Skip to content

Commit c01afa1

Browse files
djarpindjarpin
authored
Merge pull request aws#312 from EvanzzzZ/notebook_update
Notebook update
2 parents 2db7bc5 + eb59174 commit c01afa1

File tree

30 files changed

+1027
-548
lines changed

30 files changed

+1027
-548
lines changed

‎advanced_functionality/data_distribution_types/data_distribution_types.ipynb

Lines changed: 81 additions & 33 deletions
Original file line numberDiff line numberDiff line change
@@ -43,6 +43,7 @@
4343
"cell_type": "code",
4444
"execution_count": null,
4545
"metadata": {
46+
"collapsed": true,
4647
"isConfigCell": true
4748
},
4849
"outputs": [],
@@ -69,7 +70,9 @@
6970
{
7071
"cell_type": "code",
7172
"execution_count": null,
72-
"metadata": {},
73+
"metadata": {
74+
"collapsed": true
75+
},
7376
"outputs": [],
7477
"source": [
7578
"import boto3\n",
@@ -102,7 +105,9 @@
102105
{
103106
"cell_type": "code",
104107
"execution_count": null,
105-
"metadata": {},
108+
"metadata": {
109+
"collapsed": true
110+
},
106111
"outputs": [],
107112
"source": [
108113
"def get_gdelt(filename):\n",
@@ -117,7 +122,9 @@
117122
{
118123
"cell_type": "code",
119124
"execution_count": null,
120-
"metadata": {},
125+
"metadata": {
126+
"collapsed": true
127+
},
121128
"outputs": [],
122129
"source": [
123130
"data = get_gdelt('1979.csv')\n",
@@ -142,7 +149,9 @@
142149
{
143150
"cell_type": "code",
144151
"execution_count": null,
145-
"metadata": {},
152+
"metadata": {
153+
"collapsed": true
154+
},
146155
"outputs": [],
147156
"source": [
148157
"data = data[['EventCode', 'NumArticles', 'AvgTone', 'Actor1Geo_Lat', 'Actor1Geo_Long', 'Actor2Geo_Lat', 'Actor2Geo_Long']]\n",
@@ -172,7 +181,9 @@
172181
{
173182
"cell_type": "code",
174183
"execution_count": null,
175-
"metadata": {},
184+
"metadata": {
185+
"collapsed": true
186+
},
176187
"outputs": [],
177188
"source": [
178189
"events = pd.crosstab(index=data['EventCode'], columns='count').sort_values(by='count', ascending=False).index[:20]"
@@ -190,7 +201,9 @@
190201
{
191202
"cell_type": "code",
192203
"execution_count": null,
193-
"metadata": {},
204+
"metadata": {
205+
"collapsed": true
206+
},
194207
"outputs": [],
195208
"source": [
196209
"def write_to_s3(bucket, prefix, channel, file_prefix, X, y):\n",
@@ -219,7 +232,9 @@
219232
{
220233
"cell_type": "code",
221234
"execution_count": null,
222-
"metadata": {},
235+
"metadata": {
236+
"collapsed": true
237+
},
223238
"outputs": [],
224239
"source": [
225240
"for year in range(1979, 1984):\n",
@@ -244,22 +259,21 @@
244259
{
245260
"cell_type": "code",
246261
"execution_count": null,
247-
"metadata": {},
262+
"metadata": {
263+
"collapsed": true
264+
},
248265
"outputs": [],
249266
"source": [
250-
"containers = {'us-west-2': '174872318107.dkr.ecr.us-west-2.amazonaws.com/linear-learner:latest',\n",
251-
" 'us-east-1': '382416733822.dkr.ecr.us-east-1.amazonaws.com/linear-learner:latest',\n",
252-
" 'us-east-2': '404615174143.dkr.ecr.us-east-2.amazonaws.com/linear-learner:latest',\n",
253-
" 'eu-west-1': '438346466558.dkr.ecr.eu-west-1.amazonaws.com/linear-learner:latest',\n",
254-
" 'ap-northeast-1': '351501993468.dkr.ecr.ap-northeast-1.amazonaws.com/linear-learner:latest',\n",
255-
" 'ap-northeast-2': '835164637446.dkr.ecr.ap-northeast-2.amazonaws.com/linear-learner:latest'}\n",
256-
"container = containers[boto3.Session().region_name]"
267+
"from sagemaker.amazon.amazon_estimator import get_image_uri\n",
268+
"container = get_image_uri(boto3.Session().region_name, 'linear-learner')"
257269
]
258270
},
259271
{
260272
"cell_type": "code",
261273
"execution_count": null,
262-
"metadata": {},
274+
"metadata": {
275+
"collapsed": true
276+
},
263277
"outputs": [],
264278
"source": [
265279
"common_training_params = {\n",
@@ -326,7 +340,9 @@
326340
{
327341
"cell_type": "code",
328342
"execution_count": null,
329-
"metadata": {},
343+
"metadata": {
344+
"collapsed": true
345+
},
330346
"outputs": [],
331347
"source": [
332348
"sharded_job = 'DEMO-linear-sharded-' + time.strftime(\"%Y-%m-%d-%H-%M-%S\", time.gmtime())\n",
@@ -341,7 +357,9 @@
341357
{
342358
"cell_type": "code",
343359
"execution_count": null,
344-
"metadata": {},
360+
"metadata": {
361+
"collapsed": true
362+
},
345363
"outputs": [],
346364
"source": [
347365
"replicated_job = 'DEMO-linear-replicated-' + time.strftime(\"%Y-%m-%d-%H-%M-%S\", time.gmtime())\n",
@@ -363,7 +381,9 @@
363381
{
364382
"cell_type": "code",
365383
"execution_count": null,
366-
"metadata": {},
384+
"metadata": {
385+
"collapsed": true
386+
},
367387
"outputs": [],
368388
"source": [
369389
"%%time\n",
@@ -395,7 +415,9 @@
395415
{
396416
"cell_type": "code",
397417
"execution_count": null,
398-
"metadata": {},
418+
"metadata": {
419+
"collapsed": true
420+
},
399421
"outputs": [],
400422
"source": [
401423
"print('Sharded:', sm.describe_training_job(TrainingJobName=sharded_job)['TrainingJobStatus'])\n",
@@ -416,7 +438,9 @@
416438
{
417439
"cell_type": "code",
418440
"execution_count": null,
419-
"metadata": {},
441+
"metadata": {
442+
"collapsed": true
443+
},
420444
"outputs": [],
421445
"source": [
422446
"def get_train_timing(job):\n",
@@ -435,7 +459,9 @@
435459
{
436460
"cell_type": "code",
437461
"execution_count": null,
438-
"metadata": {},
462+
"metadata": {
463+
"collapsed": true
464+
},
439465
"outputs": [],
440466
"source": [
441467
"print('Sharded:', get_train_timing(sharded_job), 'minutes')\n",
@@ -467,7 +493,9 @@
467493
{
468494
"cell_type": "code",
469495
"execution_count": null,
470-
"metadata": {},
496+
"metadata": {
497+
"collapsed": true
498+
},
471499
"outputs": [],
472500
"source": [
473501
"sharded_model_response = sm.create_model(\n",
@@ -483,7 +511,9 @@
483511
{
484512
"cell_type": "code",
485513
"execution_count": null,
486-
"metadata": {},
514+
"metadata": {
515+
"collapsed": true
516+
},
487517
"outputs": [],
488518
"source": [
489519
"replicated_model_response = sm.create_model(\n",
@@ -511,7 +541,9 @@
511541
{
512542
"cell_type": "code",
513543
"execution_count": null,
514-
"metadata": {},
544+
"metadata": {
545+
"collapsed": true
546+
},
515547
"outputs": [],
516548
"source": [
517549
"sharded_endpoint_config = 'DEMO-sharded-endpoint-config-' + time.strftime(\"%Y-%m-%d-%H-%M-%S\", time.gmtime())\n",
@@ -530,7 +562,9 @@
530562
{
531563
"cell_type": "code",
532564
"execution_count": null,
533-
"metadata": {},
565+
"metadata": {
566+
"collapsed": true
567+
},
534568
"outputs": [],
535569
"source": [
536570
"replicated_endpoint_config = 'DEMO-replicated-endpoint-config-' + time.strftime(\"%Y-%m-%d-%H-%M-%S\", time.gmtime())\n",
@@ -556,7 +590,9 @@
556590
{
557591
"cell_type": "code",
558592
"execution_count": null,
559-
"metadata": {},
593+
"metadata": {
594+
"collapsed": true
595+
},
560596
"outputs": [],
561597
"source": [
562598
"%%time\n",
@@ -600,7 +636,9 @@
600636
{
601637
"cell_type": "code",
602638
"execution_count": null,
603-
"metadata": {},
639+
"metadata": {
640+
"collapsed": true
641+
},
604642
"outputs": [],
605643
"source": [
606644
"print('Sharded:', sm.describe_endpoint(EndpointName=sharded_endpoint)['EndpointStatus'])\n",
@@ -619,7 +657,9 @@
619657
{
620658
"cell_type": "code",
621659
"execution_count": null,
622-
"metadata": {},
660+
"metadata": {
661+
"collapsed": true
662+
},
623663
"outputs": [],
624664
"source": [
625665
"test_data = transform_gdelt(get_gdelt('1984.csv'), events).as_matrix()\n",
@@ -637,7 +677,9 @@
637677
{
638678
"cell_type": "code",
639679
"execution_count": null,
640-
"metadata": {},
680+
"metadata": {
681+
"collapsed": true
682+
},
641683
"outputs": [],
642684
"source": [
643685
"def np2csv(arr):\n",
@@ -656,7 +698,9 @@
656698
{
657699
"cell_type": "code",
658700
"execution_count": null,
659-
"metadata": {},
701+
"metadata": {
702+
"collapsed": true
703+
},
660704
"outputs": [],
661705
"source": [
662706
"def predict_batches(data, endpoint):\n",
@@ -684,7 +728,9 @@
684728
{
685729
"cell_type": "code",
686730
"execution_count": null,
687-
"metadata": {},
731+
"metadata": {
732+
"collapsed": true
733+
},
688734
"outputs": [],
689735
"source": [
690736
"sharded_predictions = predict_batches(test_X, sharded_endpoint)\n",
@@ -728,7 +774,9 @@
728774
{
729775
"cell_type": "code",
730776
"execution_count": null,
731-
"metadata": {},
777+
"metadata": {
778+
"collapsed": true
779+
},
732780
"outputs": [],
733781
"source": [
734782
"sm.delete_endpoint(EndpointName=sharded_endpoint)\n",

0 commit comments

Comments
 (0)