def plot_data(pickle_name,pause=False,basename=None,roi=None, cycles=None):
#read a pickle file created by simulate and create plots. Basename is directory
#plus optional characters to put in front of plot names
#cycles is (from,to) tuple of the cycle numbers to plot, inclusive, or a single int
with open(pickle_name,'rb') as fp:
data=pickle.load(fp)
plots.plot(data,roi,pause=pause,out_place=basename, cycles=None)
def simulate_and_plot(override_param, do_plots=False, pause=False,
outdir=None,params_file='params',verbose=False, roi=None):
data,ret,param,pickle_name,out_place= simulate(override_param, outdir=outdir,
params_file=params_file, verbose=verbose)
if do_plots:
plots.plot(data, roi, pause=pause, out_place=out_place)
return ret,param
def _save_metric(self, metric, metric_name='loss', y_label='Test Error'):
c_name = metric_name+'-'+self.name
print_metric_mean_and_std(metric, c_name)
if self.do_plot and (self.save_dir is not None or self.show_plot):
plot(metric, None, None, None, None, None,
'', y_label, c_name, self.save_dir, self.show_plot, self.name)
if self.save_dir is not None:
save_nparray(metric, c_name, self.save_dir)
def loop(robot, input_file, comparison_file, plot_name):
output_file = "analytic_outputs/" + plot_name + ".csv"
outputFile = open(output_file, "w")
"""column_headers = np.array(['time', 'lidar_F', 'lidar_R',
'gyro', 'compass_x', 'compass_y'])
np.savetxt(OUTPUT_FILE, column_headers)"""
w = robot.width
l = robot.length
xOffset = robot.S[0] - (l // 2)
yOffset = robot.S[1] - (w // 2)
#pygame.init()
#screen = pygame.display.set_mode((robot.room_width, robot.room_length))
with open(input_file) as csvFile:
csvReader = csv.reader(csvFile, delimiter=',')
inputs = list(csvReader)
time = np.arange(0, 20.005, 0.005)
lidar = np.zeros((len(time), 2))
gyro = np.zeros((len(time), 1))
compass = np.zeros((len(time), 2))
position = np.zeros((len(time), 3))
for i in range(len(time)):
try:
robot.state_update(inputs[i])
except IndexError:
break
data = robot.get_observation()
lidar[i] = data[:2]
gyro[i] = data[2]
compass[i] = np.array(data[3:])
position[i] = robot.S.reshape((3,))
"""# draw
screen.fill((0, 0, 0))
angle = robot.S[2] * 180 / np.pi
surf = pygame.Surface((l, w)).convert_alpha()
surf.fill((0, 128, 255))
x = xOffset + robot.S[0]
y = yOffset + robot.S[1]
blitRotate(screen, surf, (x, y), (l // 2, w // 2), -angle)
pygame.display.update() """
output_matrix = time
output_matrix = np.column_stack((output_matrix, lidar))
output_matrix = np.column_stack((output_matrix, gyro))
output_matrix = np.column_stack((output_matrix, compass))
output_matrix = np.column_stack((output_matrix, position))
np.savetxt(output_file, output_matrix, delimiter=' ', fmt='%.4f')
plot(output_file, comparison_file, plot_name)
def on_epoch_end(self, epoch, logs={}):
if epoch % self.period == 0 and epoch != 0:
[
os.remove(os.path.join(self.path, file))
for file in os.listdir(self.path)
]
used_metrics = [
metric for metric in self.model.history.params['metrics']
if 'val' not in metric
]
for label in used_metrics:
plot(history=self.model.history, path=self.path, label=label)
def main(restraints, x_axis, y_axis, one_graph=False):
df, header = load()
filt = filter_data(df, header, restraints)
xs, ys, hs = extract_plotable_data(filt, header, x_axis, y_axis)
if not (one_graph):
for x, y, h in zip(xs, ys, hs):
plot(x, y, x_axis, y_axis, h, header, one_graph)
else:
plot(xs, ys, x_axis, y_axis, hs, header, one_graph)
return None
def main():
# check if user args are velid.
user_args = check_user_args(arguments, model_list)
# read data.
data = read_data.read_data(user_args[2], user_args[3])
# call to model.
model = model_handler.model_caller(user_args[1])
# fit model to data.
result = fits.fit(data, model[0])
# plot of fitted model.
plots.plot(data, result[0], model[1], user_args[2], show=user_args[0])
# report of fit results.
logs.log(user_args[2], model[1], result, show=user_args[0])
def main():
start = time.time()
pool = Pool()
pool.map(worker, range(len(eats)))
pool.close()
pool.join()
plot(plots, xc, Vc, zc, DEL, eats, nKGB, 'n', DATA, zpoints, fpoints,
sig80)
end = time.time()
print('time taken = {0}'.format(end - start))
def launch(filepath,
k=55,
include_seq=True,
threshold=0.3,
format='pdf',
output='out',
show=False,
pause=1.2,
fix_steps=False,
make_gif=False):
"""
Launch function to create graph representation of assembly, collapse it, remove low covered vertices
:param filepath: str - path to fasta file
:param k: int - kmer size, recommended to be odd
:param include_seq: boolean - whether to display vertex and edge sequences
:param threshold: float - coefficient which determines threshold for coverage for vertex removing
:param format: str - format of output images - pdf, svg, png
:param output: str - file name of output, 'original' and 'collapsed' will be appended to it
:param show: boolean - whether to display fiery slide show
:param pause: float - pause between each slide
:param fix_steps: boolean - whether to create image of graph state at every stage
:return:
"""
# Load file
a = Graph(filepath, k)
# Populate graph and cover edges
a.fragmentate()
a.cover_edges()
a.edge_coverage()
# Collapse graph and remove low covered vertices
a.collapse_filter(threshold,
fix_steps=fix_steps,
show=show,
pause=pause,
format=format,
output=output)
# Compute edge coverage
a.edge_coverage()
a.extract(f'{output}/out')
# Create plot of collapsed graph if it wasn't created
if not fix_steps:
plot(a, f'{output}/picts/collapsed', include_seq, format, show)
# Create animation from obtained images
if make_gif:
animate(f'{output}/picts', format, pause)
def selection_sort(arr, type_plot):
# Run through all data elements
for i in range(len(arr)):
# Find the minimum element
min_idx = i
for j in range(i + 1, len(arr)):
if arr[min_idx] > arr[j]:
min_idx = j
# Swap the found minimum element with
# the first element
arr[i], arr[min_idx] = arr[min_idx], arr[i]
# Created typ with selected plot type
plot(arr=arr, fallow_point=min_idx, type_plot=type_plot, title='Selection Sort')
def bubble_sort(arr, type_plot):
# Traverse through all array elements
for j in range(len(arr)):
# Compare all elements
for i in range(len(arr)):
if i < (len(arr)) - 1:
# Swap if the element found is greater
# than the next element
if arr[i] > arr[i + 1]:
arr[i], arr[i + 1] = arr[i + 1], arr[i]
# Created typ with selected plot type
plot(arr=arr,
fallow_point=i + 1,
type_plot=type_plot,
title='Bubble Sort')
def on_epoch_end(self, epoch, logs={}):
if epoch % self.period == 0 and epoch != 0:
[
os.remove(os.path.join(self.path, file))
for file in os.listdir(self.path)
]
used_metrics = [
metric for metric in self.model.history.params['metrics']
if 'val' not in metric
]
for label in used_metrics:
plot(history=self.model.history, path=self.path, label=label)
evaluate_model(model=self.model,
path=self.path,
dataset=self.validation_data,
split='Validation',
target_names=self.validation_data.classes)
def hold_out(training_data, results):
print 'Total Data : ' + str(len(training_data))
print sum(results)
test_data = training_data[-int(0.2 * len(training_data)):]
training_data = training_data[:-int(0.2 * len(training_data))]
results_training = results[:-int(0.2 * len(results))]
results_test = results[-int(0.2 * len(results)):]
zeros = 0
ones = 0
# unique(training_data,test_data)
print 'Training Items : ' + str(len(training_data))
print 'Test Items : ' + str(len(test_data))
training_scores_logreg = []
testing_scores_logreg = []
training_scores_svm = []
testing_scores_svm = []
training_scores_rf = []
testing_scores_rf = []
training_scores_bnb = []
testing_scores_bnb = []
#training_scores_knn = []
#testing_scores_knn = []
for i in range(1, len(training_data) / 1000):
print 'Iteration : ' + str(i)
print len(testing_scores_logreg)
logreg.fit(training_data[:i * 1000], results_training[:i * 1000])
testing_scores_logreg.append(logreg.score(test_data, results_test))
training_scores_logreg.append(
logreg.score(training_data, results_training))
clf2.fit(training_data[:i * 1000], results_training[:i * 1000])
testing_scores_svm.append(clf2.score(test_data, results_test))
training_scores_svm.append(clf2.score(training_data, results_training))
rf.fit(training_data[:i * 1000], results_training[:i * 1000])
testing_scores_rf.append(rf.score(test_data, results_test))
training_scores_rf.append(rf.score(training_data, results_training))
bnb.fit(training_data[:i * 1000], results_training[:i * 1000])
testing_scores_bnb.append(bnb.score(test_data, results_test))
training_scores_bnb.append(bnb.score(training_data, results_training))
#logreg.fit(training_data[:i*1000],results_training[:i*1000])
#testing_scores_logreg.append(logreg.score(test_data,results_test))
#training_scores_logreg.append(logreg.score(training_data,results_training))
plots.plot(training_scores_logreg, testing_scores_logreg,
training_scores_svm, testing_scores_svm, training_scores_rf,
testing_scores_rf, training_scores_bnb, testing_scores_bnb)
'''
def train(audio_path, plot_matrix=False):
x_data, y_data = get_set(26, 9, audio_path)
x_data = keras.preprocessing.sequence.pad_sequences(x_data, maxlen=100)
x_train, x_test, Y_train, Y_test = train_test_split(x_data,
y_data,
test_size=0.1,
random_state=42)
y_train = keras.utils.to_categorical(Y_train, 16)
y_test = keras.utils.to_categorical(Y_test, 16)
model = getModel((x_train.shape[1], x_train.shape[2]), y_train.shape[1])
history = model.fit(x_train,
y_train,
batch_size=10,
epochs=137,
verbose=1,
validation_data=(x_test, y_test))
# Plot training & validation accuracy values
plt.plot(history.history['acc'])
plt.plot(history.history['val_acc'])
plt.title('Model accuracy')
plt.ylabel('Accuracy')
plt.xlabel('Epoch')
plt.legend(['Train', 'Test'], loc='upper left')
plt.show()
# Plot training & validation loss values
plt.plot(history.history['loss'])
plt.plot(history.history['val_loss'])
plt.title('Model loss')
plt.ylabel('Loss')
plt.xlabel('Epoch')
plt.legend(['Train', 'Test'], loc='upper left')
plt.show()
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
model.save(model_name)
if plot_matrix:
plot(x_test, Y_test, model_name)
def hold_out(training_data,results): print 'Total Data : ' + str(len(training_data)) print sum(results); test_data = training_data[-int(0.2*len(training_data)):] training_data = training_data[:-int(0.2*len(training_data))] results_training = results[:-int(0.2*len(results))] results_test = results[-int(0.2*len(results)):] zeros = 0 ones = 0 # unique(training_data,test_data) print 'Training Items : ' + str(len(training_data)) print 'Test Items : ' + str(len(test_data)) training_scores_logreg = [] testing_scores_logreg = [] training_scores_svm = [] testing_scores_svm = [] training_scores_rf = [] testing_scores_rf = [] training_scores_bnb = [] testing_scores_bnb = [] #training_scores_knn = [] #testing_scores_knn = [] for i in range(1,len(training_data)/1000): print 'Iteration : ' + str(i) print len(testing_scores_logreg) logreg.fit(training_data[:i*1000],results_training[:i*1000]) testing_scores_logreg.append(logreg.score(test_data,results_test)) training_scores_logreg.append(logreg.score(training_data,results_training)) clf2.fit(training_data[:i*1000],results_training[:i*1000]) testing_scores_svm.append(clf2.score(test_data,results_test)) training_scores_svm.append(clf2.score(training_data,results_training)) rf.fit(training_data[:i*1000],results_training[:i*1000]) testing_scores_rf.append(rf.score(test_data,results_test)) training_scores_rf.append(rf.score(training_data,results_training)) bnb.fit(training_data[:i*1000],results_training[:i*1000]) testing_scores_bnb.append(bnb.score(test_data,results_test)) training_scores_bnb.append(bnb.score(training_data,results_training)) #logreg.fit(training_data[:i*1000],results_training[:i*1000]) #testing_scores_logreg.append(logreg.score(test_data,results_test)) #training_scores_logreg.append(logreg.score(training_data,results_training)) plots.plot(training_scores_logreg,testing_scores_logreg,training_scores_svm,testing_scores_svm ,training_scores_rf ,testing_scores_rf ,training_scores_bnb , testing_scores_bnb) '''
def parflow():
INPUT_DIR = os.path.join(os.path.dirname(__file__), '..', 'parflow',
'inputs')
FORCING_DIR = os.path.join(os.path.dirname(__file__), '..', 'parflow',
'forcing')
OUTPUT_DIR = os.path.join(os.getenv('HOME'), 'RESULTS/output')
PNG_PATH = os.path.join(os.getenv('HOME'), 'RESULTS/plot.png')
# copy input files directly in the base output folder,
# because that's where Parflow expects them to be.
shutil.copytree(INPUT_DIR, OUTPUT_DIR, dirs_exist_ok=True)
run.Solver.CLM.CLMFileDir = OUTPUT_DIR
run.Solver.CLM.MetFilePath = FORCING_DIR
run.run(working_directory=OUTPUT_DIR)
plot(OUTPUT_DIR, run.get_name(), PNG_PATH, title=title)
img = open(PNG_PATH, 'rb').read()
return render_template('result.html',
img=base64.encodebytes(img).decode('ascii'))
def partition(arr, left, right, type_plot):
i = (left - 1) # smaller element
pivot = arr[right] # pivot
for j in range(left, right):
# Swap if current element is smaller than or
# equal to pivot
if arr[j] <= pivot:
i = i + 1
arr[i], arr[j] = arr[j], arr[i]
# Create plot with arr and pivot
plot(arr=arr,
fallow_point=pivot,
type_plot=type_plot,
title='Quick Sort ')
arr[i + 1], arr[right] = arr[right], arr[i + 1]
return i + 1
def plot_to_file(self):
file = filenames['basic_plot'] % curr
data = self.data['input']['processed']
args = {}
args['filename'] = file
args['data'] = data
args['y_label'] = curr + "/PLN"
args['x_label'] = 'czas [data]'
return plot(args)
def plot_to_file_log_return(self):
file = filenames['basic_log_return_plot'] % curr
data = self.data['input']['log_returns']
args = {}
args['filename'] = file
args['data'] = data
args['y_label'] = "logarytmiczna stopa zwrotu, średnia %d dni" % (
2 * moving_average_delta + 1)
args['x_label'] = 'czas [data]'
return plot(args)
def plot_data_with_equation(self, file_name, equation):
fitted_signal = self.fourier_analyse_inverse(equation)
args = {}
args['data'] = [
self.data['input']['processed'][0], # time
self.data['input']['processed'][1], # real values
fitted_signal['signal_abs'] # fitted values
]
args['filename'] = file_name
return plot(args)
def simulationAnnealing(NT, snapNT, spins, J, N, M, kB, gamma, T0, Tsteps):
#initialization for first simulation
spins = spinsInit(N, M)
#keeping track of magnetization
globals.magnetizationInit()
globals.magnetizationHistoryInit()
globals.magnetization = spins.sum()
#calculating the decreas factor of temperature
decreasFactor = (0.05 / T0)**(1 / Tsteps)
#Array of slowly decreasing temperatures
iterationArray = np.zeros(Tsteps)
for i in range(len(iterationArray)):
iterationArray[i] = T0 * decreasFactor**i
#Annealing
for t in iterationArray:
beta = 1 / t / kB
simulation(NT, snapNT, spins, J, N, M, beta, gamma)
plots.plot(spins, gamma, 1, 1)
def heapify(arr, n, i, type_plot):
largest = i # Initialize largest as root
l = 2 * i + 1 # left = 2*i + 1
r = 2 * i + 2 # right = 2*i + 2
# See if left child of root exists and is
# greater than root
if l < n and arr[i] < arr[l]:
largest = l
# See if right child of root exists and is
# greater than root
if r < n and arr[largest] < arr[r]:
largest = r
# Change root, if needed
if largest != i:
arr[i], arr[largest] = arr[largest], arr[i] # swap
# Heapify the root.
heapify(arr, n, largest, type_plot)
plot(arr=arr, fallow_point=largest, type_plot=type_plot, title='Heap Sort')
def plot_freq_basic_analysis(self):
omegas = self.freq_domain['omegas']
dft_abs = self.freq_domain['dft_abs']
points = self.find_valuable_maxims()
file = filenames['basic_fourier_analysis_freq'] % curr
args = {}
args['filename'] = file
args['data'] = [omegas, dft_abs]
args['highlight_data'] = points
args['y_label'] = 'amplituda'
args['x_label'] = 'częstość [1/dzień]'
return plot(args)
def plot_returns(self):
arguments = {}
timing = self.data['fourier_fit']['returns'][0]
values1 = self.data['input']['log_returns'][1]
values2 = self.data['fourier_fit']['returns'][1]
arguments['data'] = [timing, values1, values2]
name = filenames['both_log_return_plot'] % curr
arguments['filename'] = name
arguments['y_label'] = "logarytmiczna stopa zwrotu, średnia %d dni" % (
2 * moving_average_delta + 1)
arguments['x_label'] = "czas [data]"
return plot(arguments)
def plot_basic_data_with_dff_fit(self):
self.find_valuable_maxims(True)
equation = self.translate_maxims_to_equation()
fitted_signal = self.fourier_analyse_inverse(equation)
file = filenames['basic_fourier_analysis_fit'] % curr
args = {}
args['data'] = [
self.data['input']['processed'][0], # time
self.data['input']['processed'][1], # real values
fitted_signal['signal_abs'] # fitted values
]
args['filename'] = file
return plot(args)
def main_page():
data_list = [cases[place] for place in current_places]
if request.method == 'POST':
query = request.form.get('place')
if query in cases.keys() and query not in current_places:
current_places.append(query)
function = request.form.get('function')
try:
data_list = [eval(function) for place in current_places]
except:
data_list = [cases[place] for place in current_places]
script, div, js_resources, css_resources = plot(dates, data_list,
current_places)
return render_template(
'index.html',
current_places=current_places,
places=filter(is_country, list(cases.keys())),
plot_script=script,
plot_div=div,
js_resources=js_resources,
css_resources=css_resources,
)
from stimuli import visual_stimulus
from plots import plot
try:
Runcase=int(sys.argv[-1])
except ValueError:
print('Provide aither an integer, the key "all", or a specific key (see run.py)')
Runcase=str(sys.argv[-1])
np.random.seed(1)
Nshow = 8
if (Runcase==19) or (Runcase=='model-doc') or (Runcase=='all'):
model = earlyVis_model(from_file='data/dense-noise.npz')
ps = plot(model=model)
fig = ps.protocol_plot(cell_plot=np.random.choice(np.arange(model.Ncells), Nshow, replace=False))
fig.savefig('docs/figs/response-dense-noise.png')
ps.show()
if (Runcase==18) or (Runcase=='model-doc') or (Runcase=='all'):
model = earlyVis_model(from_file='data/sparse-noise.npz')
ps = plot(model=model)
fig = ps.protocol_plot(cell_plot=np.random.choice(np.arange(model.Ncells), Nshow, replace=False))
fig.savefig('docs/figs/response-sparse-noise.png')
ps.show()
if (Runcase==15) or (Runcase=='model-doc') or (Runcase=='all'):
model = earlyVis_model(from_file='data/static-grating.npz')
ps = plot(model=model)
fig = ps.protocol_plot(cell_plot=np.random.choice(np.arange(model.Ncells), Nshow, replace=False))
ns = 2 # number of sweeps
dE = 0.0001 # V, potential increment. This value has to be small for BI to approximate the circuit properly
t, E = wf.sweep(Eini=Eini, Efin=Efin, dE=dE, sr=sr, ns=ns) # Creates waveform
g0 = Q0/F # mol/cm2, maximum coverage for 1 monolayer
eps = n*FRT*(E-E0)
kf = ks*np.exp(alpha*eps)
kb = ks*np.exp(-(1-alpha)*eps)
# Simulation parameters
nt = np.size(t)
dt = t[1]
Th = np.ones(nt)
#%% Simulation
for j in range(1,nt):
# Backwards implicit:
Th[j] = (Th[j-1] + dt*kb[j-1])/(1 + dt*(kf[j-1]+kb[j-1]))
# Denormalisation
i = -n*F*Ageo*g0*(kb - (kf+kb)*Th)
Q = g0*F*(1-Th)
end = time.time()
print(end-start)
#%% Plot
p.plot(E, i, "$E$ / V", "$i$ / A")
p.plot(E, Q*1e6, "$E$ / V", "$Q$ / $\mu$C cm$^{-2}$")
X = np.linspace(0, Xmax, nX) # Discretisation of distance
eps = (E - E0) * n * FRT # adimensional potential waveform
delta = np.sqrt(D * t[-1]) # cm, diffusion layer thickness
K0 = ks * delta / D # Normalised standard rate constant
#%% Simulation
for k in range(1, nT):
# Boundary condition, Butler-Volmer:
C[k, 0] = (C[k - 1, 1] +
dX * K0 * np.exp(-alpha * eps[k])) / (1 + dX * K0 * (np.exp(
(1 - alpha) * eps[k]) + np.exp(-alpha * eps[k])))
# Solving finite differences:
for j in range(1, nX - 1):
C[k, j] = C[k - 1, j] + lamb * (C[k - 1, j + 1] - 2 * C[k - 1, j] +
C[k - 1, j - 1])
# Denormalising:
i = n * F * Ageo * D * cB * (-C[:, 2] + 4 * C[:, 1] - 3 * C[:, 0]) / (2 * dX *
delta)
cR = C * cB
cO = cB - cR
x = X * delta
end = time.time()
print(end - start)
#%% Plot
p.plot(E, i, "$E$ / V", "$i$ / A")
p.plot2(x, cR[-1, :] * 1e6, x, cO[-1, :] * 1e6, "[R]", "[O]", "x / cm",
"c($t_{end}$,$x$=0) / mM")
epilog='''
''')
parser.add_argument("files", nargs='+',type=str, help="pickle files to plot", default=None)
parser.add_argument('-c', '--cycles', type=parseIntRange, help="cycle or range of cycles to plot, .e.g. 4 or 0-2", default=None)
parser.add_argument('-n', '--nopause', action='store_true', help="do not pause after generating plots")
options = parser.parse_args()
print('git revision:{}'.format(util.get_git_commit()))
class AttrDict(dict):
def __init__(self, *args, **kwargs):
super(AttrDict, self).__init__(*args, **kwargs)
self.__dict__ = self
for file in options.files:
with open(file, 'rb') as fp:
data=pickle.load(fp)
print('loaded {}'.format(file))
print('cyles {}'.format(options.cycles))
head,tail=os.path.split(file)
print('head={} tail={}'.format(head,tail))
#take the part up to the dash in tail, and add something
parts=tail.split('-')
out_place=os.path.join(head,parts[0]+'-replot')
print('out_place={}'.format(out_place))
plots.plot(data, out_place=out_place, pause=not options.nopause,
cycles=options.cycles, text=parts[0])
if (t>=surround_delay) and (t<=surround_delay+surround_duration) and (t>=center_delay) and (t<=center_delay+center_duration):
self.static_center_surround_grating(i+self.it0, **args)
elif (t>=surround_delay) and (t<=surround_delay+surround_duration):
self.static_surround_grating(i+self.it0, **args)
elif (t>=center_delay) and (t<=center_delay+center_duration):
self.static_center_surround_grating(i+self.it0, **args)
if __name__=='__main__':
from plots import plot
stim = visual_stimulus(sys.argv[-1])
stim_plot= plot(stimulus=stim, graph_env_key='visual_stim')
stim_plot.screen_movie(stim)
# if stim.stimulus_params['static']:
# stim_plot.screen_plot(stim.full_array[0,:,:])
# else:
# stim_plot.screen_movie(stim)
stim_plot.show()
def hold_out_training(training_data,results): print 'Length : ' + str(len(training_data)) test_data = training_data[-int(0.3*len(training_data)):] training_data = training_data[:-int(0.3*len(training_data))] results_training = results[:-int(0.3*len(results))] results_test = results[-int(0.3*len(results)):] rf.fit(training_data,results_training) print rf.score(test_data,results_test) ''' pre_shuffle_stuff = [] counter = 0 for item in results_training: item = item.append(results_training[counter]) accuracies = [] #shuffle(training_data) ''' accuracies_svm = [] accuracies_rf = [] accuracies_knn = [] accuracies_svm_rbf = [] for j in range(10,len(training_data),1): correct = 0 wrong = 0 curr_training_data = training_data[:j] curr_results = results_training[:j] rf.fit(curr_training_data,curr_results) rf.score(test_data,results_test) y_true = [] y_pred = [] for i in range(len(test_data)): result = rf.predict(test_data[i]) if result[0] == results_test[i]: correct = correct + 1 #print 'Correct : ' + str(result) else: wrong = wrong + 1 #iprint 'Wrong : ' + str(result) print correct,wrong accuracy = (float(correct))/(float(correct) + float(wrong)) accuracies_rf.append(accuracy) print 'Accuracy : ' + str(accuracy) knn.fit(curr_training_data,curr_results) knn.score(test_data,results_test) correct = 0 wrong = 0 for i in range(len(test_data)): result = knn.predict(test_data[i]) if result[0] == results_test[i]: correct = correct + 1 #print 'Correct : ' + str(result) else: wrong = wrong + 1 #iprint 'Wrong : ' + str(result) print correct,wrong accuracy = (float(correct))/(float(correct) + float(wrong)) accuracies_knn.append(accuracy) print 'Accuracy : ' + str(accuracy) clf2.fit(curr_training_data,curr_results) clf2.score(test_data,results_test) correct = 0 wrong = 0 for i in range(len(test_data)): result = knn.predict(test_data[i]) if result[0] == results_test[i]: correct = correct + 1 #print 'Correct : ' + str(result) else: wrong = wrong + 1 #iprint 'Wrong : ' + str(result) print correct,wrong accuracy = (float(correct))/(float(correct) + float(wrong)) accuracies_svm.append(accuracy) print 'Accuracy : ' + str(accuracy) clf.fit(curr_training_data,curr_results) clf.score(test_data,results_test) correct = 0 wrong = 0 for i in range(len(test_data)): result = knn.predict(test_data[i]) if result[0] == results_test[i]: correct = correct + 1 #print 'Correct : ' + str(result) else: wrong = wrong + 1 #iprint 'Wrong : ' + str(result) print correct,wrong accuracy = (float(correct))/(float(correct) + float(wrong)) accuracies_svm_rbf.append(accuracy) print 'Accuracy : ' + str(accuracy) plots.plot(accuracies_rf,accuracies_knn,accuracies_svm,accuracies_svm_rbf,training_data)
