Reomved defekt training
This commit is contained in:
575 files changed
+382
-926
No files matched your search
+382
-65
@@ -2545,13 +2545,41 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 115,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"array([-1.53249554e-06, -1.91561943e-06, -2.39452428e-06, -2.99315535e-06,\n",
|
||||
" -3.74144419e-06, -4.67680524e-06, -5.84600655e-06, -7.30750819e-06,\n",
|
||||
" -9.13438523e-06, -1.14179815e-05, -1.42724769e-05, -1.78405962e-05,\n",
|
||||
" -2.23007452e-05, -2.78759315e-05, -3.48449144e-05, -4.35561430e-05,\n",
|
||||
" -5.44451787e-05, -6.80564734e-05, -8.50705917e-05, -1.06338240e-04,\n",
|
||||
" -1.32922800e-04, -1.66153499e-04, -2.07691874e-04, -2.59614843e-04,\n",
|
||||
" -3.24518554e-04, -4.05648192e-04, -5.07060240e-04, -6.33825300e-04,\n",
|
||||
" -7.92281625e-04, -9.90352031e-04, -1.23794004e-03, -1.54742505e-03,\n",
|
||||
" -1.93428131e-03, -2.41785164e-03, -3.02231455e-03, -3.77789319e-03,\n",
|
||||
" -4.72236648e-03, -5.90295810e-03, -7.37869763e-03, -9.22337204e-03,\n",
|
||||
" -1.15292150e-02, -1.44115188e-02, -1.80143985e-02, -2.25179981e-02,\n",
|
||||
" -2.81474977e-02, -3.51843721e-02, -4.39804651e-02, -5.49755814e-02,\n",
|
||||
" -6.87194767e-02, -8.58993459e-02, -1.07374182e-01, -1.34217728e-01,\n",
|
||||
" -1.67772160e-01, -2.09715200e-01, -2.62144000e-01, -3.27680000e-01,\n",
|
||||
" -4.09600000e-01, -5.12000000e-01, -6.40000000e-01, -6.40000000e-01,\n",
|
||||
" -8.00000000e-01, -8.00000000e-01, -1.00000000e+00, -1.00000000e+00,\n",
|
||||
" -1.00000000e+00, -1.00000000e+00, -1.00000000e+00, -1.00000000e+00,\n",
|
||||
" -1.00000000e+00, -1.00000000e+00])"
|
||||
]
|
||||
},
|
||||
"execution_count": 115,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"calculate_q_reword(\n",
|
||||
" _board_history, gamma=0.8, who_won_fraction=1, final_score_fraction=0\n",
|
||||
@@ -2560,13 +2588,41 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 116,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"array([ 1.53941436, -1.82573204, 1.46783494, -1.91520632,\n",
|
||||
" 1.3559921 , -4.55500988, 0.55623766, -3.05470293,\n",
|
||||
" -0.06837866, -6.33547333, 0.83065834, -2.71167707,\n",
|
||||
" 0.36040366, -3.29949543, 2.12563071, -3.59296161,\n",
|
||||
" 6.75879799, 4.69849749, 9.62312186, 5.77890232,\n",
|
||||
" 13.47362791, 8.09203488, 13.8650436 , 1.0813045 ,\n",
|
||||
" 5.10163063, 0.12703829, 3.90879786, 1.13599733,\n",
|
||||
" 5.16999666, -4.78750418, 0.26561978, -3.41797528,\n",
|
||||
" 1.9775309 , -3.77808638, -0.97260797, -7.46575996,\n",
|
||||
" 1.91780005, -1.35274994, 4.55906257, -8.05117178,\n",
|
||||
" 1.18603527, -4.76745591, 0.29068011, -3.38664986,\n",
|
||||
" -0.48331232, -4.3541404 , 3.3073245 , 0.38415562,\n",
|
||||
" 4.23019452, 1.53774316, 8.17217894, 1.46522368,\n",
|
||||
" 5.5815296 , -4.273088 , -1.59136 , -15.7392 ,\n",
|
||||
" -8.424 , -24.28 , -26.6 , -26.6 ,\n",
|
||||
" -17. , -17. , 0. , 0. ,\n",
|
||||
" 0. , 0. , 0. , 0. ,\n",
|
||||
" 0. , 0. ])"
|
||||
]
|
||||
},
|
||||
"execution_count": 116,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"calculate_q_reword(\n",
|
||||
" _board_history, gamma=0.8, who_won_fraction=0, final_score_fraction=0\n",
|
||||
@@ -2575,7 +2631,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 117,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
@@ -2602,7 +2658,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 118,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
@@ -2659,7 +2715,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 119,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
@@ -2675,14 +2731,25 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 120,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"((70, 100, 8, 8), (70, 100, 2))"
|
||||
]
|
||||
},
|
||||
"execution_count": 120,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"_board_history, _action_history = simulate_game(100, (RandomPolicy(1), RandomPolicy(1)))\n",
|
||||
"_board_history.shape, _action_history.shape"
|
||||
@@ -2690,14 +2757,43 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 121,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"6.31 ms ± 334 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)\n",
|
||||
"peak memory: 580.32 MiB, increment: 0.04 MiB\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"image/png": "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 truncated
|
||||
"text/plain": [
|
||||
"<Figure size 1200x600 with 8 Axes>"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"image/png": "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 truncated
|
||||
"text/plain": [
|
||||
"<Figure size 1200x600 with 8 Axes>"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"def action_to_q_learning_format(\n",
|
||||
" board_history: np.ndarray, action_history: np.ndarray\n",
|
||||
@@ -2734,14 +2830,33 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 122,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"name": "stdout",
|
||||
"output_type": "stream",
|
||||
"text": [
|
||||
"46.4 ms ± 1.36 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)\n",
|
||||
"peak memory: 576.92 MiB, increment: 0.01 MiB\n"
|
||||
]
|
||||
},
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"(2, 2, 2, 70, 100, 2, 8, 8)"
|
||||
]
|
||||
},
|
||||
"execution_count": 122,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"def build_symetry_action(\n",
|
||||
" board_history: np.ndarray, action_history: np.ndarray\n",
|
||||
@@ -2773,7 +2888,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 123,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
@@ -2810,7 +2925,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 124,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
@@ -3074,7 +3189,7 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 125,
|
||||
"metadata": {
|
||||
"collapsed": false,
|
||||
"jupyter": {
|
||||
@@ -3084,7 +3199,18 @@
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"(70, 10, 8, 8)"
|
||||
]
|
||||
},
|
||||
"execution_count": 125,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"_train_boards, _train_actions = simulate_game(10, (RandomPolicy(0), RandomPolicy(0)))\n",
|
||||
"_action_possible = ~np.all(_train_actions[:, :] == -1, axis=2)\n",
|
||||
@@ -3138,14 +3264,25 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 126,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW00-FSF10-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 126,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy1 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3160,14 +3297,36 @@
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 141,
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"ql_policys = []"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": 142,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW00-FSF10-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 142,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy1 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3177,19 +3336,31 @@
|
||||
" who_won_fraction=0,\n",
|
||||
" final_score_fraction=1,\n",
|
||||
")\n",
|
||||
"ql_policys.append(ql_policy1)\n",
|
||||
"ql_policy1.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 143,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G08-WW00-FSF10-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 143,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy2 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3199,19 +3370,31 @@
|
||||
" who_won_fraction=0,\n",
|
||||
" final_score_fraction=1,\n",
|
||||
")\n",
|
||||
"ql_policys.append(ql_policy2)\n",
|
||||
"ql_policy2.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 144,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G10-WW00-FSF10-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 144,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy3 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3221,19 +3404,31 @@
|
||||
" who_won_fraction=0,\n",
|
||||
" final_score_fraction=1,\n",
|
||||
")\n",
|
||||
"ql_policys.append(ql_policy3)\n",
|
||||
"ql_policy3.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 145,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW10-FSF00-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 145,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy4 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3243,12 +3438,13 @@
|
||||
" who_won_fraction=1,\n",
|
||||
" final_score_fraction=0,\n",
|
||||
")\n",
|
||||
"ql_policys.append(ql_policy4)\n",
|
||||
"ql_policy4.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 146,
|
||||
"metadata": {
|
||||
"collapsed": false,
|
||||
"jupyter": {
|
||||
@@ -3258,7 +3454,18 @@
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW03-FSF03-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 146,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy5 = QLPolicy(\n",
|
||||
" 0.95,\n",
|
||||
@@ -3268,12 +3475,13 @@
|
||||
" who_won_fraction=0.3,\n",
|
||||
" final_score_fraction=0.3,\n",
|
||||
")\n",
|
||||
"ql_policys.append(ql_policy5)\n",
|
||||
"ql_policy5.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 147,
|
||||
"metadata": {
|
||||
"collapsed": false,
|
||||
"jupyter": {
|
||||
@@ -3283,7 +3491,18 @@
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW03-FSF07-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 147,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy6 = QLPolicy(\n",
|
||||
" 0.95,\n",
|
||||
@@ -3293,19 +3512,31 @@
|
||||
" who_won_fraction=0.3,\n",
|
||||
" final_score_fraction=0.65,\n",
|
||||
")\n",
|
||||
"ql_policys.append(ql_policy6)\n",
|
||||
"ql_policy6.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 148,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW02-FSF07-DQLSimple-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 148,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy7 = QLPolicy(\n",
|
||||
" 0.95,\n",
|
||||
@@ -3315,19 +3546,31 @@
|
||||
" who_won_fraction=0.2,\n",
|
||||
" final_score_fraction=0.65,\n",
|
||||
")\n",
|
||||
"ql_policy6.policy_name"
|
||||
"ql_policys.append(ql_policy7)\n",
|
||||
"ql_policy7.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 149,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G08-WW00-FSF10-DQLNet-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 149,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy8 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3337,19 +3580,31 @@
|
||||
" who_won_fraction=0,\n",
|
||||
" final_score_fraction=1,\n",
|
||||
")\n",
|
||||
"ql_policy2.policy_name"
|
||||
"ql_policys.append(ql_policy8)\n",
|
||||
"ql_policy8.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 150,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G10-WW00-FSF10-DQLNet-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 150,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy9 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3359,19 +3614,31 @@
|
||||
" who_won_fraction=0,\n",
|
||||
" final_score_fraction=1,\n",
|
||||
")\n",
|
||||
"ql_policy3.policy_name"
|
||||
"ql_policys.append(ql_policy9)\n",
|
||||
"ql_policy9.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 151,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW10-FSF00-DQLNet-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 151,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy10 = QLPolicy(\n",
|
||||
" 0.92,\n",
|
||||
@@ -3381,12 +3648,13 @@
|
||||
" who_won_fraction=1,\n",
|
||||
" final_score_fraction=0,\n",
|
||||
")\n",
|
||||
"ql_policy4.policy_name"
|
||||
"ql_policys.append(ql_policy10)\n",
|
||||
"ql_policy10.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 152,
|
||||
"metadata": {
|
||||
"collapsed": false,
|
||||
"jupyter": {
|
||||
@@ -3396,7 +3664,18 @@
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW03-FSF03-DQLNet-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 152,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy11 = QLPolicy(\n",
|
||||
" 0.95,\n",
|
||||
@@ -3406,12 +3685,13 @@
|
||||
" who_won_fraction=0.3,\n",
|
||||
" final_score_fraction=0.3,\n",
|
||||
")\n",
|
||||
"ql_policy5.policy_name"
|
||||
"ql_policys.append(ql_policy11)\n",
|
||||
"ql_policy11.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 153,
|
||||
"metadata": {
|
||||
"collapsed": false,
|
||||
"jupyter": {
|
||||
@@ -3421,7 +3701,18 @@
|
||||
"is_executing": true
|
||||
}
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW03-FSF07-DQLNet-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 153,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy12 = QLPolicy(\n",
|
||||
" 0.95,\n",
|
||||
@@ -3431,19 +3722,31 @@
|
||||
" who_won_fraction=0.3,\n",
|
||||
" final_score_fraction=0.65,\n",
|
||||
")\n",
|
||||
"ql_policy6.policy_name"
|
||||
"ql_policys.append(ql_policy12)\n",
|
||||
"ql_policy12.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"execution_count": 154,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"text/plain": [
|
||||
"'QL-M-G09-WW02-FSF07-DQLNet-MSELoss'"
|
||||
]
|
||||
},
|
||||
"execution_count": 154,
|
||||
"metadata": {},
|
||||
"output_type": "execute_result"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"ql_policy13 = QLPolicy(\n",
|
||||
" 0.95,\n",
|
||||
@@ -3453,7 +3756,8 @@
|
||||
" who_won_fraction=0.2,\n",
|
||||
" final_score_fraction=0.65,\n",
|
||||
")\n",
|
||||
"ql_policy6.policy_name"
|
||||
"ql_policys.append(ql_policy13)\n",
|
||||
"ql_policy13.policy_name"
|
||||
]
|
||||
},
|
||||
{
|
||||
@@ -3480,26 +3784,39 @@
|
||||
},
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"ql_policy = ql_policy1\n",
|
||||
"ql_policy.load()\n",
|
||||
"ql_policy.train(15, 10, 1000, 250, [RandomPolicy(0), GreedyPolicy(0)])"
|
||||
"outputs": [
|
||||
{
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "d5b65d806968471e925990e0a73b862c",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"text/plain": [
|
||||
" 0%| | 0/1 [00:00<?, ?epoch/s]"
|
||||
]
|
||||
},
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"metadata": {
|
||||
"pycharm": {
|
||||
"is_executing": true
|
||||
"data": {
|
||||
"application/vnd.jupyter.widget-view+json": {
|
||||
"model_id": "ec4fbeb7458d4ed98cd863dc0eda19ac",
|
||||
"version_major": 2,
|
||||
"version_minor": 0
|
||||
},
|
||||
"tags": []
|
||||
"text/plain": [
|
||||
" 0%| | 0/10 [00:00<?, ?batch/s]"
|
||||
]
|
||||
},
|
||||
"outputs": [],
|
||||
"metadata": {},
|
||||
"output_type": "display_data"
|
||||
}
|
||||
],
|
||||
"source": [
|
||||
"for i in range(100):\n",
|
||||
" for ql_policy in [ql_policy8, ql_policy9, ql_policy10, ql_policy11, ql_policy12, ql_policy13]:\n",
|
||||
" for ql_policy in ql_policys:\n",
|
||||
" ql_policy.load()\n",
|
||||
" ql_policy.train(1, 10, 1000, 250, [RandomPolicy(0), GreedyPolicy(0)])"
|
||||
]
|
||||
|
||||
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Binary file not shown.
Loaded 100 of 575 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user