outputting all the acinqu... (84d1a063) · Commits · DMT-development / DMT-core

DMT/Hdev/dut_hdev.py

+58 −56

Original line number	Diff line number	Diff line
		@@ -453,6 +453,7 @@ class DutHdev(DutTcad):
		try:
		df_ = h5py.File(os.path.join(sim_folder, "simulation_data.h5")) # hdf5 simulation data
		df_iv = self.get_df(df_, None, "iv")
		df_dc = copy.deepcopy(df_iv)
		try:
		df_cap = self.get_df(df_, None, "cap")
		except:
		@@ -516,7 +517,7 @@ class DutHdev(DutTcad):
		n = 0
		for df_ac in dfs_ac:
		# read in ac df
		df_dc = df_iv.iloc[[n]]
		df_dc_temp = df_iv.iloc[[n]]
		# extend ac df with dc data
		for col_dc in df_iv.columns:
		if col_dc not in df_ac.columns:
		@@ -524,10 +525,10 @@ class DutHdev(DutTcad):

		# extend dc df with ac data
		for col_ac in df_ac.columns:
		if col_ac not in df_dc.columns:
		df_dc.loc[:, col_ac] = 0 # set AC values to zero in DC
		if col_ac not in df_dc_temp.columns:
		df_dc_temp.loc[:, col_ac] = 0 # set AC values to zero in DC

		dfs_temp.append(df_dc)
		dfs_temp.append(df_dc_temp)
		dfs_temp.append(df_ac)

		n = n + 1
		@@ -556,9 +557,9 @@ class DutHdev(DutTcad):
		df_iv.ensure_specifier_column(specifiers.CAPACITANCE + "B" + "E", ports=["B", "C"])
		df_iv.ensure_specifier_column(specifiers.CAPACITANCE + "B" + "C", ports=["B", "C"])
		df_iv.ensure_specifier_column(specifiers.TRANSIT_FREQUENCY, ports=["B", "C"])
		df_iv.ensure_specifier_column(specifiers.TRANSCONDUCTANCE, ports=["B", "C"])
		# df_iv.ensure_specifier_column(specifiers.TRANSCONDUCTANCE, ports=["B", "C"])
		df_iv.ensure_specifier_column(
		specifiers.SS_PARA_Y + "C" + "B" + sub_specifiers.REAL,
		specifiers.SS_PARA_Y + ["C", "B"] + sub_specifiers.REAL,
		ports=["B", "C"],
		)
		except KeyError:
		@@ -604,42 +605,43 @@ class DutHdev(DutTcad):
		df_iv["Q\|P"] = qp

		# where inqu df is required for every op
		gm = np.array(df_iv[specifiers.TRANSCONDUCTANCE])
		tau_e = np.ones(len(df_iv))
		tau_be = np.ones(len(df_iv))
		tau_b = np.ones(len(df_iv))
		tau_c = np.ones(len(df_iv))
		tau_bc = np.ones(len(df_iv))
		x_be = np.ones(len(df_iv))
		x_bc = np.ones(len(df_iv))

		if len(df_dc) == 1:
		# gm from low f Y_CB
		gm = [
		np.real(
		df_iv.at[
		df_iv[specifiers.FREQUENCY].idxmin(),
		specifiers.SS_PARA_Y + ["C", "B"],
		]
		)
		]
		else:
		# gm using gradient
		df_dc.ensure_specifier_column(specifiers.TRANSCONDUCTANCE, ports=["B", "C"])
		gm = np.array(df_dc[specifiers.TRANSCONDUCTANCE])
		# tau_e = np.ones(len(df_dc))
		# tau_be = np.ones(len(df_dc))
		# tau_b = np.ones(len(df_dc))
		# tau_c = np.ones(len(df_dc))
		# tau_bc = np.ones(len(df_dc))
		# x_be = np.ones(len(df_dc))
		# x_bc = np.ones(len(df_dc))

		try:
		for i_row, row in enumerate(df_iv.iterrows()):
		key_inqu = self.join_key(key, f"op{i_row+1}_inqu")
		# next(
		# _key
		# for _key in self.data.keys()
		# if key in _key and "_inqu" in _key and "op" + str(i_row + 1) in _key
		# )
		key_ac_inqu = self.join_key(key, f"acinqu_op{i_row+1}_dVb")
		# key_ac_inqu = next(
		# _key
		# for _key in self.data.keys()
		# if key in _key
		# and "acinqu" in _key
		# and "op" + str(i_row + 1) in _key
		# and "dVb" in _key
		# )
		for i_row, row in df_dc.iterrows():
		key_inqu = self.join_key(key, f"op{i_row+1:.0f}_inqu")
		key_ac_inqu = self.join_key(key, f"acinqu_op{i_row+1:.0f}_dVb_f1")

		# get the necessary data
		df_ac_inqu_i = self.data[key_ac_inqu]
		dn_dic = df_ac_inqu_i["re_d_n_x"].to_numpy() / gm[i_row]
		dn_dic = df_ac_inqu_i["re_d_n_x"].to_numpy() / gm[int(i_row)]
		try:
		dn2_dic = df_ac_inqu_i["re_d_n2_x"].to_numpy() / gm[i_row]
		dn2_dic = df_ac_inqu_i["re_d_n2_x"].to_numpy() / gm[int(i_row)]
		except KeyError:
		dn2_dic = df_ac_inqu_i["re_d_n_x"].to_numpy() / gm[i_row]
		dn2_dic = df_ac_inqu_i["re_d_n_x"].to_numpy() / gm[int(i_row)]

		dp_dic = df_ac_inqu_i["re_d_p_x"].to_numpy() / gm[i_row]
		dp_dic = df_ac_inqu_i["re_d_p_x"].to_numpy() / gm[int(i_row)]
		droh_dic = dp_dic - dn_dic
		# transit time
		changes = (
		@@ -647,20 +649,20 @@ class DutHdev(DutTcad):
		)
		index_be = changes[np.argmin(np.abs(changes - junctions[0]))]
		index_bc = changes[np.argmin(np.abs(changes - junctions[1]))]
		x_be[i_row] = x[index_be]
		x_bc[i_row] = x[index_bc]
		x_be = x[index_be]
		x_bc = x[index_bc]

		tau = np.zeros(len(x))
		taup = np.zeros(len(x))
		taun = np.zeros(len(x))
		taun2 = np.zeros(len(x))
		for j, x_ in enumerate(x):
		if x_ <= x_be[i_row]:
		if x_ <= x_be:
		tau[j] += np.trapezoid(dp_dic[:j], x[:j])
		tau[j] += np.trapezoid(dn_dic[:j], x[:j]) - np.trapezoid(
		dp_dic[:j], x[:j]
		)
		elif x_ <= x_bc[i_row]:
		elif x_ <= x_bc:
		tau[j] += np.trapezoid(dn_dic[:j], x[:j])
		else:
		tau[j] += np.trapezoid(dp_dic[:j], x[:j])
		@@ -677,23 +679,19 @@ class DutHdev(DutTcad):
		taun2 = taun2 * constants.P_Q
		taup = taup * constants.P_Q

		tau_e[i_row] = (
		np.trapezoid(dp_dic[:index_be], x[:index_be]) * constants.P_Q
		)
		tau_be[i_row] = (
		tau_e = np.trapezoid(dp_dic[:index_be], x[:index_be]) * constants.P_Q
		tau_be = (
		np.trapezoid(dn_dic[:index_be], x[:index_be]) * constants.P_Q
		- tau_e[i_row]
		- tau_e
		)
		tau_b[i_row] = (
		tau_b = (
		np.trapezoid(dn_dic[index_be:index_bc], x[index_be:index_bc])
		* constants.P_Q
		)
		tau_c[i_row] = (
		np.trapezoid(dp_dic[index_bc:], x[index_bc:]) * constants.P_Q
		)
		tau_bc[i_row] = (
		tau_c = np.trapezoid(dp_dic[index_bc:], x[index_bc:]) * constants.P_Q
		tau_bc = (
		np.trapezoid(dn_dic[index_bc:], x[index_bc:]) * constants.P_Q
		- tau_c[i_row]
		- tau_c
		)

		dm_dic = np.where(dn_dic < dp_dic, dn_dic, dn_dic)
		@@ -706,14 +704,18 @@ class DutHdev(DutTcad):
		self.data[key_inqu]["TAUN"] = taun
		self.data[key_inqu]["TAUN2"] = taun2

		df_iv["tau_e"] = tau_e
		df_iv["tau_be"] = tau_be
		df_iv["tau_b"] = tau_b
		df_iv["tau_bc"] = tau_bc
		df_iv["tau_c"] = tau_c
		indexes_op = (
		df_iv[specifiers.CURRENT + "C"] == row[specifiers.CURRENT + "C"]
		)

		df_iv.loc[indexes_op, "tau_e"] = tau_e
		df_iv.loc[indexes_op, "tau_be"] = tau_be
		df_iv.loc[indexes_op, "tau_b"] = tau_b
		df_iv.loc[indexes_op, "tau_bc"] = tau_bc
		df_iv.loc[indexes_op, "tau_c"] = tau_c

		df_iv["x_be"] = x_be
		df_iv["x_bc"] = x_bc
		df_iv.loc[indexes_op, "x_be"] = x_be
		df_iv.loc[indexes_op, "x_bc"] = x_bc

		except Exception as err:
		print(err)