residual paper fixes from the paper review (#304)

* ensure the mask_visible is set to what was used to produce the samples in 2023
jpata · Mar 29, 2024 · cd2ed41 · cd2ed41
1 parent 4d35e51
commit cd2ed41
Show file tree

Hide file tree

Showing 6 changed files with 177 additions and 195 deletions.
diff --git a/mlpf/plotting/plot_utils.py b/mlpf/plotting/plot_utils.py
@@ -980,7 +980,6 @@ def plot_particles(yvals, epoch=None, cp_dir=None, comet_experiment=None, title=
 
 
 def plot_jet_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=None, title=None):
-
     pf_genjet_pt = yvals["jet_gen_to_cand_genpt"]
     mlpf_genjet_pt = yvals["jet_gen_to_pred_genpt"]
 
@@ -1038,7 +1037,6 @@ def plot_jet_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=No
         plt.title(labels["gen_jet_range"].format(lim_low, lim_hi))
         plt.yscale("log")
 
-    plt.tight_layout()
     save_img(
         "jet_response_binned.png",
         epoch,
@@ -1051,22 +1049,12 @@ def plot_jet_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=No
     mlpf_vals = np.array(mlpf_vals)
 
     # Plot median and IQR as a function of gen pt
-    fig, axs = plt.subplots(2, 1, sharex=True)
-    plt.sca(axs[0])
-    plt.plot(x_vals, pf_vals[:, 1], marker="o", label="PF")
-    plt.plot(x_vals, mlpf_vals[:, 1], marker="o", label="MLPF")
-    plt.ylim(0.75, 1.25)
-    plt.axhline(1.0, color="black", ls="--")
-    plt.ylabel("Response median")
-    plt.legend(title=title)
-
-    plt.sca(axs[1])
-    plt.plot(x_vals, pf_vals[:, 2] - pf_vals[:, 0], marker="o", label="PF")
-    plt.plot(x_vals, mlpf_vals[:, 2] - mlpf_vals[:, 0], marker="o", label="MLPF")
-    plt.ylabel("Response IQR")
+    plt.figure()
+    plt.plot(x_vals, (pf_vals[:, 2] - pf_vals[:, 0]) / pf_vals[:, 1], marker="o", label="PF")
+    plt.plot(x_vals, (mlpf_vals[:, 2] - mlpf_vals[:, 0]) / mlpf_vals[:, 1], marker="o", label="MLPF")
+    plt.legend(loc=1, title=title, fontsize=30)
+    plt.ylabel("Response IQR / median")
     plt.xlabel(labels["gen_jet"])
-
-    plt.tight_layout()
     save_img(
         "jet_response_med_iqr.png",
         epoch,
@@ -1076,7 +1064,6 @@ def plot_jet_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=No
 
 
 def plot_met_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=None, title=None):
-
     genmet = yvals["gen_met"]
 
     pf_response = yvals["ratio_cand"]
@@ -1131,7 +1118,6 @@ def plot_met_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=No
         plt.title(labels["gen_met_range"].format(lim_low, lim_hi))
         plt.yscale("log")
 
-    plt.tight_layout()
     save_img(
         "met_response_binned.png",
         epoch,
@@ -1144,27 +1130,12 @@ def plot_met_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=No
     mlpf_vals = np.array(mlpf_vals)
 
     # Plot median and IQR as a function of gen pt
-    fig, axs = plt.subplots(2, 1, sharex=True)
-    plt.sca(axs[0])
-    plt.plot(x_vals, pf_vals[:, 1], marker="o", label="PF")
-    plt.plot(x_vals, mlpf_vals[:, 1], marker="o", label="MLPF")
-    plt.ylim(0.75, 1.25)
-    plt.axhline(1.0, color="black", ls="--")
-    plt.ylabel("Response median")
-    if title:
-        plt.title(title)
-    plt.legend()
-
-    plt.sca(axs[1])
-    plt.plot(x_vals, pf_vals[:, 2] - pf_vals[:, 0], marker="o", label="PF")
-    plt.plot(x_vals, mlpf_vals[:, 2] - mlpf_vals[:, 0], marker="o", label="MLPF")
-    plt.ylabel("Response IQR")
-    plt.legend()
-    if title:
-        plt.title(title)
+    plt.figure()
+    plt.plot(x_vals, (pf_vals[:, 2] - pf_vals[:, 0]) / pf_vals[:, 1], marker="o", label="PF")
+    plt.plot(x_vals, (mlpf_vals[:, 2] - mlpf_vals[:, 0]) / mlpf_vals[:, 1], marker="o", label="MLPF")
+    plt.ylabel("Response IQR / median")
+    plt.legend(loc=1, title=title, fontsize=30)
     plt.xlabel(labels["gen_met"])
-
-    plt.tight_layout()
     save_img(
         "met_response_med_iqr.png",
         epoch,
@@ -1174,7 +1145,6 @@ def plot_met_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=No
 
 
 def plot_3dmomentum_response_binned(yvals, epoch=None, cp_dir=None, comet_experiment=None, title=None):
-
     genmet = yvals["gen_mom"]
 
     pf_response = yvals["ratio_cand"]
@@ -1229,7 +1199,6 @@ def plot_3dmomentum_response_binned(yvals, epoch=None, cp_dir=None, comet_experi
         plt.title(labels["gen_mom_range"].format(lim_low, lim_hi))
         plt.yscale("log")
 
-    plt.tight_layout()
     save_img(
         "mom_response_binned.png",
         epoch,
@@ -1242,22 +1211,12 @@ def plot_3dmomentum_response_binned(yvals, epoch=None, cp_dir=None, comet_experi
     mlpf_vals = np.array(mlpf_vals)
 
     # Plot median and IQR as a function of gen pt
-    fig, axs = plt.subplots(2, 1, sharex=True)
-    plt.sca(axs[0])
-    plt.plot(x_vals, pf_vals[:, 1], marker="o", label="PF")
-    plt.plot(x_vals, mlpf_vals[:, 1], marker="o", label="MLPF")
-    plt.ylim(0.75, 1.25)
-    plt.axhline(1.0, color="black", ls="--")
-    plt.ylabel("Response median")
-    plt.legend(title=title)
-
-    plt.sca(axs[1])
-    plt.plot(x_vals, pf_vals[:, 2] - pf_vals[:, 0], marker="o", label="PF")
-    plt.plot(x_vals, mlpf_vals[:, 2] - mlpf_vals[:, 0], marker="o", label="MLPF")
-    plt.ylabel("Response IQR")
+    plt.figure()
+    plt.plot(x_vals, (pf_vals[:, 2] - pf_vals[:, 0]) / pf_vals[:, 1], marker="o", label="PF")
+    plt.plot(x_vals, (mlpf_vals[:, 2] - mlpf_vals[:, 0]) / mlpf_vals[:, 1], marker="o", label="MLPF")
+    plt.ylabel("Response IQR / median")
     plt.xlabel(labels["gen_mom"])
-
-    plt.tight_layout()
+    plt.legend(loc=1, fontsize=30, title=title)
     save_img(
         "mom_response_med_iqr.png",
         epoch,

diff --git a/notebooks/clic/paper-2023-overlapping-lepton-jet.ipynb b/notebooks/clic/paper-2023-overlapping-lepton-jet.ipynb
@@ -246,7 +246,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
+   "version": "3.10.13"
   }
  },
  "nbformat": 4,

diff --git a/notebooks/clic/paper-2023-single-particle.ipynb b/notebooks/clic/paper-2023-single-particle.ipynb
@@ -179,7 +179,7 @@
     "    plt.savefig(\"{}/res_icls{}_ivar{}.pdf\".format(outpath, pid, ivar), bbox_inches=\"tight\")\n",
     "\n",
     "\n",
-    "def plot_eff_and_fake_rate(icls=1, ivar=4, ielem=1, bins=np.linspace(-3, 6, 100), xlabel=\"PFElement log[E/GeV]\", log=True, ylim_fake=1.5, legtitle=\"\"):\n",
+    "def plot_eff_and_fake_rate(icls=1, ivar=4, ielem=1, bins=np.linspace(-3, 6, 100), xlabel=\"PFElement log[E/GeV]\", log=True, ylim_eff=(0.0, 1.5), ylim_fake=(0.0, 1.5), legtitle=\"\"):\n",
     "\n",
     "    #choose only events with a cluster\n",
     "    #this makes the definition of the single particle reconstruction efficiency straightforward\n",
@@ -266,7 +266,7 @@
     "    ax = plt.axes(sharex=ax)\n",
     "    mplhep.histplot(eff_pf, label=\"PF\")\n",
     "    mplhep.histplot(eff_mlpf, label=\"MLPF\")\n",
-    "    plt.ylim(0, 1.5)\n",
+    "    plt.ylim(ylim_eff[0], ylim_eff[1])\n",
     "    plt.ylabel(\"Efficiency\")\n",
     "    plt.xlabel(xlabel)\n",
     "    #cms_label(ax)\n",
@@ -281,7 +281,7 @@
     "    ax = plt.axes(sharex=ax)\n",
     "    mplhep.histplot(fake_pf, label=\"PF\")\n",
     "    mplhep.histplot(fake_mlpf, label=\"MLPF\")\n",
-    "    plt.ylim(0, ylim_fake)\n",
+    "    plt.ylim(ylim_fake[0], ylim_fake[1])\n",
     "    plt.ylabel(\"Fake rate\")\n",
     "    plt.xlabel(xlabel)\n",
     "    #cms_label(ax)\n",
@@ -308,7 +308,17 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "!rm *.pdf"
+    "!ls *.pdf"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a09d8c1b-f55b-4c62-9bd2-e64de7124897",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "nfiles = -1"
    ]
   },
   {
@@ -318,7 +328,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "yvals, X, _ = load_eval_data(\"../../experiments/mlpf-clic-2023-results/clusters_best_tuned_gnn_clic_v130/evaluation/epoch_96/clic_edm_single_pi_pf/*/*.parquet\", -1)"
+    "yvals, X, _ = load_eval_data(\"../../mlpf-clic-2023-results/clusters_best_tuned_gnn_clic_v130/evaluation/epoch_96/clic_edm_single_pi_pf/*/*.parquet\", nfiles)"
    ]
   },
   {
@@ -328,7 +338,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "plot_eff_and_fake_rate(icls=1, ivar=1, ielem=1, bins=np.logspace(0, 2, 41), xlabel=\"track $p_T$ [GeV]\", log=True, ylim_fake=0.1, legtitle=\"charged hadrons\")"
+    "plot_eff_and_fake_rate(icls=1, ivar=1, ielem=1, bins=np.logspace(0, 2, 21), xlabel=\"track $p_T$ [GeV]\", log=True, ylim_eff=(0.6, 1.1), ylim_fake=(0, 0.01), legtitle=\"charged hadrons\")"
    ]
   },
   {
@@ -349,7 +359,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "yvals, X, _ = load_eval_data(\"../../experiments/mlpf-clic-2023-results/clusters_best_tuned_gnn_clic_v130/evaluation/epoch_96/clic_edm_single_gamma_pf/*/*.parquet\", -1)"
+    "yvals, X, _ = load_eval_data(\"../../mlpf-clic-2023-results/clusters_best_tuned_gnn_clic_v130/evaluation/epoch_96/clic_edm_single_gamma_pf/*/*.parquet\", nfiles)"
    ]
   },
   {
@@ -359,7 +369,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "plot_eff_and_fake_rate(icls=3, ivar=5, ielem=2, bins=np.logspace(0, 2, 41), xlabel=\"cluster $E$ [GeV]\", log=True, ylim_fake=0.1, legtitle=\"$\\gamma$\")"
+    "plot_eff_and_fake_rate(icls=3, ivar=5, ielem=2, bins=np.logspace(0, 2, 21), xlabel=\"cluster $E$ [GeV]\", log=True, ylim_eff=(0.8, 1.1), ylim_fake=(0, 0.02), legtitle=\"$\\gamma$\")"
    ]
   },
   {
@@ -380,7 +390,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "yvals, X, _ = load_eval_data(\"../../experiments/mlpf-clic-2023-results/clusters_best_tuned_gnn_clic_v130/evaluation/epoch_96/clic_edm_single_kaon0l_pf/*/*.parquet\", -1)"
+    "yvals, X, _ = load_eval_data(\"../../mlpf-clic-2023-results/clusters_best_tuned_gnn_clic_v130/evaluation/epoch_96/clic_edm_single_kaon0l_pf/*/*.parquet\", nfiles)"
    ]
   },
   {
@@ -390,7 +400,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "plot_eff_and_fake_rate(icls=2, ivar=5, ielem=2, bins=np.logspace(0, 2, 41), xlabel=\"cluster $E$ [GeV]\", log=True, ylim_fake=0.1, legtitle=\"neutral hadrons\")"
+    "plot_eff_and_fake_rate(icls=2, ivar=5, ielem=2, bins=np.logspace(0, 2, 21), xlabel=\"cluster $E$ [GeV]\", log=True, ylim_eff=(0.5, 1.1), ylim_fake=(0, 0.05), legtitle=\"neutral hadrons\")"
    ]
   },
   {
@@ -421,7 +431,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.12"
+   "version": "3.10.13"
   }
  },
  "nbformat": 4,