markdown stringlengths 0 1.02M | code stringlengths 0 832k | output stringlengths 0 1.02M | license stringlengths 3 36 | path stringlengths 6 265 | repo_name stringlengths 6 127 |
|---|---|---|---|---|---|
Now when we list our clusters again, we should see the cluster we created. | !gcloud container clusters list | NAME LOCATION MASTER_VERSION MASTER_IP MACHINE_TYPE NODE_VERSION NUM_NODES STATUS
asl-cluster us-central1-a 1.20.10-gke.1600 34.71.2.241 e2-medium 1.20.10-gke.1600 3 RUNNING
cluster-1 us-central1-a 1.19.14-gke.1900 35.193.84.252 custom-2-4352 1.19.14-gke.1900 2... | Apache-2.0 | notebooks/docker_and_kubernetes/labs/2_intro_k8s.ipynb | ctivanovich/asl-ml-immersion |
Get authentication credentials and deploy and applicationAfter creating your cluster, you need authentication credentials to interact with it. Use `get-credentials` to authenticate the cluster.**Exercise**Use `gcloud container clusters get-credentials` to authenticate the cluster you created. | %%bash
gcloud container clusters get-credentials asl-cluster --zone $ZONE | Fetching cluster endpoint and auth data.
kubeconfig entry generated for asl-cluster.
| Apache-2.0 | notebooks/docker_and_kubernetes/labs/2_intro_k8s.ipynb | ctivanovich/asl-ml-immersion |
You can now deploy a containerized application to the cluster. For this lab, you'll run `hello-app` in your cluster.GKE uses Kubernetes objects to create and manage your cluster's resources. Kubernetes provides the [Deployment](https://kubernetes.io/docs/concepts/workloads/controllers/deployment/) object for deploying ... | %%bash
kubectl create deployment hello-server --image=gcr.io/google-samples/hello-app:1.0 | deployment.apps/hello-server created
| Apache-2.0 | notebooks/docker_and_kubernetes/labs/2_intro_k8s.ipynb | ctivanovich/asl-ml-immersion |
This Kubernetes command creates a Deployment object that represents `hello-server`. To create a Kubernetes Service, which is a Kubernetes resource that lets you expose your application to external traffic, run the `kubectl expose` command. **Exercise**Use the `kubectl expose` to expose the application. In this command,... | %%bash
kubectl expose deployment hello-server --type=LoadBalancer --port 8080 | service/hello-server exposed
| Apache-2.0 | notebooks/docker_and_kubernetes/labs/2_intro_k8s.ipynb | ctivanovich/asl-ml-immersion |
Use the `kubectl get service` command to inspect the `hello-server` Service.**Note**: It might take a minute for an external IP address to be generated. Run the previous command again if the `EXTERNAL-IP` column for `hello-server` status is pending. | !kubectl get service hello-server | NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
hello-server LoadBalancer 10.7.245.50 34.135.234.4 8080:30333/TCP 73s
| Apache-2.0 | notebooks/docker_and_kubernetes/labs/2_intro_k8s.ipynb | ctivanovich/asl-ml-immersion |
You can now view the application from your web browser, open a new tab and enter the following address, replacing `EXTERNAL IP` with the EXTERNAL-IP for `hello-server`:```bashhttp://[EXTERNAL_IP]:8080```You should see a simple page which displays```bashHello, world!Version: 1.0.0Hostname: hello-server-5bfd595c65-7jqkn`... | %%bash
kubectl delete deployment hello-server | deployment.apps "hello-server" deleted
| Apache-2.0 | notebooks/docker_and_kubernetes/labs/2_intro_k8s.ipynb | ctivanovich/asl-ml-immersion |
Advanced Usage Exampes for Seldon Client Istio Gateway Request with token over HTTPS - no SSL verificationTest against a current kubeflow cluster with Dex token authentication. 1. Install kubeflow with Dex authentication | INGRESS_HOST=!kubectl -n istio-system get service istio-ingressgateway -o jsonpath='{.status.loadBalancer.ingress[0].ip}'
ISTIO_GATEWAY=INGRESS_HOST[0]
ISTIO_GATEWAY | _____no_output_____ | Apache-2.0 | notebooks/seldon_client.ipynb | OrthoDex/seldon-core |
Get a token from the Dex gateway. At present as Dex does not support curl password credentials you will need to get it from your browser logged into the cluster. Open up a browser console and run `document.cookie` | TOKEN="eyJhbGciOiJIUzUxMiIsInR5cCI6IkpXVCJ9.eyJleHAiOjE1NjM2MjA0ODYsImlhdCI6MTU2MzUzNDA4NiwiaXNzIjoiMzQuNjUuNzMuMjU1IiwianRpIjoiYjllNDQxOGQtZjNmNC00NTIyLTg5ODEtNDcxOTY0ODNmODg3IiwidWlmIjoiZXlKcGMzTWlPaUpvZEhSd2N6b3ZMek0wTGpZMUxqY3pMakkxTlRvMU5UVTJMMlJsZUNJc0luTjFZaUk2SWtOcFVYZFBSMFUwVG1wbk1GbHBNV3RaYW1jMFRGUlNhVTU2VFhS... | _____no_output_____ | Apache-2.0 | notebooks/seldon_client.ipynb | OrthoDex/seldon-core |
Start Seldon CoreUse the setup notebook to [Install Seldon Core](seldon_core_setup.ipynbInstall-Seldon-Core) with [Istio Ingress](seldon_core_setup.ipynbIstio). Instructions [also online](./seldon_core_setup.html).**Note** When running helm install for this example you will need to set the istio.gateway flag to kubefl... | deployment_name="test1"
namespace="default"
from seldon_core.seldon_client import SeldonClient, SeldonChannelCredentials, SeldonCallCredentials
sc = SeldonClient(deployment_name=deployment_name,namespace=namespace,gateway_endpoint=ISTIO_GATEWAY,debug=True,
channel_credentials=SeldonChannelCredentials(v... | _____no_output_____ | Apache-2.0 | notebooks/seldon_client.ipynb | OrthoDex/seldon-core |
Its not presently possible to use gRPC without getting access to the certificates. We will update this once its clear how to obtain them from a Kubeflow cluser setup. Istio - SSL Endpoint - Client Side Verification - No Authentication 1. First run through the [Istio Secure Gateway SDS example](https://istio.io/docs/t... | # Set to folder where the httpbin certificates are
ISTIO_HTTPBIN_CERT_FOLDER='/home/clive/work/istio/httpbin.example.com' | _____no_output_____ | Apache-2.0 | notebooks/seldon_client.ipynb | OrthoDex/seldon-core |
Start Seldon CoreUse the setup notebook to [Install Seldon Core](seldon_core_setup.ipynbInstall-Seldon-Core) with [Istio Ingress](seldon_core_setup.ipynbIstio). Instructions [also online](./seldon_core_setup.html).**Note** When running ```helm install``` for this example you will need to set the ```istio.gateway``` fl... | deployment_name="mymodel"
namespace="default"
from seldon_core.seldon_client import SeldonClient, SeldonChannelCredentials, SeldonCallCredentials
sc = SeldonClient(deployment_name=deployment_name,namespace=namespace,gateway_endpoint="httpbin.example.com",debug=True,
channel_credentials=SeldonChannelCre... | _____no_output_____ | Apache-2.0 | notebooks/seldon_client.ipynb | OrthoDex/seldon-core |
<div id="image_img" class="header_image_13"> Download, Install and Execute Anaconda Difficu... | from biosignalsnotebooks.__notebook_support__ import css_style_apply
css_style_apply()
%%html
<script>
// AUTORUN ALL CELLS ON NOTEBOOK-LOAD!
require(
['base/js/namespace', 'jquery'],
function(jupyter, $) {
$(jupyter.events).on("kernel_ready.Kernel", function () {
con... | _____no_output_____ | MIT | notebookToHtml/biosignalsnotebooks_html_publish/Categories/Install/prepare_anaconda.ipynb | biosignalsnotebooks/biosignalsnotebooks |
Testing cosmoganApril 19, 2021Borrowing pieces of code from : - https://github.com/pytorch/tutorials/blob/11569e0db3599ac214b03e01956c2971b02c64ce/beginner_source/dcgan_faces_tutorial.py- https://github.com/exalearn/epiCorvid/tree/master/cGAN | import os
import random
import logging
import sys
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.backends.cudnn as cudnn
import torch.optim as optim
import torch.utils.data
from torchsummary import summary
from torch.utils.data import DataLoader, TensorDataset
import torch.distributed as dist... | _____no_output_____ | BSD-3-Clause-LBNL | code/4_basic_3d_GAN/1_main_code/train_3dgan.ipynb | vmos1/cosmogan_pytorch |
Modules | ### Transformation functions for image pixel values
def f_transform(x):
return 2.*x/(x + 4.) - 1.
def f_invtransform(s):
return 4.*(1. + s)/(1. - s)
# Generator Code
class View(nn.Module):
def __init__(self, shape):
super(View, self).__init__()
self.shape = shape
def forward(self, ... | _____no_output_____ | BSD-3-Clause-LBNL | code/4_basic_3d_GAN/1_main_code/train_3dgan.ipynb | vmos1/cosmogan_pytorch |
Train loop | ### Train code ###
def f_train_loop(gan_model,Dset,metrics_df,gdict,fixed_noise):
''' Train epochs '''
## Define new variables from dict
keys=['image_size','start_epoch','epochs','iters','best_chi1','best_chi2','save_dir','device','flip_prob','nz','batch_size','bns']
image_size,start_epoch,epochs,iters,... | _____no_output_____ | BSD-3-Clause-LBNL | code/4_basic_3d_GAN/1_main_code/train_3dgan.ipynb | vmos1/cosmogan_pytorch |
Start | ### Setup modules ###
def f_manual_add_argparse():
''' use only in jpt notebook'''
args=argparse.Namespace()
args.config='config_3dgan_128_cori.yaml'
args.mode='fresh'
args.local_rank=0
args.facility='cori'
args.distributed=False
# args.mode='continue'
return args
def f_parse_... | _____no_output_____ | BSD-3-Clause-LBNL | code/4_basic_3d_GAN/1_main_code/train_3dgan.ipynb | vmos1/cosmogan_pytorch |
Main | #########################
### Main code #######
#########################
if __name__=="__main__":
jpt=False
jpt=True ##(different for jupyter notebook)
t0=time.time()
t0=time.time()
args=f_parse_args() if not jpt else f_manual_add_argparse()
#################################
### Set up gl... | _____no_output_____ | BSD-3-Clause-LBNL | code/4_basic_3d_GAN/1_main_code/train_3dgan.ipynb | vmos1/cosmogan_pytorch |
Debug | # class Generator(nn.Module):
# def __init__(self, gdict):
# super(Generator, self).__init__()
# ## Define new variables from dict
# keys=['ngpu','nz','nc','ngf','kernel_size','stride','g_padding']
# ngpu, nz,nc,ngf,kernel_size,stride,g_padding=list(collections.OrderedDict({key:gdic... | _____no_output_____ | BSD-3-Clause-LBNL | code/4_basic_3d_GAN/1_main_code/train_3dgan.ipynb | vmos1/cosmogan_pytorch |
Exploring Clustering ResultsThe file containing the clustering results is stored in the processed data folder with the suffix clean. The index is set to the first __Product group key__.As a reminder the file is organized in three columns: _Product Group Key_, _Cluster Number_ and the corresponding _Centroid_ of the cl... | import os
import sys
# add the 'src' directory as one where we can import modules
root_dir = os.path.join(os.getcwd(),os.pardir,os.pardir)
import pandas as pd
import math
import numpy as np
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import copy as cp
import seaborn as sns
import statsmo... | (382, 2)
| MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Get clients description | file_name1 = "data_client_bnd_ita.csv"
file_name2 = "data_client_bnd_ita2.csv"
non_unique_features = ["Key_lvl1","Key_lvl2","Key_lvl3","CONCESS PDL","CLIENTE FATT","CONCESS FATT","PTF SPEDIZIONE","TYPE_DISTRIB"]
unique_features = ["Key_lvl4","Key_lvl5"]
client_df1 = pd.read_csv(raw_path+file_name1, sep=';', encodin... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Get Products descriptionIn order to get the product features description, an inner join on the product group key is operated on the cluster result with the products description file.Since the clustering was calculated on the second level group, some columns of the description file must be dropped in order to avoid dup... | file_name1 = "bnd_products_desc.csv"
file_name2 = "bnd_products_desc2.csv"
non_unique_features=["Key","Description","CONFEZIONE",\
"CONFEZIONE (Description)","IMBALLO","STD/PROMO","IMBALLO (Description)","STD/PROMO (Description)",\
"TIPO ARTICOLO","TIPO ARTICOLO (De... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Merge Products and Clients tables | # produit_client_cluster = pd.merge(product_cluster,client_df,how='inner', left_on=["Client"],right_on =["Key_lvl4"] )#.drop(["Key_lvl4"],axis=1)
# clusters = produit_client_cluster["Cluster"]
# centroids = produit_client_cluster["Centroid"]
# produit_client_cluster = produit_client_cluster.drop(["Cluster","Centroid... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Save the final result into a csv file for further exploration | filename = 'bnd_product_cluster_clean.csv'
file_name = "p2_clustering_clean_mois.csv"
product_cluster.to_csv(processed_path+filename,sep=';',encoding='iso8859_2')
| _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Homogeneity TestIn order to detect specific caraterstics for each resulted cluster we perform a statistic test based on Pearsons chi-square score with the hypothesis of a uniform distribution.Features with the pvalues lower than 0.1 are displayed for analysis | def cramer_v(chisq,n,k,r=1):
return math.sqrt(chisq/(n * min(k-1,r-1) )) | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Calculate modalities frequency through clustersAs a first step, all the distrubtions of modalities across features and clusters are calculated and stored in one array structered as follows:One array for each cluster which contains a dictionnary of features. Each feature is again a dictionary of modalities and their oc... | #get the features
features = all_features
#get the clusters (actually its a range(1,nb_cluster))
clusters = set(product_cluster['Cluster'].values)
#array to store each cluster and freq for all the features
clusters_feature_dist = [0] #to shift the indices to clusters
#loop trhough features
for c in clusters:
fe... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Chi-square test over clusters | from scipy.stats import chisquare
pthreashold = 0.2
#get the features
features = all_features
clusters = [6]
res_features_over_cluster = [0]
for c in clusters:
#align each feature with its distrubtion in this cluster c
cluster_feature_dist = clusters_feature_dist[c]
dist = [len(x) for x in list(cluster... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Calculate modalities frequency through features | #get the features
features = all_features
#get the clusters (actually its a range(1,nb_cluster))
clusters = set(product_cluster['Cluster'].values)
#dict to store each feater and freq for all the clusters
features_clust_dist = dict()
#invert the dict and get it by feature
for f in features:
freq = dict()
for... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Chi-square test over features | pthreashold = 0.2
clusters = set(product_cluster['Cluster'].values)
features = all_features
features = ["FAM MARKETING (Description)"]
for f in features:
for c in clusters:
#get information from the previous array
feature_clust_dist = features_clust_dist[f]
feature_distribution = list(feat... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Modalities distribution | clusters = set(product_cluster['Cluster'].values)
nclusters = len(clusters)
#get the features
features = product_cluster.columns[0:-2]
features = features.drop(unbalanced)
modalities_clust_dist = dict()
for f in features:
feature_sum=[]
modalities = set(product_cluster[f].values)
modalities_distribution=... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Chi-square test for modalities over clusters | %matplotlib inline
clusters = set(product_cluster['Cluster'].values)
nclusters = len(clusters)
pthreashold = 0.2
n_min_dist = 1
min_members = 5
min_dust = True
for f in features:
modalities = set(product_cluster[f].values)
r = len(modalities)
for m in modalities:
modality_dist = modalities_cl... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
MCA Analysis Remove unbalanced columns | features_df = product_cluster.iloc[:,3:-2]
plt.figure(figsize=(16,20))
features = features_df.columns
for i,f in enumerate(features):
counts = features_df.groupby([f])[f].count().to_dict()
dist = list(counts.values())
keys = list(counts.keys())
chisq, p = chisquare(dist)
plt.subplot(6,3,i+1)
plt... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Apply MCA on Products | import prince
unbalanced = ["CANALE DISTRIB (Description)","CLASSE COGE (Description)","BIOLOGICO (Description)"]
features_df = product_cluster.iloc[:,1:-2].drop(unbalanced,axis=1)
features_df = df_produit
mca = prince.MCA(features_df)
mca.plot_relationship_square()
plt.show() | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Apply MCA on Clients | import prince
features_df = client_df.astype(str).fillna("NA")
mca = prince.MCA(features_df)
mca.plot_relationship_square()
plt.show() | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Classification Tree | from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
import subprocess
from sklearn.tree import export_graphviz
from sklearn.preprocessing import OneHotEncoder,LabelBinarizer,LabelEncoder
def visualize_tree(tree, feature_names,class_names=None):
with open(r... | _____no_output_____ | MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Evaluation the algorithm | from sklearn.metrics import classification_report, confusion_matrix
from sklearn.metrics import precision_recall_fscore_support as report
#print(confusion_matrix(y_test, y_pred))
precision,recall,fscore,support = report(y_test, y_pred,warn_for=())
print(precision.mean(),recall.mean(),fscore.mean(),support.sum())
#p... | <sklearn.tree._tree.Tree object at 0x000000000A407C60>
| MIT | notebooks/p2c4/0.4-imad-cluster-analysis.ipynb | imad24/pc_clustering |
Decoding in time-frequency space using Common Spatial Patterns (CSP)The time-frequency decomposition is estimated by iterating over raw data thathas been band-passed at different frequencies. This is used to compute acovariance matrix over each epoch or a rolling time-window and extract the CSPfiltered signals. A line... | # Authors: Laura Gwilliams <laura.gwilliams@nyu.edu>
# Jean-Remi King <jeanremi.king@gmail.com>
# Alex Barachant <alexandre.barachant@gmail.com>
# Alexandre Gramfort <alexandre.gramfort@inria.fr>
#
# License: BSD (3-clause)
import numpy as np
import matplotlib.pyplot as plt
from mne import... | _____no_output_____ | MIT | practical_classes/Week_7/plot_decoding_csp_timefreq_practical_7.ipynb | victor-m-p/adv_cognitive_neuroscience |
Set parameters and read data | event_id = dict(hands=2, feet=3) # motor imagery: hands vs feet
subject = 1
runs = [6, 10, 14]
raw_fnames = eegbci.load_data(subject, runs)
raw = concatenate_raws([read_raw_edf(f) for f in raw_fnames])
# Extract information from the raw file
sfreq = raw.info['sfreq']
events, _ = events_from_annotations(raw, event_id=... | _____no_output_____ | MIT | practical_classes/Week_7/plot_decoding_csp_timefreq_practical_7.ipynb | victor-m-p/adv_cognitive_neuroscience |
Loop through frequencies, apply classifier and save scores | # init scores
freq_scores = np.zeros((n_freqs - 1,))
# Loop through each frequency range of interest
for freq, (fmin, fmax) in enumerate(freq_ranges):
# Infer window size based on the frequency being used
w_size = n_cycles / ((fmax + fmin) / 2.) # in seconds
# Apply band-pass filter to isolate the speci... | _____no_output_____ | MIT | practical_classes/Week_7/plot_decoding_csp_timefreq_practical_7.ipynb | victor-m-p/adv_cognitive_neuroscience |
Plot frequency results | plt.bar(freqs[:-1], freq_scores, width=np.diff(freqs)[0],
align='edge', edgecolor='black')
plt.xticks(freqs)
plt.ylim([0, 1])
plt.axhline(len(epochs['feet']) / len(epochs), color='k', linestyle='--',
label='chance level')
plt.legend()
plt.xlabel('Frequency (Hz)')
plt.ylabel('Decoding Scores')
plt.ti... | _____no_output_____ | MIT | practical_classes/Week_7/plot_decoding_csp_timefreq_practical_7.ipynb | victor-m-p/adv_cognitive_neuroscience |
Loop through frequencies and time, apply classifier and save scores | # init scores
tf_scores = np.zeros((n_freqs - 1, n_windows))
# Loop through each frequency range of interest
for freq, (fmin, fmax) in enumerate(freq_ranges):
# Infer window size based on the frequency being used
w_size = n_cycles / ((fmax + fmin) / 2.) # in seconds
# Apply band-pass filter to isolate t... | _____no_output_____ | MIT | practical_classes/Week_7/plot_decoding_csp_timefreq_practical_7.ipynb | victor-m-p/adv_cognitive_neuroscience |
Plot time-frequency results | # Set up time frequency object
av_tfr = AverageTFR(create_info(['freq'], sfreq), tf_scores[np.newaxis, :],
centered_w_times, freqs[1:], 1)
chance = np.mean(y) # set chance level to white in the plot
av_tfr.plot([0], vmin=chance, title="Time-Frequency Decoding Scores",
cmap=plt.cm.Reds) | _____no_output_____ | MIT | practical_classes/Week_7/plot_decoding_csp_timefreq_practical_7.ipynb | victor-m-p/adv_cognitive_neuroscience |
week03: Логистическая регрессия и анализ изображенийВ этом ноутбуке предлагается построить классификатор изображений на основе логистической регрессии. *Забегая вперед, мы попробуем решить задачу классификации изображений используя лишь простые методы. В третьей части нашего курса мы вернемся к этой задаче.* | import numpy as np
import matplotlib.pyplot as plt
import h5py
%matplotlib inline | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
1. Постановка задачи **Задача**: Есть датасет [прямая ссылка](https://drive.google.com/file/d/15tOimf2QYWsMtPJXTUCwgZaOTF8Nxcsm/view?usp=sharing) ("catvnoncat.h5") состоящий из: - обучающей выборки из m_train изображений, помеченных "cat" (y=1) или "non-cat" (y=0) - тестовой выборки m_test изображений, помеченны... | # Uncomment this cell to download the data
# !wget "https://downloader.disk.yandex.ru/disk/7ef1d1e30e23740a4a30799a825319154815ddc85bf689542add0a3d11ccb91c/5d7fdcb0/3dcxK38Q0fG3ui0g2gMZgKkLls8ULwVpoYNkWpBm9d24EceJ6mIoH5l3_wKkFv3PfZ0WMGYjfJULynuJkuGaug%3D%3D?uid=76549735&filename=data.zip&disposition=attachment&hash=&l... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
1.1 Загрузка данных и визуализация | def load_dataset():
train_data = h5py.File("data/train_catvnoncat.h5", "r")
train_set_x_orig = np.array(train_data["train_set_x"][:]) # признаки
train_set_y_orig = np.array(train_data["train_set_y"][:]) # метки классов
test_data = h5py.File("data/test_catvnoncat.h5", "r")
test_set_x_orig = np.array... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
Цветные изображения в формате RGB представлены в виде трёхмерных numpy.array.Порядок измерений $H \times W \times C$: $H$ - высота, $W$ - ширина и $C$ - число каналов.Значение каждого пиксела находится в интервале $[0;255]$. | from ipywidgets import interact, interactive, fixed, interact_manual
import ipywidgets as widgets
def show_image_interact(i=0):
f, ax = plt.subplots(1,4, figsize=(15,20), sharey=True)
ax[0].imshow(train_set_x_orig[i])
ax[0].set_title('RGB image')
ax[1].imshow(train_set_x_orig[i][:,:,0], cmap='gray... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
При работе с данными полезно будет сохранить размерности входных изображений для дальнейшей обработки. | m_train = train_set_x_orig.shape[0]
m_test = test_set_x_orig.shape[0]
src_size = train_set_x_orig.shape[1]
print ("Размер обучающей выборки: m_train = " + str(m_train))
print ("Размер тестовой выборки: m_test = " + str(m_test))
print ("Ширина/Высота каждого изображения: src_size = " + str(src_size))
print ("Размерны т... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
2. Предварительная обработка Преобразуем входные изображения размера (num_px, num_px, 3) в вектор признаков размера (num_px $*$ num_px $*$ 3, 1), чтобы сформировать матрицы объект-признак в виде numpy-array для обучающей и тестовой выборок.Каждой строке матрицы объект-признак соответствует входное развёрнутое в вектор... | def image_preprocessing_simple(data):
assert type(data) == np.ndarray
assert data.ndim == 4
n,h,w,c = data.shape
data_vectorized = <ваш код>
data_normalized = <ваш код>
return data_normalized
# Изменить размеры входных данных
train_set_x_vectorized = image_preprocessing_simple(train_s... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
3. Классификация | from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import GridSearchCV
import warnings
warnings.filterwarnings('ignore') | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
**Вопрос**: Какую метрику качества стоит использовать? 3.1 Построение модели Построим модель с параметрами по умолчанию и посмотрим, как хорошо она справится с задачей. | clf = # <ваш код>
score = <ваш код>
print('Точность для простой модели с параметрами по умолчанию: {:.4f}'.format(score))
# from sklearn.metrics import f1_score
y_predicted = clf.predict(test_set_x_vectorized)
correct_score = <ваш код>
print('<Имя метрики> для простой модели: {:.4f}'.format(correct_score)) | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
Попробуем подобрать параметры регуляризации в надежде, что это повысит точность предсказаний. | <ваш код>
print('Оптимальные параметры: {}'.format(<ваш код>))
print('Наилучшее значение метрики качества: {}'.format(<ваш код>)) | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
Обучим модель с оптимальными параметрами на всей обучающей выборке и посмотрим на метрики качества: | best_clf = <ваш код>
best_clf.fit(train_set_x_vectorized, train_set_y)
y_predicted = best_clf.predict(test_set_x_vectorized)
metric_score = <ваш код>(y_predicted, test_set_y)
print('Optimal model hyperparameters accuracy score: {:.4f}'.format(metric_score)) | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
3.2 Анализ ошибок | is_outlier = (y_predicted != test_set_y)
test_outliers_x, test_outliers_y, predicted_y = test_set_x_orig[is_outlier], test_set_y[is_outlier], y_predicted[is_outlier]
def show_image_outliers(i=0):
f = plt.figure(figsize=(5,5))
plt.imshow(test_outliers_x[i])
plt.title('RGB image')
fmt_string = "Sampl... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
**Вопрос**: Как по-вашему можно повысить точность? Каким недостатком обладает данный подход к классификации? 3.3 Модель с аугментациями Как можно увеличить количество данных для обучения?Сформировать новые примеры из уже имеющихся!Например, можно пополнить class 'cat' обучающей выборки [зеркально отображёнными](https:... | def augment_sample(src, label):
<ваш код>
def image_preprocessing_augment(data, labels):
assert type(data) == np.ndarray
assert data.ndim == 4
## ВАШ КОД ##
data_augmented =
labels_augmented =
## ВАШ КОД ЗАКАНЧИВАЕТСЯ ЗДЕСЬ ##
n,h,w,c = data_augmented.shape
dat... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
4. Проверьте работу классификатора на своей картинке Библиотека [OpenCV](https://opencv.org) для работы с изображениями для [python](https://pypi.org/project/opencv-python/):`pip install opencv-python`Вместе с contrib-модулями:`pip install opencv-contrib-python` | import cv2
# Путь к картинке на вашем ПК
fname = "cat-non-cat.jpg"
# Считываем картинку через scipy
src = cv2.cvtColor(cv2.imread((fname)), cv2.COLOR_BGR2RGB)
src_resized = cv2.resize(src, (src_size,src_size), interpolation=cv2.INTER_LINEAR).reshape(1, src_size*src_size*3)
my_image_predict = clf.predict(src_resized)[0... | _____no_output_____ | MIT | week0_03_logistic/week03_extra_image_classifier.ipynb | GendalfSeriyy/ml-mipt |
Extracted from text tokentweet_feature_mentions: list of ints (or None):- Mentions extracted from the tweet.tweet_feature_number_of_mentions: int:- Number of mentions in the tweet.tweet_feature_token_length: int:- Number of BERT tokens in the tweet.tweet_feature_token_length_unique: int:- Number of unique bert tokens ... | import sys
sys.path.append('..')
import core.config as conf
from utils.preprocessing import *
import numpy as np
from tqdm import tqdm
from datetime import datetime
import matplotlib.pyplot as plt
import tensorflow as tf
from transformers import *
from tensorflow.keras.preprocessing.sequence import pad_sequences
fro... | _____no_output_____ | Apache-2.0 | preprocessing/tweet_features.ipynb | RecKIE7/recsys2021-twitter |
Load data | data_path = '/hdd/twitter/dataset_mini/train'
df = read_data(data_path)
text_tokens = df['text_tokens']
text_tokens
df['len_text_tokens'] = df['text_tokens'].apply(lambda x: len(x.split('\t')))
df['decoded_text_tokens'] = df['text_tokens'].progress_apply(lambda x: tokenizer.decode(x.split('\t'), skip_special_tokens=Tru... | _____no_output_____ | Apache-2.0 | preprocessing/tweet_features.ipynb | RecKIE7/recsys2021-twitter |
Using resultsSince json is a dictionary, you can pull out a single datapoint using the key.```{ "source": "ensembl_havana", "object_type": "Gene", "logic_name": "ensembl_havana_gene", "version": 12, "species": "homo_sapiens", "description": "B-Raf proto-oncogene, serine/threonine kinase [Source:HGNC Symbol;Acc:H... | import requests, json
from pprint import pprint
def fetch_endpoint(server, request, content_type):
r = requests.get(server+request, headers={ "Accept" : content_type})
if not r.ok:
r.raise_for_status()
sys.exit()
if content_type == 'application/json':
return r.json()
else:
... | BRAF
| Apache-2.0 | Python3/3_Using_results.ipynb | fabiolib/rest-api-jupyter-course |
Exercises 31\. Write a script to lookup the gene called *ESPN* in human and print the stable ID of this gene. | # Exercise 3.1
#!/usr/bin/env python
# Get modules needed for script
import sys, requests, json
from pprint import pprint
def fetch_endpoint(server, request, content_type):
r = requests.get(server+request, headers={ "Accept" : content_type})
if not r.ok:
r.raise_for_status()
sys.exit()
... | 'ENSG00000187017'
| Apache-2.0 | Python3/3_Using_results.ipynb | fabiolib/rest-api-jupyter-course |
2\. Get all variants that are associated with the phenotype 'Coffee consumption'. For each variant print a. the p-value for the association b. the PMID for the publication which describes the association between that variant and ‘Coffee consumption’ c. the risk allele and the associated gene. | # Exercise 3.2
pprint(get_lookup) | {'assembly_name': 'GRCh38',
'biotype': 'protein_coding',
'db_type': 'core',
'description': 'espin [Source:HGNC Symbol;Acc:HGNC:13281]',
'display_name': 'ESPN',
'end': 6461367,
'id': 'ENSG00000187017',
'logic_name': 'ensembl_havana_gene_homo_sapiens',
'object_type': 'Gene',
'seq_region_name': '1',
'source': 'e... | Apache-2.0 | Python3/3_Using_results.ipynb | fabiolib/rest-api-jupyter-course |
3\. Get the mouse homologue of the human BRCA2 and print the ID and sequence of both.Note that the JSON for the endpoint you need is several layers deep, containing nested lists (appear as square brackets [ ] in the JSON) and key value sets (dictionary; appear as curly brackets { } in the JSON). Pretty print (pprint) c... | # Exercise 3.3 | _____no_output_____ | Apache-2.0 | Python3/3_Using_results.ipynb | fabiolib/rest-api-jupyter-course |
1. Welcome to the Hour of Callysto!Lesson created and taught by [Discover Coding](https://discovercoding.ca).With support and funding from [Callysto](https://callysto.ca/) and the [Pacific Institute for Mathematical Sciences](https://www.pims.math.ca/) 2. What is Callysto?- **Callysto** is a free and online collectio... | # My First Turtle Program!
from mobilechelonian import Turtle
ted = Turtle()
ted.forward(50) | _____no_output_____ | MIT | hour-of-callysto-2.ipynb | callysto/hour-of-callysto |
Let's break down what we each line of code that we see:1. ` My First Turtle Program!` - Any line starting with a `` is a comment (or notes). It is NOT CODE!1. `from mobilechelonian import Turtle` - This lets use some `Turtle` code that someone else wrote. We won't worry about it for this class.1. `ted = Turtle()` - Thi... | # My Turtle is now 2D!
from mobilechelonian import Turtle
ted = Turtle()
ted.speed(5)
ted.forward(100)
ted.left(120)
ted.forward(100)
ted.left(120)
ted.forward(100) | _____no_output_____ | MIT | hour-of-callysto-2.ipynb | callysto/hour-of-callysto |
*ACTIVITY 5.2*Can you change the color of the triangle?Try inserting this line of code into Example 5.1:`ted.pencolor('red')`Try other [colors named here](https://www.w3schools.com/tags/ref_colornames.asp)**Observe**1. What happens if you write this BEFORE the first move `forward` (on line 5)?1. What happens if you wr... | # My Turtle is now 2D!
from mobilechelonian import Turtle
ted = Turtle()
ted.speed(1)
# hint 1 - copy and paste the previous code
# hint 2 - the next turn should be 30 degrees
# hint 3 - forward, turn, forward, turn, forward... | _____no_output_____ | MIT | hour-of-callysto-2.ipynb | callysto/hour-of-callysto |
6. LOOPSDid you notice that drawing shapes used the same lines of code over and over?Instead of typing the same things over and over, we can use a *loop* to run the same code multiple times.Loops run code __*OVER and OVER and OVER and . . .*__Loops have two parts:1. A line starting with the special keyword `for` or `w... | from mobilechelonian import Turtle
ted = Turtle()
ted.speed(5)
# This loop runs (how many?) times
for index in range(10):
ted.forward(50)
ted.right(80)
| _____no_output_____ | MIT | hour-of-callysto-2.ipynb | callysto/hour-of-callysto |
*ACTIVITY 6.2*Experiment with the code in Example 6.1, and try different values.Can you:1. Make turtle draw faster?1. Change how BIG the shape is? (Hint: change the distance turtle moves `forward`)1. Change how much the turtle turns each time? 1. What happens when turtle turns less than 90? 1. What happens when ... | from mobilechelonian import Turtle
import random
ted = Turtle()
ted.speed(10)
# This loop runs 10 times
for index in range(10):
red = random.randint(0,255)
green = random.randint(0,255)
blue = random.randint(0,255)
random_color = "RGB(%d,%d,%d)" % (red,green,blue)
ted.pencolor(random_color)
... | _____no_output_____ | MIT | hour-of-callysto-2.ipynb | callysto/hour-of-callysto |
*ACTIVITY 7.2*Let's experiment with the code in Example 7.11. On line 9, we pick a random number between 0 to 255 for `red`. What happens if we pick a random number between 200 to 255?1. What happens if we pick a random number between 0 and 10 for the color `red`?1. Try to change the range of random numbers for `green... | from mobilechelonian import Turtle
import random
ted = Turtle()
ted.speed(10)
# Now we'll run a loop.
for index in range(5):
red = random.randint(0,255)
green = random.randint(0,255)
blue = random.randint(0,255)
random_color = "RGB(%d,%d,%d)" % (red,green,blue)
ted.pencolor(random_color)
... | _____no_output_____ | MIT | hour-of-callysto-2.ipynb | callysto/hour-of-callysto |
ZigZag | das_template_x = np.array([2.5*np.sqrt(2)*i for i in range(24)])
das_template_y = np.array([2.5*np.sqrt(2)*i for i in range(24)])
das_template_x2 = np.hstack([das_template_x,das_template_x[::-1],das_template_x,das_template_x[::-1]])
das_template_y2 = np.hstack([das_template_y,das_template_y+das_template_y[-1],das_templ... | _____no_output_____ | MIT | OptimalDesign/DAS_Designs.ipynb | jbmuir/DAS-Reconstruction |
Spiral | das_theta2 = np.linspace(0,(360*4)**2, 192*2)
das_theta = np.deg2rad(np.sqrt(das_theta2))
a = 0
b = 1
das_r = b*das_theta
das_x = das_r * np.cos(das_theta)
das_y = das_r * np.sin(das_theta)
raz = np.pi/2-np.arctan2(b*np.tan(das_theta)+(a+b*das_theta), b-(a+b*das_theta)*np.tan(das_theta))
das_az = np.rad2deg(raz)
xwid... | _____no_output_____ | MIT | OptimalDesign/DAS_Designs.ipynb | jbmuir/DAS-Reconstruction |
Crossing | template = np.linspace(0,320, 65)
template = (template[1:]+template[:-1])/2
das_x = np.hstack([template, template, template,[80 for i in range(len(template))], [160 for i in range(len(template))], [240 for i in range(len(template))]])
das_y = np.hstack([[80 for i in range(len(template))], [160 for i in range(len(templ... | _____no_output_____ | MIT | OptimalDesign/DAS_Designs.ipynb | jbmuir/DAS-Reconstruction |
Random | template = np.linspace(0,320, 65)
template = (template[1:]+template[:-1])/2
np.random.seed(94899109)
das_x = np.random.uniform(5,315,384)
das_y = np.random.uniform(5,315,384)
das_az = np.random.uniform(0,360,384)
raz = np.deg2rad(das_az)
L = 10 #gauge length
ll = np.linspace(-L/2, L/2, 2**5+1)
dl = ll[1]-ll[0]
p1 = ... | _____no_output_____ | MIT | OptimalDesign/DAS_Designs.ipynb | jbmuir/DAS-Reconstruction |
Eigenvalue Spectrum | G_full = np.vstack([np.hstack([G3, np.zeros(G3.shape)]), np.hstack([np.zeros(G3.shape), G3])])
idet = 1e-10*np.eye(G_zigzag.shape[1])
ezig = np.sort(np.linalg.eigvals(G_zigzag.T@G_zigzag+idet))[::-1]
espi = np.sort(np.linalg.eigvals(G_spiral.T@G_spiral+idet))[::-1]
ecro = np.sort(np.linalg.eigvals(G_cross.T@G_cross+id... | _____no_output_____ | MIT | OptimalDesign/DAS_Designs.ipynb | jbmuir/DAS-Reconstruction |
$\newcommand{\xv}{\mathbf{x}} \newcommand{\wv}{\mathbf{w}} \newcommand{\yv}{\mathbf{y}} \newcommand{\zv}{\mathbf{z}} \newcommand{\Chi}{\mathcal{X}} \newcommand{\R}{\rm I\!R} \newcommand{\sign}{\text{sign}} \newcommand{\Tm}{\mathbf{T}} \newcommand{\Xm}{\mathbf{X}} \newcommand{\Xlm}{\mathbf{X1}} \newcommand{\Wm}{\mathbf{... | import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
fig, ax = plt.subplots()
# x - y axis
ax.axhline(y=0, color='k', linewidth=1)
ax.axvline(x=0, color='k', linewidth=1)
# step function in blue
plt.plot([0, 6], [1, 1], 'b-', linewidth=3)
plt.plot([-6, 0], [-1, -1], 'b-', linewidth=3)
plt.plot([0,... | _____no_output_____ | MIT | reading_assignments/6_Note-Neural Networks.ipynb | biqar/Fall-2020-ITCS-8156-MachineLearning |
Non-linear Extension of Linear ModelAs we discussed, feed forward neural networks have a rich representation. Thus, it can represent the linear model with single layer. Considering the multiple outputs, we formulated this in matrix: $$\begin{align}E &= \frac{1}{N} \frac{1}{K} \sum_{n=1}^{N} \sum_{k=1}^{K} (t_{nk} - y_... | # polinomial function
def h_poly(x):
return x**4 + 3 * x**2 + 7 * x + 3
# sigmoid function
def h_sigmoid(x):
return np.tanh(x)
##### Gradient functions
# polinomial function
def dh_poly(x):
return 4 * x**3 + 6 * x + 7
# polinomial function
def dh_sigmoid(x):
h = h_sigmoid(x)
return 1 - h ** ... | _____no_output_____ | MIT | reading_assignments/6_Note-Neural Networks.ipynb | biqar/Fall-2020-ITCS-8156-MachineLearning |
Here, we can see the polinomial gradients are very huge when $x$ is moving away from 0. A gradient descent procedure takes this huge step for the large positive or negative $x$ values, which can make learning divergent and unstable.In the right figure, we can see the gradient is nearly turned off for large $x$ values. ... | import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
import IPython.display as ipd # for display and clear_output
import time # for sleep
# Make some training data
n = 20
X = np.linspace(0.,20.0,n).reshape((n,1)) - 10
T = 0.2 + 0.05 * (X+10) + 0.4 * np.sin(X+10) + 0.2 * np.random.normal(size=(n,1))
... | V: (2, 10)
X1: (2, 20)
error: (20, 1)
W.T: (1, 11)
Z: (20, 10)
(20, 10)
| MIT | reading_assignments/6_Note-Neural Networks.ipynb | biqar/Fall-2020-ITCS-8156-MachineLearning |
$\newcommand{\xv}{\mathbf{x}} \newcommand{\wv}{\mathbf{w}} \newcommand{\yv}{\mathbf{y}} \newcommand{\zv}{\mathbf{z}} \newcommand{\av}{\mathbf{a}} \newcommand{\Chi}{\mathcal{X}} \newcommand{\R}{\rm I\!R} \newcommand{\sign}{\text{sign}} \newcommand{\Tm}{\mathbf{T}} \newcommand{\Xm}{\mathbf{X}} \newcommand{\Xlm}{\mathbf{X... | import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
import scipy.optimize as opt
from scipy.optimize import rosen, minimize
# examples are from http://people.duke.edu/~ccc14/sta-663-2017/14C_Optimization_In_Python.html
x = np.linspace(-5, 5, 1000)
y = np.linspace(-5, 5, 1000)
xs, ys = np.meshgrid(x... | _____no_output_____ | MIT | reading_assignments/6_Note-Neural Networks.ipynb | biqar/Fall-2020-ITCS-8156-MachineLearning |
Vs. others?Now, let us take a look at other optimization tools including naive steepest descent and scaled conjugate gradient ([Moller, 1997](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.50.8063&rep=rep1&type=pdf)). To run this properly, you need to download [grad.py](https://webpages.uncc.edu/mlee173/teac... | from grad import steepest
res = steepest(np.array(x0), rosen_der, rosen, stepsize=0.0001, wtracep=True, ftracep=True)
ps = np.array(res['wtrace'])
plt.figure(figsize=(16, 8))
plt.subplot(121)
plt.contour(xs, ys, zs, np.arange(10)**5, cmap='jet')
plt.plot(ps[:, 0], ps[:, 1], '-ro')
plt.subplot(122)
plt.semilogy(range(... | _____no_output_____ | MIT | reading_assignments/6_Note-Neural Networks.ipynb | biqar/Fall-2020-ITCS-8156-MachineLearning |
Neural Network! Now, let us use this optimization trick for our neural networks. | # standardization class
class Standardizer:
""" class version of standardization """
def __init__(self, X, explore=False):
self._mu = np.mean(X,0)
self._sigma = np.std(X,0)
if explore:
print ("mean: ", self._mu)
print ("sigma: ", self._sigma)
print ... | _____no_output_____ | MIT | reading_assignments/6_Note-Neural Networks.ipynb | biqar/Fall-2020-ITCS-8156-MachineLearning |
Reinforcement Learning for seq2seqThis time we'll solve a problem of transribing hebrew words in english, also known as g2p (grapheme2phoneme) * word (sequence of letters in source language) -> translation (sequence of letters in target language)Unlike what most deep learning practicioners do, we won't only train it t... | import sys
if 'google.colab' in sys.modules:
!wget https://raw.githubusercontent.com/yandexdataschool/Practical_RL/spring20/week07_seq2seq/basic_model_torch.py -O basic_model_torch.py
!wget https://raw.githubusercontent.com/yandexdataschool/Practical_RL/spring20/week07_seq2seq/main_dataset.txt -O main_dataset.t... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Step 1: preprocessingWe shall store dataset as a dictionary`{ word1:[translation1,translation2,...], word2:[...],...}`.This is mostly due to the fact that many words have several correct translations.We have implemented this thing for you so that you can focus on more interesting parts.__Attention python2 users!__ You... | import numpy as np
from collections import defaultdict
word_to_translation = defaultdict(list) # our dictionary
bos = '_'
eos = ';'
with open("main_dataset.txt", encoding="utf-8") as fin:
for line in fin:
en, he = line[:-1].lower().replace(bos, ' ').replace(eos,
... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
split the datasetWe hold out 10% of all words to be used for validation. | from sklearn.model_selection import train_test_split
train_words, test_words = train_test_split(
all_words, test_size=0.1, random_state=42) | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Building vocabulariesWe now need to build vocabularies that map strings to token ids and vice versa. We're gonna need these fellas when we feed training data into model or convert output matrices into english words. | from voc import Vocab
inp_voc = Vocab.from_lines(''.join(all_words), bos=bos, eos=eos, sep='')
out_voc = Vocab.from_lines(''.join(all_translations), bos=bos, eos=eos, sep='')
# Here's how you cast lines into ids and backwards.
batch_lines = all_words[:5]
batch_ids = inp_voc.to_matrix(batch_lines)
batch_lines_restored =... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Draw word/translation length distributions to estimate the scope of the task. | import matplotlib.pyplot as plt
%matplotlib inline
plt.figure(figsize=[8, 4])
plt.subplot(1, 2, 1)
plt.title("words")
plt.hist(list(map(len, all_words)), bins=20)
plt.subplot(1, 2, 2)
plt.title('translations')
plt.hist(list(map(len, all_translations)), bins=20) | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Step 3: deploy encoder-decoder (1 point)__assignment starts here__Our architecture consists of two main blocks:* Encoder reads words character by character and outputs code vector (usually a function of last RNN state)* Decoder takes that code vector and produces translations character by characterThan it gets fed int... | import torch
import torch.nn as nn
import torch.nn.functional as F
from basic_model_torch import BasicTranslationModel
model = BasicTranslationModel(inp_voc, out_voc,
emb_size=64, hid_size=256)
# Play around with symbolic_translate and symbolic_score
inp = torch.tensor(np.random.randint(0,... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Scoring functionLogLikelihood is a poor estimator of model performance.* If we predict zero probability once, it shouldn't ruin entire model.* It is enough to learn just one translation if there are several correct ones.* What matters is how many mistakes model's gonna make when it translates!Therefore, we will use mi... | import editdistance # !pip install editdistance
def get_distance(word, trans):
"""
A function that takes word and predicted translation
and evaluates (Levenshtein's) edit distance to closest correct translation
"""
references = word_to_translation[word]
assert len(references) != 0, "wrong/unk... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Step 2: Supervised pre-training (2 points)Here we define a function that trains our model through maximizing log-likelihood a.k.a. minimizing crossentropy. | import random
def sample_batch(words, word_to_translation, batch_size):
"""
sample random batch of words and random correct translation for each word
example usage:
batch_x,batch_y = sample_batch(train_words, word_to_translations,10)
"""
# choose words
batch_words = np.random.choice(words,... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Actually train the modelMinibatches and stuff... | from IPython.display import clear_output
from tqdm import tqdm, trange # or use tqdm_notebook,tnrange
loss_history = []
editdist_history = []
entropy_history = []
opt = torch.optim.Adam(model.parameters())
for i in trange(25000):
loss = compute_loss_on_batch(*sample_batch(train_words, word_to_translation, 32))
... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
__How to interpret the plots:__* __Train loss__ - that's your model's crossentropy over minibatches. It should go down steadily. Most importantly, it shouldn't be NaN :)* __Val score distribution__ - distribution of translation edit distance (score) within batch. It should move to the left over time.* __Val score / tra... | for word in train_words[:10]:
print("%s -> %s" % (word, translate([word])[0]))
test_scores = []
for start_i in trange(0, len(test_words), 32):
batch_words = test_words[start_i:start_i+32]
batch_trans = translate(batch_words)
distances = list(map(get_distance, batch_words, batch_trans))
test_scores.e... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Self-critical policy gradient (2 points)In this section you'll implement algorithm called self-critical sequence training (here's an [article](https://arxiv.org/abs/1612.00563)).The algorithm is a vanilla policy gradient with a special baseline. $$ \nabla J = E_{x \sim p(s)} E_{y \sim \pi(y|x)} \nabla log \pi(y|x) \cd... | def compute_reward(input_sequence, translations):
""" computes sample-wise reward given token ids for inputs and translations """
distances = list(map(get_distance,
inp_voc.to_lines(input_sequence.data.numpy()),
out_voc.to_lines(translations.data.numpy())))
... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Policy gradient training | entropy_history = [np.nan] * len(loss_history)
opt = torch.optim.Adam(model.parameters(), lr=1e-5)
for i in trange(100000):
loss, ent = scst_objective_on_batch(
sample_batch(train_words, word_to_translation, 32)[0]) # [0] = only source sentence
# train with backprop
loss.backward()
opt.step()
... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
__Debugging tips:__ * As usual, don't expect improvements right away, but in general the model should be able to show some positive changes by 5k steps. * Entropy is a good indicator of many problems. * If it reaches zero, you may need greater entropy regularizer. * If it has rapid changes time to time, you may ne... | for word in train_words[:10]:
print("%s -> %s" % (word, translate([word])[0]))
test_scores = []
for start_i in trange(0, len(test_words), 32):
batch_words = test_words[start_i:start_i+32]
batch_trans = translate(batch_words)
distances = list(map(get_distance, batch_words, batch_trans))
test_scores.e... | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
Step 6: Make it actually work (5++ pts)In this section we want you to finally __restart with EASY_MODE=False__ and experiment to find a good model/curriculum for that task.We recommend you to start with the following architecture```encoder---decoder P(y|h) ^ LSTM -> LSTM ^ ^ biLSTM ... | assert not EASY_MODE, "make sure you set EASY_MODE = False at the top of the notebook." | _____no_output_____ | Unlicense | week07_seq2seq/practice_torch.ipynb | dmgirdyuk/Practical_RL |
DEAP DEAP is a novel evolutionary computation framework for rapid prototyping and testing of ideas. It seeks to make algorithms explicit and data structures transparent. It works in perfect harmony with parallelisation mechanism such as multiprocessing and SCOOP. The following documentation presents the key concepts a... | import random
from deap import base
from deap import creator
from deap import tools
# creator is a class factory that can build new classes at run-time
creator.create("FitnessMax", base.Fitness, weights=(1.0,))
creator.create("Individual", list, fitness=creator.FitnessMax)
# a toolbox stores functions and their argume... | Best individual is [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],... | MIT | ipython-notebooks/notebooks/libraries/DEAP.ipynb | gopala-kr/ds-notebooks |
Symbolic Regression (GP) Symbolic regression is one of the best known problems in GP. It is commonly used as a tuning problem for new algorithms, but is also widely used with real-life distributions, where other regression methods may not work.All symbolic regression problems use an arbitrary data distribution, and tr... | import operator
import math
import random
import numpy
from deap import algorithms
from deap import base
from deap import creator
from deap import tools
from deap import gp
# define a new function for divison that guards against divide by 0
def protectedDiv(left, right):
try:
return left / right
exce... | fitness size
--------------------------------------- -------------------------------
gen nevals avg max min std avg max min std
0 300 2.39949 59.2593 0.165572 4.64122 3.69667 7 2 1.61389
1 146 1.0971 ... | MIT | ipython-notebooks/notebooks/libraries/DEAP.ipynb | gopala-kr/ds-notebooks |
Introdução | import pandas as pd;
dados_enem = pd.read_csv('https://github.com/alura-cursos/imersao-dados-2-2020/blob/master/MICRODADOS_ENEM_2019_SAMPLE_43278.csv?raw=true')
# retorna as 5 primeiras linhas da colecao
dados_enem.head()
# para acessar apenas uma coluna -> dados_enem['nome_da_coluna']
dados_enem['SG_UF_RESIDENCIA'... | _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
Continuação | # visualização em histograma
#dados_enem['NU_IDADE'].hist()
dados_enem['NU_IDADE'].hist(bins=100,figsize=(11,9),legend=True)
treineiros = dados_enem.query('IN_TREINEIRO == 1')
treineiros['NU_IDADE'].value_counts()
# Notas Redacao
dados_enem['NU_NOTA_REDACAO'].hist(bins=20)
# analise geral
provas = ["NU_NOTA_CN","NU_NOT... | _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
Desafios - 01 : Informar a proporção de inscritos por idades | # desafio 1
def proporcao(total_itens):
def funcao_calculo(x):
res = x * 100 / total_itens
return round(res, 6);
return funcao_calculo
total = len(dados_enem)
inscritos_por_idade = dados_enem['NU_IDADE'].value_counts()
inscritos_por_idade.apply(proporcao(total)) | _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
- 02: Descobrir de quais estados são os inscritos com 13 anos | # desafio 02
cabecalhos= ['SG_UF_RESIDENCIA','NU_IDADE']
inscritos=dados_enem[cabecalhos]
inscritos.query('NU_IDADE==13')
| _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
- 03: Qual a proporcao dos alunos com 18 anos por estado | # desafio 3
cabecalhos = ['SG_UF_RESIDENCIA','NU_IDADE']
novo_df = dados_enem[cabecalhos]
inscritos = novo_df.query('NU_IDADE==18')
total_inscritos = len(inscritos)
inscritos.value_counts().apply(proporcao(total_inscritos))
| _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
- 04: Plotar Histogramas das idades de treineiros e não treineiros | # desafio 4
inscritos_treineiros = dados_enem.query('IN_TREINEIRO == 1')['NU_IDADE'].value_counts();
inscritos_treineiros.hist(bins=30,figsize=(10,7),legend=True)
inscritos_nao_treineiros = dados_enem.query('IN_TREINEIRO == 0')['NU_IDADE'].value_counts();
inscritos_nao_treineiros.hist(bins=30,figsize=(10,7),legend=True... | _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
- 05: Comparar as distribuições das provas em ingles e espanhol | # desafio 5
# TP_LINGUA==0 -> Ingles
# TP_LINGUA==1 -> Espanhol
dados_enem.query('TP_LINGUA==1')['TP_LINGUA'].hist(bins=20,figsize=(10,7),legend=True)
dados_enem.query('TP_LINGUA==0')['TP_LINGUA'].hist(bins=20,figsize=(10,7),legend=True) | _____no_output_____ | MIT | imersao_dados_aula_1.ipynb | CarlosNeto2804/imersao-dados-2 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.