cetin/Projektpraktikum
1
1
Fork 0
Code Issues Pull Requests repo.project_board Releases Wiki Activity

lasso

Browse Source
This commit is contained in:
cetin 2024-01-24 16:17:45 +01:00
parent c6bb97b942
commit 02d26d974f
3 changed files with 85 additions and 67 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
methods/fit_linear_regression_model.py
View File

@ -1,6 +1,6 @@
import awkward as ak import awkward as ak
from sklearn.preprocessing import PolynomialFeatures from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression from sklearn.linear_model import LinearRegression, Lasso
from sklearn.model_selection import train_test_split from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error from sklearn.metrics import mean_squared_error
import numpy as np import numpy as np
@ -71,7 +71,7 @@ def fit_linear_regression_model(
X_test_model = np.delete(X_test_model, remove, axis=1) X_test_model = np.delete(X_test_model, remove, axis=1)
poly_features = np.delete(poly_features, remove) poly_features = np.delete(poly_features, remove)
lin_reg = LinearRegression(fit_intercept=fit_intercept) lin_reg = LinearRegression(fit_intercept=fit_intercept) # Lasso(alpha=0.01)
lin_reg.fit(X_train_model, y_train) lin_reg.fit(X_train_model, y_train)
y_pred_test = lin_reg.predict(X_test_model) y_pred_test = lin_reg.predict(X_test_model)
print(f"Parameterisation for {target_feat}:") print(f"Parameterisation for {target_feat}:")

102
trackinglosses_endVelo_momEff.ipynb
View File

File diff suppressed because one or more lines are too long

46
trackinglosses_rad_length_endVelo.ipynb
View File

@ -38,12 +38,10 @@
"\n", "\n",
"# selektiere nur elektronen von B->K*ee\n", "# selektiere nur elektronen von B->K*ee\n",
"allcolumns = file.arrays()\n", "allcolumns = file.arrays()\n",
"found = allcolumns[\n", "found = allcolumns[(allcolumns.isElectron) & (~allcolumns.lost) &\n",
" (allcolumns.isElectron) & (~allcolumns.lost) & (allcolumns.fromB)\n", " (allcolumns.fromB)] # B: 9056\n",
"] # B: 9056\n", "lost = allcolumns[(allcolumns.isElectron) & (allcolumns.lost) &\n",
"lost = allcolumns[\n", " (allcolumns.fromB)] # B: 1466\n",
" (allcolumns.isElectron) & (allcolumns.lost) & (allcolumns.fromB)\n",
"] # B: 1466\n",
"\n", "\n",
"electrons = allcolumns[(allcolumns.isElectron) & (allcolumns.fromB)]\n", "electrons = allcolumns[(allcolumns.isElectron) & (allcolumns.fromB)]\n",
"\n", "\n",
@ -225,14 +223,16 @@
" y = found[\"ideal_state_9410_y\"]\n", " y = found[\"ideal_state_9410_y\"]\n",
" qop = found[\"ideal_state_9410_qop\"]\n", " qop = found[\"ideal_state_9410_qop\"]\n",
"\n", "\n",
"data = ak.zip({\n", "data = ak.zip(\n",
" \"rad_length_frac\": rad_length_frac,\n", " {\n",
" \"x\": x,\n", " \"rad_length_frac\": rad_length_frac,\n",
" \"y\": y,\n", " \"x\": x,\n",
" \"tx\": slopex,\n", " \"y\": y,\n",
" \"ty\": slopey,\n", " \"tx\": slopex,\n",
" \"qop\": qop,\n", " \"ty\": slopey,\n",
"})\n", " \"qop\": qop,\n",
" }\n",
")\n",
"lin_reg, features, xx0_test, xx0_predicted = fit_linear_regression_model(\n", "lin_reg, features, xx0_test, xx0_predicted = fit_linear_regression_model(\n",
" data,\n", " data,\n",
" \"rad_length_frac\",\n", " \"rad_length_frac\",\n",
@ -357,16 +357,14 @@
" y = lost[\"ideal_state_9410_y\"]\n", " y = lost[\"ideal_state_9410_y\"]\n",
" qop = lost[\"ideal_state_9410_qop\"]\n", " qop = lost[\"ideal_state_9410_qop\"]\n",
"\n", "\n",
"data = ak.zip(\n", "data = ak.zip({\n",
" {\n", " \"rad_length_frac\": rad_length_frac,\n",
" \"rad_length_frac\": rad_length_frac,\n", " \"x\": x,\n",
" \"x\": x,\n", " \"y\": y,\n",
" \"y\": y,\n", " \"tx\": slopex,\n",
" \"tx\": slopex,\n", " \"ty\": slopey,\n",
" \"ty\": slopey,\n", " \"qop\": qop,\n",
" \"qop\": qop,\n", "})\n",
" }\n",
")\n",
"lin_reg, features, xx0_test, xx0_predicted = fit_linear_regression_model(\n", "lin_reg, features, xx0_test, xx0_predicted = fit_linear_regression_model(\n",
" data,\n", " data,\n",
" \"rad_length_frac\",\n", " \"rad_length_frac\",\n",