'such-and-such' - and what is yours? And whatever it is - the difference between them is taken and raised, for example, to a square. Then 'D' 'approaches' the group 'A, B, C' ('C' is already a member of the group - he was accepted) and the same thing happens - the difference between the average of the group ('A, B, C') and the new representative ('D') is raised to a square. All the same happens with element E. The essence of this is simple and is based on existing evolutions - if one submit showed an advantage over another submit - then why not try to increase/decrease this value, or rather their difference in a slightly superlative degree. This cisk-architecture - when one of the submitters approaches the group was taken from small studies of this area - they exist but in very small quantities, but even they in some competitions of the Playground series put the top solutions of their blends and their ensembles into great question - that is, neither one nor the other could cope with them. But this was not always the case. And the cases of victories of this approach are less than blends and ensembles. Let's say 1 to 5. But if they showed themselves, they confidently went up. It would be possible to move on to the risk-architecture of this approach - but due to the small number of studies in this area, the question of this is not being raised yet. We are happy when this works out. Therefore, to be honest - we ourselves consider this approach to be a coin tossed up. Yes, we know that it will land on the heavier side. But we cannot give an exact assessment of this approach yet.To business, let's see what happens here?!\n\n------------------------------------\n\nOn the penultimate day of the competition, a light FE was performed on the basis of existing and proven notebooks (Taylor S Amarel) - groups of simple distances in hyperspace were taken. In order to try to shift the \"main axis\" (the entire architecture remains untouched, only a little FE). And also this new element was connected to our blend, which also showed itself a little from the best side here. Well, let's see what comes of it. p.s. tomorrow on the last day we have 5 more launches - 2-3 of which we will leave as in the plan indicated above - we will take the top solution and impose or rather add these \"hyperspace heroes\" to it. To business..\n\n------------------------------------\n\nWe made two mistakes - one \"good\" and the other \"not so good\"\n\n\"Good\": not one day left - but two\n\n\"Not so good\" - we \"forgot to return\" from the corresponding function the names of the new feats, on which the supposed \"bet\" was made!","metadata":{}},{"cell_type":"code","source":"import pandas as pd","metadata":{"papermill":{"duration":1.879293,"end_time":"2025-07-16T08:55:42.280345","exception":false,"start_time":"2025-07-16T08:55:40.401052","status":"completed"},"tags":[],"trusted":true,"execution":{"iopub.status.busy":"2025-07-22T17:11:07.189535Z","iopub.execute_input":"2025-07-22T17:11:07.189818Z","iopub.status.idle":"2025-07-22T17:11:07.194026Z","shell.execute_reply.started":"2025-07-22T17:11:07.189793Z","shell.execute_reply":"2025-07-22T17:11:07.193120Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def iBlend(path_to_ds, file_short_names, sls):\n\n def ida(sls):\n def read_subm(sls,i):\n tnm = sls[\"subm\"][i][\"name\"]\n FiN = sls[\"path\"] + tnm + \".csv\"\n return pd.read_csv(FiN).rename(columns={'target':tnm, sls[\"target\"]:tnm})\n \n dfs_subm = [read_subm(sls,i) for i in range(len(sls[\"subm\"]))]\n df_subms = pd.merge(dfs_subm[0], dfs_subm[1], on=['ID'])\n \n for i in range(2, len(sls[\"subm\"])): \n df_subms = pd.merge(df_subms, dfs_subm[i], on=['ID'])\n \n cols = [col for col in df_subms.columns if col != \"ID\"]\n short_name_cols = [c.replace(sls[\"prefix\"], '') for c in cols]\n \n weights1,corrects1 = [subm['weight'] for subm in sls[\"subm\"]], [wt for wt in sls[\"subwts\"] ]\n weights2,corrects2 = [subm['weight'] for subm in sls[\"subm2\"]],[wt for wt in sls[\"subwts2\"]]\n weights3,corrects3 = [subm['weight'] for subm in sls[\"subm3\"]],[wt for wt in sls[\"subwts3\"]]\n weights4,corrects4 = [subm['weight'] for subm in sls[\"subm4\"]],[wt for wt in sls[\"subwts4\"]]\n weights5,corrects5 = [subm['weight'] for subm in sls[\"subm5\"]],[wt for wt in sls[\"subwts5\"]]\n \n def alls(x, cs=cols):\n tes = {c: x[c] for c in cs}.items()\n subms_sorted = [\n t[0].replace(sls[\"prefix\"], '')\n for t in sorted(tes,key=lambda k:k[1],reverse=True if sls[\"sort\"]=='desc' else False)]\n return subms_sorted\n \n def correct(x, cs=cols, \n w1=weights1, cw1=corrects1, \n w2=weights2, cw2=corrects2,\n w3=weights3, cw3=corrects3,\n w4=weights4, cw4=corrects4,\n w5=weights5, cw5=corrects5,\n ):\n ic = [x['alls'].index(c) for c in short_name_cols]\n\n mxm = x['abs(mx-m)']\n\n if 0.00 < mxm <= 0.50:\n cS = [x[cols[j]] * (w1[j] + cw1[ic[j]]) for j in range(len(cols))]\n elif 0.50 < mxm <= 1.00:\n cS = [x[cols[j]] * (w2[j] + cw2[ic[j]]) for j in range(len(cols))]\n elif 1.00 < mxm <= 1.50:\n cS = [x[cols[j]] * (w3[j] + cw3[ic[j]]) for j in range(len(cols))]\n elif 1.50 < mxm <= 2.00:\n cS = [x[cols[j]] * (w4[j] + cw4[ic[j]]) for j in range(len(cols))]\n else:\n cS = [x[cols[j]] * (w5[j] + cw5[ic[j]]) for j in range(len(cols))]\n return sum(cS)\n\n def amxm(x, cs=cols):\n list_values = x[cs].to_list()\n mxm = abs(max(list_values)-min(list_values))\n return mxm\n\n df_subms['abs(mx-m)'] = df_subms.apply(lambda x: amxm (x), axis=1)\n \n df_subms['alls'] = df_subms.apply(lambda x: alls (x), axis=1)\n df_subms[sls[\"target\"]] = df_subms.apply(lambda x: correct(x), axis=1)\n \n schema_rename = { old_nc:new_shnc for old_nc, new_shnc in zip(cols, short_name_cols) }\n \n df_subms = df_subms.rename(columns=schema_rename)\n df_subms = df_subms.rename(columns={sls[\"target\"]:\"ensemble\"})\n \n df_subms.insert(loc=1, column=' _ ', value=[' '] * sls[\"q_rows\"])\n \n df_subms[' _ '] = df_subms[' _ '].astype(str)\n pd.set_option('display.max_rows',100)\n pd.set_option('display.float_format', '{:.4f}'.format)\n vcols = ['ID'] + [' _ '] + short_name_cols + [' _ '] + ['abs(mx-m)'] + [' _ '] + ['alls'] + [' _ '] + ['ensemble']\n df_subms = df_subms[vcols]\n display(df_subms.head(8))\n pd.set_option('display.float_format', '{:.7f}'.format)\n df_subms = df_subms.rename(columns={\"ensemble\":sls[\"target\"]})\n return df_subms\n\n sample_subm = pd.read_csv(path_to_ds + file_short_names[1] + \".csv\")\n\n def ensemble_ida(sls,submission=sample_subm): \n sls['sort'] = 'desc'\n dfs = ida(sls)\n dfD = dfs[['ID', sls['target']]]\n dfD.to_csv(f'tida_desc.csv', index=False)\n sls['sort'] = 'asc'\n dfs = ida(sls)\n dfA = dfs[['ID', sls['target']]]\n dfA.to_csv(f'tida_asc.csv', index=False)\n target,d,a = sls['target'],sls['desc'],sls['asc']\n submission[target] = dfD[target] * d + a * dfA[target]\n return submission\n\n submission = ensemble_ida(sls)\n \n return submission","metadata":{"_kg_hide-input":false,"execution":{"iopub.status.busy":"2025-07-22T17:11:11.084703Z","iopub.execute_input":"2025-07-22T17:11:11.085617Z","iopub.status.idle":"2025-07-22T17:11:11.105365Z","shell.execute_reply.started":"2025-07-22T17:11:11.085580Z","shell.execute_reply":"2025-07-22T17:11:11.104415Z"},"papermill":{"duration":0.023404,"end_time":"2025-07-16T08:55:42.306375","exception":false,"start_time":"2025-07-16T08:55:42.282971","status":"completed"},"tags":[],"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.83975','0.86767','0.89178']\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts': [+1.00, -0.40, -0.60], # LB = 0.93598\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.40, },\n# { 'name':file_short_names[1],'weight':0.60, },\n# { 'name':file_short_names[2],'weight':1.00, },\n# ]\n# }\n\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.88377','0.89178','0.90038']\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.45,\n# 'asc' : 0.55,\n# 'subwts': [+0.55, -0.20, -0.35], # LB = 0.89864\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.27, },\n# { 'name':file_short_names[1],'weight':0.33, },\n# { 'name':file_short_names[2],'weight':0.40, },\n# ]\n# }\n\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.88377','0.89178','0.90038']\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.20,\n# 'asc' : 0.80,\n# 'subwts': [+0.50, -0.20, -0.30], # LB = 0.90174\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.20, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.50, },\n# ]\n# }\n\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.83975','0.86767','0.88377','0.89178','0.90038']\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts': [+0.40, +0.05, -0.05,-0.15,-0.25], # LB = 0.93794\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.20, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.47, },\n# { 'name':file_short_names[4],'weight':0.02, },\n# ]\n# }\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.83975','0.86767','0.88377','0.89178','0.90038']\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts': [+0.20, +0.10, -0.05,-0.10,-0.15], # LB = 0.94828\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.20, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.20, },\n# { 'name':file_short_names[3],'weight':0.20, },\n# { 'name':file_short_names[4],'weight':0.20, },\n# ]\n# }\n\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts': [+0.20, +0.10, -0.05,-0.10,-0.15], # LB = 0.94857 (only Horizont)\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.20, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23, },\n# ],\n# 'subwts2': [+0.18, +0.09, -0.04,-0.09,-0.14], # LB = 0.94859\n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23, },\n# ]\n# }\n\n# # 5 solutions\n\n# # path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# # file_short_names = ['0.83975','0.86767','0.88377','0.89178','0.90038']\n\n# # option.1 ------- ------- ------- ------- ------- ------- ------- LB = 0.94915\n\n# # params = {\n# # 'path' : path_to_ds, \n# # 'sort' : \"dynamic\",\n# # 'target': \"prediction\",\n# # 'q_rows': 538_150,\n# # 'prefix': \"subm_\",\n# # 'desc' : 0.30,\n# # 'asc' : 0.70,\n# # 'subwts': [+0.19, +0.10, -0.05,-0.10,-0.14], \n# # 'subm' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # 'subwts2': [+0.17, +0.09, -0.04,-0.09,-0.13], \n# # 'subm2' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # 'subwts3': [+0.15, +0.08, -0.03,-0.08,-0.12], \n# # 'subm3' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # 'subwts4': [+0.14, +0.07, -0.02,-0.07,-0.11], \n# # 'subm4' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # 'subwts5': [+0.11, +0.06, -0.01,-0.06,-0.10], \n# # 'subm5' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # }\n\n# # ID\tprediction\n# # 0\t1\t-0.1542508\n# # 1\t2\t 0.2589274\n# # 2\t3\t-1.4203646\n# # 3\t4\t-0.2054324\n# # 4\t5\t 0.1035272\n# # ...\t...\t...\n# # 538145\t538146\t-0.3137427\n# # 538146\t538147\t 0.1481161\n# # 538147\t538148\t-0.9100966\n# # 538148\t538149\t 0.7786920\n# # 538149\t538150\t-0.4061519\n# # 538150 rows × 2 columns\n\n\n# # option.2 ------- ------- ------- ------- ------- ------- ------- LB = 0.94889\n\n\n# # params = {\n# # 'path' : path_to_ds, \n# # 'sort' : \"dynamic\",\n# # 'target': \"prediction\",\n# # 'q_rows': 538_150,\n# # 'prefix': \"subm_\",\n# # 'desc' : 0.30,\n# # 'asc' : 0.70,\n# # 'subwts': [+0.19, +0.10, -0.05,-0.10,-0.14], \n# # 'subm' : [\n# # { 'name':file_short_names[0],'weight':0.21, },\n# # { 'name':file_short_names[1],'weight':0.22, },\n# # { 'name':file_short_names[2],'weight':0.23, },\n# # { 'name':file_short_names[3],'weight':0.24, },\n# # { 'name':file_short_names[4],'weight':0.25, },\n# # ],\n# # 'subwts2': [+0.17, +0.09, -0.04,-0.09,-0.13], \n# # 'subm2' : [\n# # { 'name':file_short_names[0],'weight':0.22, },\n# # { 'name':file_short_names[1],'weight':0.23, },\n# # { 'name':file_short_names[2],'weight':0.24, },\n# # { 'name':file_short_names[3],'weight':0.25, },\n# # { 'name':file_short_names[4],'weight':0.26, },\n# # ],\n# # 'subwts3': [+0.15, +0.08, -0.03,-0.08,-0.12], \n# # 'subm3' : [\n# # { 'name':file_short_names[0],'weight':0.23, },\n# # { 'name':file_short_names[1],'weight':0.24, },\n# # { 'name':file_short_names[2],'weight':0.25, },\n# # { 'name':file_short_names[3],'weight':0.26, },\n# # { 'name':file_short_names[4],'weight':0.27, },\n# # ],\n# # 'subwts4': [+0.14, +0.07, -0.02,-0.07,-0.11], \n# # 'subm4' : [\n# # { 'name':file_short_names[0],'weight':0.24, },\n# # { 'name':file_short_names[1],'weight':0.25, },\n# # { 'name':file_short_names[2],'weight':0.26, },\n# # { 'name':file_short_names[3],'weight':0.27, },\n# # { 'name':file_short_names[4],'weight':0.28, },\n# # ],\n# # 'subwts5': [+0.11, +0.06, -0.01,-0.06,-0.10], \n# # 'subm5' : [\n# # { 'name':file_short_names[0],'weight':0.25, },\n# # { 'name':file_short_names[1],'weight':0.26, },\n# # { 'name':file_short_names[2],'weight':0.27, },\n# # { 'name':file_short_names[3],'weight':0.28, },\n# # { 'name':file_short_names[4],'weight':0.29, },\n# # ],\n# # }\n\n# # \tID\tprediction\n# # 0\t1\t-0.1673348\n# # 1\t2\t 0.2849654\n# # 2\t3\t-1.8152673\n# # 3\t4\t-0.2242138\n# # 4\t5\t 0.1223858\n# # ...\t...\t...\n# # 538145\t538146\t-0.3409893\n# # 538146\t538147\t 0.1720437\n# # 538147\t538148\t-1.1151354\n# # 538148\t538149\t 1.0286348\n# # 538149\t538150\t-0.4889277\n# # 538150 rows × 2 columns\n\n\n# # # option.3 ------- ------- ------- ------- ------- ------- ------- LB = 0.94717\n\n# # params = {\n# # 'path' : path_to_ds, \n# # 'sort' : \"dynamic\",\n# # 'target': \"prediction\",\n# # 'q_rows': 538_150,\n# # 'prefix': \"subm_\",\n# # 'desc' : 0.30,\n# # 'asc' : 0.70,\n# # 'subwts': [+0.19, +0.10, -0.05,-0.10,-0.14], \n# # 'subm' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # 'subwts2': [+0.17, +0.09, -0.04,-0.09,-0.13], \n# # 'subm2' : [\n# # { 'name':file_short_names[0],'weight':0.18, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.24, },\n# # ],\n# # 'subwts3': [+0.15, +0.08, -0.03,-0.08,-0.12], \n# # 'subm3' : [\n# # { 'name':file_short_names[0],'weight':0.17, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.25, },\n# # ],\n# # 'subwts4': [+0.14, +0.07, -0.02,-0.07,-0.11], \n# # 'subm4' : [\n# # { 'name':file_short_names[0],'weight':0.16, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.26, },\n# # ],\n# # 'subwts5': [+0.11, +0.06, -0.01,-0.06,-0.10], \n# # 'subm5' : [\n# # { 'name':file_short_names[0],'weight':0.15, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.27, },\n# # ],\n# # }\n\n\n# # ID\tprediction\n# # 0\t1\t-0.1542508\n# # 1\t2\t 0.2589274\n# # 2\t3\t-1.3990823\n# # 3\t4\t-0.2054324\n# # 4\t5\t 0.1073946\n# # ...\t...\t...\n# # 538145\t538146\t-0.3137427\n# # 538146\t538147\t 0.1482668\n# # 538147\t538148\t-0.8857567\n# # 538148\t538149\t 0.7211133\n# # 538149\t538150\t-0.4089252\n\n\n# # option.4 ------- ------- ------- ------- ------- ------- ------- LB = 0.94900\n\n# # params = {\n# # 'path' : path_to_ds, \n# # 'sort' : \"dynamic\",\n# # 'target': \"prediction\",\n# # 'q_rows': 538_150,\n# # 'prefix': \"subm_\",\n# # 'desc' : 0.26,\n# # 'asc' : 0.74,\n# # 'subwts': [+0.19, +0.10, -0.05,-0.10,-0.14], \n# # 'subm' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.23, },\n# # ],\n# # 'subwts2': [+0.17, +0.09, -0.04,-0.09,-0.13], \n# # 'subm2' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.225,},\n# # ],\n# # 'subwts3': [+0.15, +0.08, -0.03,-0.08,-0.12], \n# # 'subm3' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.220,},\n# # ],\n# # 'subwts4': [+0.14, +0.07, -0.02,-0.07,-0.11], \n# # 'subm4' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.215,},\n# # ],\n# # 'subwts5': [+0.11, +0.06, -0.01,-0.06,-0.10], \n# # 'subm5' : [\n# # { 'name':file_short_names[0],'weight':0.19, },\n# # { 'name':file_short_names[1],'weight':0.20, },\n# # { 'name':file_short_names[2],'weight':0.21, },\n# # { 'name':file_short_names[3],'weight':0.22, },\n# # { 'name':file_short_names[4],'weight':0.210,},\n# # ],\n# # }\n\n\n# # ID\tprediction\n# # 0\t1\t-0.1575648\n# # 1\t2\t 0.2557592\n# # 2\t3\t-1.4114128\n# # 3\t4\t-0.2071729\n# # 4\t5\t 0.0966335\n# # ...\t...\t...\n# # 538145\t538146\t-0.3189531\n# # 538146\t538147\t 0.1430990\n# # 538147\t538148\t-0.9122206\n# # 538148\t538149\t 0.7518473\n# # 538149\t538150\t-0.4125638\n# # 538150 rows × 2 columns\n\n\n# # option.5 ------- ------- ------- ------- ------- ------- ------- LB = ?\n\n# params = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.35,\n# 'asc' : 0.65,\n# 'subwts': [+0.19, +0.10, -0.05,-0.10,-0.14], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23, },\n# ],\n# 'subwts2': [+0.14, +0.10, +0.05,-0.10,-0.19], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23,},\n# ],\n# 'subwts3': [+0.25, +0.25, 0, -0.25,-0.25], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23,},\n# ],\n# 'subwts4': [-0.05, -0.10, +0.30,-0.10,-0.05], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23,},\n# ],\n# 'subwts5': [+0.05, +0.10, -0.30, +0.10, +0.05], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# { 'name':file_short_names[4],'weight':0.23,},\n# ],\n# }\n\n# # \tID\tprediction\n# # 0\t1\t-0.1501083\n# # 1\t2\t 0.2628877\n# # 2\t3\t-1.4322685\n# # 3\t4\t-0.2032566\n# # 4\t5\t 0.1207345\n# # ...\t...\t...\n# # 538145\t538146\t-0.3072296\n# # 538146\t538147\t 0.1369802\n# # 538147\t538148\t-0.7683847\n# # 538148\t538149\t 0.9015924\n# # 538149\t538150\t-0.3144665\n# # 538150 rows × 2 columns\n","metadata":{"execution":{"iopub.execute_input":"2025-07-16T08:55:42.312371Z","iopub.status.busy":"2025-07-16T08:55:42.312057Z","iopub.status.idle":"2025-07-16T08:55:42.326646Z","shell.execute_reply":"2025-07-16T08:55:42.325848Z"},"papermill":{"duration":0.019492,"end_time":"2025-07-16T08:55:42.328151","exception":false,"start_time":"2025-07-16T08:55:42.308659","status":"completed"},"tags":[],"jupyter":{"source_hidden":true},"_kg_hide-input":true,"_kg_hide-output":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.83975','0.86767','0.89178','0.90038']\n\n# # option.1 -------------------------------------------------- LB = 0.94608 v.4\n\n# params_1 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts': [+0.15, +0.07, -0.07,-0.15], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts2': [+0.16, +0.08, -0.08,-0.16], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts3': [+0.17, +0.09, -0.09,-0.17], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts4': [+0.18, +0.10, -0.10,-0.18], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# }\n\n# # option.2 -------------------------------------------------- LB = 0.93874 v.5\n\n# params_2 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.24, -0.03, -0.08, -0.13], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts2': [+0.27, -0.04, -0.09, -0.14], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.18, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.20, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# ],\n# 'subwts3': [+0.30, -0.05, -0.10, -0.15], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.18, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.20, },\n# { 'name':file_short_names[3],'weight':0.22, },\n# ],\n# 'subwts4': [+0.33, +0.06, -0.11, -0.16], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.19, },\n# { 'name':file_short_names[1],'weight':0.20, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# }\n\n\n# path_to_ds ='/kaggle/input/15-juli-2025-drw/submission '\n\n# file_short_names = ['0.83975','0.86767','0.88377','0.89178']\n\n\n# # option.3 -------------------------------------------------- LB = 0.94671 v.6\n\n# params_3 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target': \"prediction\",\n# 'q_rows': 538_150,\n# 'prefix': \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts': [+0.15, +0.07, -0.07,-0.15], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts2': [+0.16, +0.08, -0.08,-0.16], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts3': [+0.17, +0.09, -0.09,-0.17], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts4': [+0.18, +0.10, -0.10,-0.18], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# }\n\n# # option.4 -------------------------------------------------- LB = 0.94807, v.8\n\n# params_4 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.10, +0.04, -0.04, -0.10], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts2': [+0.11, +0.05, -0.05, -0.11], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts3': [+0.12, +0.06, -0.06, -0.12], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts4': [+0.13, +0.07, -0.07, -0.13], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# }\n\n# # option.5 -------------------------------------------------- LB = 0.94812\n\n# params_5 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.08, +0.03, -0.03, -0.08], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts2': [+0.09, +0.04, -0.04, -0.09], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts3': [+0.10, +0.05, -0.05, -0.10], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# 'subwts4': [+0.13, +0.07, -0.07, -0.13], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.17, },\n# { 'name':file_short_names[1],'weight':0.19, },\n# { 'name':file_short_names[2],'weight':0.21, },\n# { 'name':file_short_names[3],'weight':0.23, },\n# ],\n# }","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-18T00:48:11.036550Z","iopub.execute_input":"2025-07-18T00:48:11.036862Z","iopub.status.idle":"2025-07-18T00:48:11.058876Z","shell.execute_reply.started":"2025-07-18T00:48:11.036840Z","shell.execute_reply":"2025-07-18T00:48:11.057631Z"},"_kg_hide-input":true,"_kg_hide-output":true,"jupyter":{"source_hidden":true}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# params = params_5\n\n# if params == params_1 or params == params_2: \n# file_short_names = ['0.83975','0.86767', '0.89178','0.90038' ]\n\n# if params == params_3 or params == params_4: \n# file_short_names = ['0.83975','0.86767','0.88377','0.89178' ]\n\n# if params == params_5: \n# file_short_names = ['0.83975','0.86767', '0.89178', '0.95002']","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-18T00:48:42.814863Z","iopub.execute_input":"2025-07-18T00:48:42.815178Z","iopub.status.idle":"2025-07-18T00:48:42.821492Z","shell.execute_reply.started":"2025-07-18T00:48:42.815158Z","shell.execute_reply":"2025-07-18T00:48:42.820343Z"},"jupyter":{"source_hidden":true},"_kg_hide-input":true,"_kg_hide-output":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# file_short_names = ['0.95004','0.95002', '0.95001']\n\n# path_to_ds ='/kaggle/input/19-juli-2025-drw/submission '\n\n# params_1 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.011, -0.004, -0.007], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.334, },\n# { 'name':file_short_names[1],'weight':0.333, },\n# { 'name':file_short_names[2],'weight':0.333, },\n# ],\n# 'subwts2': [+0.010, -0.003, -0.007], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.335, },\n# { 'name':file_short_names[1],'weight':0.333, },\n# { 'name':file_short_names[2],'weight':0.332, },\n# ],\n# 'subwts3': [+0.009, -0.004, -0.005], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.335, },\n# { 'name':file_short_names[1],'weight':0.332, },\n# { 'name':file_short_names[2],'weight':0.033, },\n# ],\n# }\n\n\n# params_2 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.50,\n# 'asc' : 0.50,\n# 'subwts' : [+0.003, +0.002, +0.001], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.334, },\n# { 'name':file_short_names[1],'weight':0.333, },\n# { 'name':file_short_names[2],'weight':0.333, },\n# ],\n# 'subwts2': [+0.003, +0.002, +0.001], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.335, },\n# { 'name':file_short_names[1],'weight':0.333, },\n# { 'name':file_short_names[2],'weight':0.332, },\n# ],\n# 'subwts3': [+0.003, +0.002, +0.001], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.335, },\n# { 'name':file_short_names[1],'weight':0.332, },\n# { 'name':file_short_names[2],'weight':0.033, },\n# ],\n# }\n\n\n# params_3 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.011, -0.004, -0.007], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.970, },\n# { 'name':file_short_names[1],'weight':0.015, },\n# { 'name':file_short_names[2],'weight':0.015, },\n# ],\n# 'subwts2': [+0.010, -0.003, -0.007], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.970, },\n# { 'name':file_short_names[1],'weight':0.020, },\n# { 'name':file_short_names[2],'weight':0.010, },\n# ],\n# 'subwts3': [+0.009, -0.004, -0.005], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.970, },\n# { 'name':file_short_names[1],'weight':0.010, },\n# { 'name':file_short_names[2],'weight':0.020, },\n# ],\n# }\n\n\n# params_4 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.50,\n# 'asc' : 0.50,\n# 'subwts' : [+0.010, +0.005, +0.002], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.334, },\n# { 'name':file_short_names[1],'weight':0.333, },\n# { 'name':file_short_names[2],'weight':0.333, },\n# ],\n# 'subwts2': [+0.015, +0.010, +0.005], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.335, },\n# { 'name':file_short_names[1],'weight':0.333, },\n# { 'name':file_short_names[2],'weight':0.332, },\n# ],\n# 'subwts3': [+0.020, +0.015, +0.010], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.335, },\n# { 'name':file_short_names[1],'weight':0.332, },\n# { 'name':file_short_names[2],'weight':0.033, },\n# ],\n# }\n\n\n# params_5 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.29,\n# 'asc' : 0.71,\n# 'subwts' : [+0.011, -0.004, -0.007], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.974, },\n# { 'name':file_short_names[1],'weight':0.013, },\n# { 'name':file_short_names[2],'weight':0.013, },\n# ],\n# 'subwts2': [+0.005, +0.003, +0.002], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.974, },\n# { 'name':file_short_names[1],'weight':0.013, },\n# { 'name':file_short_names[2],'weight':0.013, },\n# ],\n# 'subwts3': [+0.009, -0.004, -0.005], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.980, },\n# { 'name':file_short_names[1],'weight':0.010, },\n# { 'name':file_short_names[2],'weight':0.010, },\n# ],\n# }\n\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-19T00:38:18.228082Z","iopub.execute_input":"2025-07-19T00:38:18.228413Z","iopub.status.idle":"2025-07-19T00:38:18.237058Z","shell.execute_reply.started":"2025-07-19T00:38:18.228366Z","shell.execute_reply":"2025-07-19T00:38:18.235888Z"},"_kg_hide-input":true,"_kg_hide-output":true,"jupyter":{"source_hidden":true}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# file_short_names = ['0.95129','0.95109', 'nodata_1', 'nodata_3', 'nodata_7']\n\n# path_to_ds ='/kaggle/input/20-juli-2025-drw/submission '\n\n# params_1 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.10, -0.01, -0.02, -0.03, -0.04], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.40, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.10, },\n# { 'name':file_short_names[3],'weight':0.10, },\n# { 'name':file_short_names[4],'weight':0.10, },\n# ],\n# 'subwts2': [+0.10, -0.01, -0.02, -0.03, -0.04], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.43, },\n# { 'name':file_short_names[1],'weight':0.33, },\n# { 'name':file_short_names[2],'weight':0.08, },\n# { 'name':file_short_names[3],'weight':0.08, },\n# { 'name':file_short_names[4],'weight':0.08, },\n# ],\n# 'subwts3': [+0.10, -0.01, -0.02, -0.03, -0.04], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.45, },\n# { 'name':file_short_names[1],'weight':0.34, },\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# { 'name':file_short_names[4],'weight':0.07, },\n# ],\n# 'subwts4': [+0.10, -0.01, -0.02, -0.03, -0.04], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.36, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.12, },\n# { 'name':file_short_names[3],'weight':0.12, },\n# { 'name':file_short_names[4],'weight':0.12, },\n# ],\n# 'subwts5': [+0.10, -0.01, -0.02, -0.03, -0.04], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.33, },\n# { 'name':file_short_names[1],'weight':0.25, },\n# { 'name':file_short_names[2],'weight':0.14, },\n# { 'name':file_short_names[3],'weight':0.14, },\n# { 'name':file_short_names[4],'weight':0.14, },\n# ],\n# }\n\n# params_2 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.05, +0.03, -0.01, -0.02, -0.05], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.55, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.05, },\n# ],\n# 'subwts2' : [+0.05, +0.03, -0.01, -0.02, -0.05], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.55, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.05, },\n# ],\n# 'subwts3' : [+0.05, +0.03, -0.01, -0.02, -0.05], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.55, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.05, },\n# ],\n# 'subwts4' : [+0.05, +0.03, -0.01, -0.02, -0.05], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.55, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.05, },\n# ],\n# 'subwts5' : [+0.05, +0.03, -0.01, -0.02, -0.05], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.55, },\n# { 'name':file_short_names[1],'weight':0.30, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.05, },\n# ],\n# }\n\n# params_3 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts3' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts4' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts5' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }\n\n# params_4 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.05, +0.04, +0.03, +0.02, +0.01], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.05, +0.04, +0.03, +0.02, +0.01], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts3' : [+0.05, +0.04, +0.03, +0.02, +0.01], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts4' : [+0.05, +0.04, +0.03, +0.02, +0.01], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts5' : [+0.05, +0.04, +0.03, +0.02, +0.01], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.10, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }\n\n# params_5 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.95, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.95, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts3' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.95, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts4' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.95, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts5' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.95, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-20T08:58:51.588652Z","iopub.execute_input":"2025-07-20T08:58:51.589180Z","iopub.status.idle":"2025-07-20T08:58:51.598207Z","shell.execute_reply.started":"2025-07-20T08:58:51.589157Z","shell.execute_reply":"2025-07-20T08:58:51.597458Z"},"_kg_hide-input":true,"_kg_hide-output":true,"jupyter":{"source_hidden":true}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# file_short_names = ['0.95129','0.95109', 'nodata_1', 'nodata_3', 'nodata_7']\n\n# path_to_ds ='/kaggle/input/20-juli-2025-drw/submission '\n\n# params_8 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.90, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.90, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts3' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.90, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts4' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.90, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts5' : [+0.02, +0.01, -0.005, -0.01, -0.015], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.90, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.03, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }\n\n# params_9 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.025, +0.01, -0.005, -0.01, -0.02], \n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.94, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.01, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.025, +0.01, -0.005, -0.01, -0.02], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.94, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.01, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts3' : [+0.025, +0.01, -0.005, -0.01, -0.02], \n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.94, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.01, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts4' : [+0.025, +0.01, -0.005, -0.01, -0.02], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.94, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.01, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts5' : [+0.025, +0.01, -0.005, -0.01, -0.02], \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.94, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.01, },\n# { 'name':file_short_names[3],'weight':0.01, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }\n\n\n# params = params_1 # LB=0.94780\n# params = params_2 # LB=0.94950\n# params = params_3 # LB=0.95096\n# params = params_4 # LB=?\n# params = params_5 # LB=0.95098\n# params = params_8 # LB=0.95096\n# params = params_9 # LB=0.95103","metadata":{"trusted":true,"_kg_hide-input":true,"_kg_hide-output":true,"jupyter":{"source_hidden":true}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# file_short_names = ['0.95109','0.95004', '0.95002', '0.94857']\n\n# path_to_ds ='/kaggle/input/21-juli-2025-drw/submission '\n\n# params_10 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.015, 0.000, -0.005, -0.010],\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# 'subwts2' : [+0.015, 0.000, -0.005, -0.010], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# 'subwts3' : [+0.015, 0.000, -0.005, -0.010],\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# 'subwts4' : [+0.015, 0.000, -0.005, -0.010], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# }\n\n# params_11 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.015, 0.000, -0.005, -0.010],\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# 'subwts2' : [+0.020, -0.003, -0.007, -0.010], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.87, },\n# { 'name':file_short_names[1],'weight':0.043,},\n# { 'name':file_short_names[2],'weight':0.043,},\n# { 'name':file_short_names[3],'weight':0.044,},\n# ],\n# 'subwts3' : [+0.025, -0.003, -0.007, -0.015],\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.89, },\n# { 'name':file_short_names[1],'weight':0.037,},\n# { 'name':file_short_names[2],'weight':0.037,},\n# { 'name':file_short_names[3],'weight':0.037,},\n# ],\n# 'subwts4' : [+0.030, -0.005, -0.010, -0.015], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.91, },\n# { 'name':file_short_names[1],'weight':0.030,},\n# { 'name':file_short_names[2],'weight':0.030,},\n# { 'name':file_short_names[3],'weight':0.030,},\n# ],\n# }\n\n# params_12 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.015, 0.000, -0.005, -0.010],\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# 'subwts2' : [+0.020, 0.000, -0.007, -0.013], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.053,},\n# { 'name':file_short_names[2],'weight':0.053,},\n# { 'name':file_short_names[3],'weight':0.054,},\n# ],\n# 'subwts3' : [+0.025, 0.000, -0.010, -0.015],\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.057,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# ],\n# 'subwts4' : [+0.030, 0.000, -0.010, -0.020], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.07, },\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# ],\n# }","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-21T04:06:07.571977Z","iopub.execute_input":"2025-07-21T04:06:07.572282Z","iopub.status.idle":"2025-07-21T04:06:07.580852Z","shell.execute_reply.started":"2025-07-21T04:06:07.572258Z","shell.execute_reply":"2025-07-21T04:06:07.579945Z"},"_kg_hide-input":true,"_kg_hide-output":true,"jupyter":{"source_hidden":true}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Archive\n\n# file_short_names = ['0.95109','0.95004', '0.95002', '0.94857'] # LB=0.95159\n\n# params_12 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.015, 0.000, -0.005, -0.010],\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.85, },\n# { 'name':file_short_names[1],'weight':0.05, }, # LB=0.95159\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# ],\n# 'subwts2' : [+0.020, 0.000, -0.007, -0.013], \n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.053,}, # LB=0.95159\n# { 'name':file_short_names[2],'weight':0.053,},\n# { 'name':file_short_names[3],'weight':0.054,},\n# ],\n# 'subwts3' : [+0.025, 0.000, -0.010, -0.015],\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.057,}, # LB=0.95159\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# ],\n# 'subwts4' : [+0.030, 0.000, -0.010, -0.020], \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.07, }, # LB=0.95159\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# ],\n# }\n\n\n# file_short_names = ['0.95109','0.95004', '0.95002', '0.94857', '0.89954']\n\n# params_13 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.30,\n# 'asc' : 0.70,\n# 'subwts' : [+0.015, +0.002, -0.002, -0.005, -0.010], # LB=0.95155\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.020, +0.002, -0.002, -0.007, -0.013], # LB=0.95155\n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.053,},\n# { 'name':file_short_names[2],'weight':0.053,},\n# { 'name':file_short_names[3],'weight':0.054,},\n# { 'name':file_short_names[4],'weight':0.010,},\n# ],\n# 'subwts3' : [+0.025, +0.002, -0.002, -0.010, -0.015], # LB=0.95155\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.82, },\n# { 'name':file_short_names[1],'weight':0.057,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# { 'name':file_short_names[4],'weight':0.010,},\n# ],\n# 'subwts4' : [+0.030, +0.002, -0.002, -0.010, -0.020], # LB=0.95155 \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.07, },\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# { 'name':file_short_names[4],'weight':0.00, },\n# ],\n# 'subwts5' : [+0.035, +0.002, -0.002, -0.012, -0.023], # LB=0.95155 \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.78, },\n# { 'name':file_short_names[1],'weight':0.07, },\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }\n\n# file_short_names = ['0.95109','0.95004', '0.95002', '0.94857', '0.90222']\n\n# params_14 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.35,\n# 'asc' : 0.65,\n# 'subwts' : [+0.015, +0.002, -0.002, -0.005, -0.010], # LB=?\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.05, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# 'subwts2' : [+0.020, +0.002, -0.002, -0.007, -0.013], # LB=?\n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.053,},\n# { 'name':file_short_names[2],'weight':0.053,},\n# { 'name':file_short_names[3],'weight':0.054,},\n# { 'name':file_short_names[4],'weight':0.010,},\n# ],\n# 'subwts3' : [+0.025, +0.002, -0.002, -0.010, -0.015], # LB=?\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.82, },\n# { 'name':file_short_names[1],'weight':0.057,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# { 'name':file_short_names[4],'weight':0.010,},\n# ],\n# 'subwts4' : [+0.030, +0.002, -0.002, -0.010, -0.020], # LB=? \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.07, },\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# { 'name':file_short_names[4],'weight':0.00, },\n# ],\n# 'subwts5' : [+0.035, +0.002, -0.002, -0.012, -0.023], # LB=? \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.78, },\n# { 'name':file_short_names[1],'weight':0.07, },\n# { 'name':file_short_names[2],'weight':0.07, },\n# { 'name':file_short_names[3],'weight':0.07, },\n# { 'name':file_short_names[4],'weight':0.01, },\n# ],\n# }","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-22T17:13:56.308531Z","iopub.execute_input":"2025-07-22T17:13:56.308800Z","iopub.status.idle":"2025-07-22T17:13:56.323413Z","shell.execute_reply.started":"2025-07-22T17:13:56.308779Z","shell.execute_reply":"2025-07-22T17:13:56.322502Z"},"jupyter":{"source_hidden":true},"_kg_hide-input":true,"_kg_hide-output":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"path_to_ds ='/kaggle/input/21-juli-2025-drw/submission '","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-22T17:11:53.269221Z","iopub.execute_input":"2025-07-22T17:11:53.269521Z","iopub.status.idle":"2025-07-22T17:11:53.274199Z","shell.execute_reply.started":"2025-07-22T17:11:53.269497Z","shell.execute_reply":"2025-07-22T17:11:53.272992Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# file_short_names = ['0.95109','hs90038', '0.95002', '0.94857', '0.90222']\n\n# params_15 = {\n# 'path' : path_to_ds, \n# 'sort' : \"dynamic\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.35,\n# 'asc' : 0.65,\n# 'subwts' : [+0.015, +0.002, -0.002, -0.005, -0.010], # LB=0.95163\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.05, },\n# { 'name':file_short_names[3],'weight':0.05, },\n# { 'name':file_short_names[4],'weight':0.03, },\n# ],\n# 'subwts2' : [+0.020, +0.002, -0.002, -0.007, -0.013], # LB=0.95163\n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.033,},\n# { 'name':file_short_names[2],'weight':0.053,},\n# { 'name':file_short_names[3],'weight':0.054,},\n# { 'name':file_short_names[4],'weight':0.030,},\n# ],\n# 'subwts3' : [+0.025, +0.002, -0.002, -0.010, -0.015], # LB=0.95163\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.82, },\n# { 'name':file_short_names[1],'weight':0.037,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# { 'name':file_short_names[4],'weight':0.030,},\n# ],\n# 'subwts4' : [+0.030, +0.002, -0.002, -0.010, -0.020], # LB=0.95163 \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.03, },\n# { 'name':file_short_names[2],'weight':0.08, },\n# { 'name':file_short_names[3],'weight':0.08, },\n# { 'name':file_short_names[4],'weight':0.02, },\n# ],\n# 'subwts5' : [+0.035, +0.002, -0.002, -0.012, -0.023], # LB=0.95163 \n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.78, },\n# { 'name':file_short_names[1],'weight':0.02, },\n# { 'name':file_short_names[2],'weight':0.09, },\n# { 'name':file_short_names[3],'weight':0.09, },\n# { 'name':file_short_names[4],'weight':0.02, },\n# ],\n# }\n\n# params_18 = {\n# 'path' : path_to_ds, \n# 'sort' : \"desc/asc\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.355, # !!!\n# 'asc' : 0.655, # !!!\n# 'subwts' : [+0.015, +0.002, -0.002, -0.005, -0.010], # LB=0.95158 # !!!\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.025, },\n# { 'name':file_short_names[2],'weight':0.055, },\n# { 'name':file_short_names[3],'weight':0.055, },\n# { 'name':file_short_names[4],'weight':0.025, },\n# ],\n# 'subwts2' : [+0.020, +0.002, -0.002, -0.007, -0.013], # LB=0.95158 # !!!\n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.028,},\n# { 'name':file_short_names[2],'weight':0.058,},\n# { 'name':file_short_names[3],'weight':0.055,},\n# { 'name':file_short_names[4],'weight':0.029,},\n# ],\n# 'subwts3' : [+0.025, +0.002, -0.002, -0.010, -0.015], # LB=0.95158 # !!!\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.82, },\n# { 'name':file_short_names[1],'weight':0.037,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# { 'name':file_short_names[4],'weight':0.030,},\n# ],\n# 'subwts4' : [+0.030, +0.002, -0.002, -0.010, -0.020], # LB=0.95158 # !!! \n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.02, },\n# { 'name':file_short_names[2],'weight':0.085,},\n# { 'name':file_short_names[3],'weight':0.085,},\n# { 'name':file_short_names[4],'weight':0.02, },\n# ],\n# 'subwts5' : [+0.035, +0.002, -0.002, -0.012, -0.023], # LB=0.95158 # !!!\n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.78, },\n# { 'name':file_short_names[1],'weight':0.02, },\n# { 'name':file_short_names[2],'weight':0.09, },\n# { 'name':file_short_names[3],'weight':0.09, },\n# { 'name':file_short_names[4],'weight':0.02, },\n# ],\n# }\n\n# params_19 = {\n# 'path' : path_to_ds, \n# 'sort' : \"desc/asc\",\n# 'target' : \"prediction\",\n# 'q_rows' : 538_150,\n# 'prefix' : \"subm_\",\n# 'desc' : 0.355,\n# 'asc' : 0.645,\n# 'subwts' : [+0.015, +0.002, -0.002, -0.005, -0.010], # LB=?\n# 'subm' : [\n# { 'name':file_short_names[0],'weight':0.84, },\n# { 'name':file_short_names[1],'weight':0.028, },\n# { 'name':file_short_names[2],'weight':0.052, },\n# { 'name':file_short_names[3],'weight':0.052, },\n# { 'name':file_short_names[4],'weight':0.028, },\n# ],\n# 'subwts2' : [+0.020, +0.002, -0.002, -0.007, -0.013], # LB=?\n# 'subm2' : [\n# { 'name':file_short_names[0],'weight':0.83, },\n# { 'name':file_short_names[1],'weight':0.029,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.054,},\n# { 'name':file_short_names[4],'weight':0.030,},\n# ],\n# 'subwts3' : [+0.025, +0.002, -0.002, -0.010, -0.015], # LB=?\n# 'subm3' : [\n# { 'name':file_short_names[0],'weight':0.82, },\n# { 'name':file_short_names[1],'weight':0.037,},\n# { 'name':file_short_names[2],'weight':0.057,},\n# { 'name':file_short_names[3],'weight':0.057,},\n# { 'name':file_short_names[4],'weight':0.030,},\n# ],\n# 'subwts4' : [+0.030, +0.002, -0.002, -0.010, -0.020], # LB=?\n# 'subm4' : [\n# { 'name':file_short_names[0],'weight':0.79, },\n# { 'name':file_short_names[1],'weight':0.02, },\n# { 'name':file_short_names[2],'weight':0.085,},\n# { 'name':file_short_names[3],'weight':0.085,},\n# { 'name':file_short_names[4],'weight':0.02, },\n# ],\n# 'subwts5' : [+0.035, +0.002, -0.002, -0.012, -0.023], # LB=?\n# 'subm5' : [\n# { 'name':file_short_names[0],'weight':0.78, },\n# { 'name':file_short_names[1],'weight':0.02, },\n# { 'name':file_short_names[2],'weight':0.09, },\n# { 'name':file_short_names[3],'weight':0.09, },\n# { 'name':file_short_names[4],'weight':0.02, },\n# ],\n# }","metadata":{"trusted":true,"_kg_hide-input":true,"_kg_hide-output":true,"jupyter":{"source_hidden":true}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"file_short_names = ['0.95165','hs90038', '0.95109', '0.94857', '0.90222']\n\nparams_20 = {\n 'path' : path_to_ds, \n 'sort' : \"asc/desc\",\n 'target' : \"prediction\",\n 'q_rows' : 538_150,\n 'prefix' : \"subm_\",\n 'desc' : 0.355,\n 'asc' : 0.645,\n \n 'subwts' : [+0.015, +0.002, -0.002, -0.005, -0.010], # LB=? # LB=0.95163.[0.350 0.650]\n 'subm' : [\n { 'name':file_short_names[0],'weight':0.84, },\n { 'name':file_short_names[1],'weight':0.03, },\n { 'name':file_short_names[2],'weight':0.05, },\n { 'name':file_short_names[3],'weight':0.05, },\n { 'name':file_short_names[4],'weight':0.03, },\n ],\n 'subwts2' : [+0.020, +0.002, -0.002, -0.007, -0.013], # LB=? # LB=0.95163.[0.350 0.650]\n 'subm2' : [\n { 'name':file_short_names[0],'weight':0.83, },\n { 'name':file_short_names[1],'weight':0.033,},\n { 'name':file_short_names[2],'weight':0.053,},\n { 'name':file_short_names[3],'weight':0.054,},\n { 'name':file_short_names[4],'weight':0.030,},\n ],\n 'subwts3' : [+0.025, +0.002, -0.002, -0.010, -0.015], # LB=? # LB=0.95163.[0.350 0.650]\n 'subm3' : [\n { 'name':file_short_names[0],'weight':0.82, },\n { 'name':file_short_names[1],'weight':0.037,},\n { 'name':file_short_names[2],'weight':0.057,},\n { 'name':file_short_names[3],'weight':0.057,},\n { 'name':file_short_names[4],'weight':0.030,},\n ],\n 'subwts4' : [+0.030, +0.002, -0.002, -0.010, -0.020], # LB=? # LB=0.95163.[0.350 0.650]\n 'subm4' : [\n { 'name':file_short_names[0],'weight':0.79, },\n { 'name':file_short_names[1],'weight':0.03, },\n { 'name':file_short_names[2],'weight':0.08, },\n { 'name':file_short_names[3],'weight':0.08, },\n { 'name':file_short_names[4],'weight':0.02, },\n ],\n 'subwts5' : [+0.035, +0.002, -0.002, -0.012, -0.023], # LB=? # LB=0.95163.[0.350 0.650]\n 'subm5' : [\n { 'name':file_short_names[0],'weight':0.78, },\n { 'name':file_short_names[1],'weight':0.02, },\n { 'name':file_short_names[2],'weight':0.09, },\n { 'name':file_short_names[3],'weight':0.09, },\n { 'name':file_short_names[4],'weight':0.02, },\n ],\n }","metadata":{"trusted":true},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# params = params_10 # LB=0.95131\n# params = params_11 # LB=0.95057\n# params = params_12 # LB=0.95159\n# params = params_13 # Lb=0.95155\n# params = params_14 # Lb=0.95164 <\n# params = params_15 # Lb=0.95163 <\n# params = params_18 # Lb=0.95158\n# params = params_19 # Lb=0.95161\n\nparams = params_20","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-07-22T17:14:19.597912Z","iopub.execute_input":"2025-07-22T17:14:19.598250Z","iopub.status.idle":"2025-07-22T17:14:19.602718Z","shell.execute_reply.started":"2025-07-22T17:14:19.598227Z","shell.execute_reply":"2025-07-22T17:14:19.601772Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df = iBlend ( path_to_ds, file_short_names, params )\n\ndf.to_csv('submission.csv', index=False)\n\ndisplay(df)","metadata":{"execution":{"iopub.status.busy":"2025-07-22T17:14:26.756612Z","iopub.execute_input":"2025-07-22T17:14:26.757439Z","iopub.status.idle":"2025-07-22T17:19:56.548680Z","shell.execute_reply.started":"2025-07-22T17:14:26.757408Z","shell.execute_reply":"2025-07-22T17:19:56.547894Z"},"papermill":{"duration":366.852199,"end_time":"2025-07-16T09:01:49.182681","exception":false,"start_time":"2025-07-16T08:55:42.330482","status":"completed"},"tags":[],"trusted":true},"outputs":[],"execution_count":null}]}