sum_vals.append(vals) fig = plt.figure() ax = plt.subplot(111) adjust_spines(ax, ['bottom', 'left']) xvals = [] xlbls = [] offset = 0 divider = 2 for i in range(len(groups)): base_ind = i * divider print base_ind dt = np.array(csv_vals[base_ind: base_ind + divider]).T offsets = np.arange(dt.shape[1]) + offset mean_adjust = (i != 1) do_point_line_plot(dt, offsets, mean_adjust=mean_adjust) # for j in range(dt.shape[1]): # p_offset = -0.2 # if dt[j].sum() > 0: # for ind in range(j + 1, dt.shape[1]): # if dt[ind].sum() == 0: # continue # stat, p = scipy.stats.ttest_ind(dt[j], dt[ind]) # print 'p', p, j, ind # if p < 0.05: # plt.scatter(offset + j + p_offset, # 0.95, c=colors[ind], # edgecolor=colors[ind], # marker='*') # p_offset *= -1
fig8 = plt.figure(figsize=(4.5, 3.5)) fig8.set_facecolor('white') plt.hold(True) offset = 1 xvals = [] xlbls = [] ax = plt.subplot(111) adjust_spines(ax, ['left', 'bottom']) for exp_type in exp_types: dt = [] offsets = offset + np.arange(len(stim_types)) plt.text(offsets.mean(), 1.02, exp_type, ha='center') for i, k in enumerate(stim_types): dt.append(cell_max_time[exp_type][k][:, 1]) dt = np.array(dt).T do_point_line_plot(dt, offsets, mean_adjust=True, width=0.7) xvals += offsets.tolist() xlbls += stim_types_short offset = offsets.max() + 2 print xvals plt.ylabel('Corr of Cells to Prediction') #plt.ylim(-0.01, 1) plt.xticks(xvals, xlbls, rotation='vertical') plt.subplots_adjust(left=0.12, bottom=0.2, right=0.97, top=0.95, wspace=0.23, hspace=0.23) fig_path = startup.fig_path + 'Sparseness/summary/' fig8.savefig(fig_path + '%s_pred_summary.eps' % (str(filt)))
fig8 = plt.figure(figsize=(4.5, 3.5)) fig8.set_facecolor('white') plt.hold(True) offset = 1 xvals = [] xlbls = [] ax = plt.subplot(111) adjust_spines(ax, ['left', 'bottom']) for exp_type in exp_types: dt = [] offsets = offset + np.arange(len(stim_types)) plt.text(offsets.mean(), 1.02, exp_type, ha='center') for i, k in enumerate(stim_types): dt.append(cell_max_time[exp_type][k][:, 1]) dt = np.array(dt).T do_point_line_plot(dt, offsets, mean_adjust=True, width=0.7) xvals += offsets.tolist() xlbls += stim_types_short offset = offsets.max() + 2 print xvals plt.ylabel('Corr of Cells to Prediction') #plt.ylim(-0.01, 1) plt.xticks(xvals, xlbls, rotation='vertical') plt.subplots_adjust(left=0.12, bottom=0.2, right=0.97, top=0.95, wspace=0.23, hspace=0.23) fig_path = startup.fig_path + 'Sparseness/summary/' fig8.savefig(fig_path + '%s_pred_summary.eps' % (str(filt))) fig8.savefig(fig_path + '%s_pred_summary.png' % (str(filt))) #plt.show() plt.close(fig8)
sum_vals.append(vals) fig = plt.figure() ax = plt.subplot(111) adjust_spines(ax, ['bottom', 'left']) xvals = [] xlbls = [] offset = 0 divider = 2 for i in range(len(groups)): base_ind = i * divider print base_ind dt = np.array(csv_vals[base_ind:base_ind + divider]).T offsets = np.arange(dt.shape[1]) + offset mean_adjust = (i != 1) do_point_line_plot(dt, offsets, mean_adjust=mean_adjust) # for j in range(dt.shape[1]): # p_offset = -0.2 # if dt[j].sum() > 0: # for ind in range(j + 1, dt.shape[1]): # if dt[ind].sum() == 0: # continue # stat, p = scipy.stats.ttest_ind(dt[j], dt[ind]) # print 'p', p, j, ind # if p < 0.05: # plt.scatter(offset + j + p_offset, # 0.95, c=colors[ind], # edgecolor=colors[ind], # marker='*') # p_offset *= -1
adjust_spines(ax, ['bottom', 'left']) xvals = [] xlbls = [] offset = 1 if exp_type == 'PYR': divider = 2 else: divider = 3 for i in range(len(groups)): base_ind = i * divider mean_adjust = (i != 1) dt = csv_vals[:, base_ind: base_ind + divider] offsets = np.arange(dt.shape[1]) + offset plt.text(offsets.mean(), 1, groups[i], ha='center') do_point_line_plot(dt, offsets, width=0.7, mean_adjust=mean_adjust, alpha=0.5, c=colors) # do_spot_scatter_plot(csv_vals[:, i], offset, col, # width=0.7, mean_adjust=mean_adjust) xvals += offsets.tolist() xlbls += headers[1 + base_ind: base_ind + divider + 1] offset += divider + 1 plt.xticks(xvals, xlbls, rotation='vertical') plt.ylim(0, 1.15) plt.subplots_adjust(left=0.05, bottom=0.25, right=0.98, top=0.98, wspace=0.3, hspace=0.34)
xvals = [] xlbls = [] offset = 1 if exp_type == 'PYR': divider = 2 else: divider = 3 for i in range(len(groups)): base_ind = i * divider mean_adjust = (i != 1) dt = csv_vals[:, base_ind:base_ind + divider] offsets = np.arange(dt.shape[1]) + offset plt.text(offsets.mean(), 1, groups[i], ha='center') do_point_line_plot(dt, offsets, width=0.7, mean_adjust=mean_adjust, alpha=0.5, c=colors) # do_spot_scatter_plot(csv_vals[:, i], offset, col, # width=0.7, mean_adjust=mean_adjust) xvals += offsets.tolist() xlbls += headers[1 + base_ind:base_ind + divider + 1] offset += divider + 1 plt.xticks(xvals, xlbls, rotation='vertical') plt.ylim(0, 1.15) plt.subplots_adjust(left=0.05, bottom=0.25,