s434765/ium_434765
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

sacred
All checks were successful
s434765-training/pipeline/head This commit looks good
Details

Browse Source
This commit is contained in:
Karolina Oparczyk 2021-05-20 22:10:16 +02:00
parent b0346d0b62
commit 3e23841578
29 changed files with 1405 additions and 500 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Dockerfile
View File

@ -8,6 +8,8 @@ RUN pip3 install pandas
RUN pip3 install kaggle
RUN pip3 install tensorflow
RUN pip3 install sklearn
RUN pip3 install pymongo
RUN pip3 install sacred
COPY ./data_train ./
COPY ./data_dev ./
COPY ./neural_network.sh ./

7
JenkinsfileNeural
View File

@ -17,7 +17,7 @@ node {
}
stage('Clone repo') {
try { docker.image("karopa/ium:19").inside {
/*try {*/ docker.image("karopa/ium:21").inside {
stage('Test') {
checkout([$class: 'GitSCM', branches: [[name: '*/master']], doGenerateSubmoduleConfigurations: false, extensions: [], submoduleCfg: [], userRemoteConfigs: [[url: 'https://git.wmi.amu.edu.pl/s434765/ium_434765']]])
copyArtifacts fingerprintArtifacts: true, projectName: 's434765-create-dataset', selector: buildParameter("BUILD_SELECTOR")
@ -28,8 +28,9 @@ node {
'''
archiveArtifacts 'output.txt'
archiveArtifacts 'model/**/*.*'
archiveArtifacts 'my_runs/**/*.*'
}
emailext body: 'Successful build',
/* emailext body: 'Successful build',
subject: "s434765",
to: "26ab8f35.uam.onmicrosoft.com@emea.teams.ms"
}
@ -39,7 +40,7 @@ node {
emailext body: 'Failed build',
subject: "s434765",
to: "26ab8f35.uam.onmicrosoft.com@emea.teams.ms"
throw e
throw e*/
}
}
stage ("build evaluation") {

BIN
evaluation.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 28 KiB

9
model/keras_metadata.pb Normal file
View File

File diff suppressed because one or more lines are too long

BIN
model/saved_model.pb Normal file
View File

Binary file not shown.

BIN
model/variables/variables.data-00000-of-00001 Normal file
View File

Binary file not shown.

BIN
model/variables/variables.index Normal file
View File

Binary file not shown.

4
my_runs/1/config.json Normal file
View File

@ -0,0 +1,4 @@
{
"epochs_amount": 30,
"seed": 511320143
}

79
my_runs/1/cout.txt Normal file
View File

@ -0,0 +1,79 @@
views 0
dtype: int32
views 488
dtype: int32
likes 1
dtype: int32
likes 3345
dtype: int32
Epoch 1/30
1/19 [>.............................] - ETA: 6s - loss: 0.0834 - mean_absolute_error: 0.0834 19/19 [==============================] - 1s 10ms/step - loss: 0.0679 - mean_absolute_error: 0.0679 - val_loss: 0.0670 - val_mean_absolute_error: 0.0670
Epoch 2/30
1/19 [>.............................] - ETA: 0s - loss: 0.1142 - mean_absolute_error: 0.1142 19/19 [==============================] - 0s 2ms/step - loss: 0.0657 - mean_absolute_error: 0.0657 - val_loss: 0.0530 - val_mean_absolute_error: 0.0530
Epoch 3/30
1/19 [>.............................] - ETA: 0s - loss: 0.0940 - mean_absolute_error: 0.0940 19/19 [==============================] - 0s 2ms/step - loss: 0.0608 - mean_absolute_error: 0.0608 - val_loss: 0.0600 - val_mean_absolute_error: 0.0600
Epoch 4/30
1/19 [>.............................] - ETA: 0s - loss: 0.0524 - mean_absolute_error: 0.0524 19/19 [==============================] - 0s 2ms/step - loss: 0.0521 - mean_absolute_error: 0.0521 - val_loss: 0.0541 - val_mean_absolute_error: 0.0541
Epoch 5/30
1/19 [>.............................] - ETA: 0s - loss: 0.0440 - mean_absolute_error: 0.0440 19/19 [==============================] - 0s 2ms/step - loss: 0.0518 - mean_absolute_error: 0.0518 - val_loss: 0.0541 - val_mean_absolute_error: 0.0541
Epoch 6/30
1/19 [>.............................] - ETA: 0s - loss: 0.0576 - mean_absolute_error: 0.0576 19/19 [==============================] - 0s 2ms/step - loss: 0.0579 - mean_absolute_error: 0.0579 - val_loss: 0.0523 - val_mean_absolute_error: 0.0523
Epoch 7/30
1/19 [>.............................] - ETA: 0s - loss: 0.0310 - mean_absolute_error: 0.0310 19/19 [==============================] - 0s 2ms/step - loss: 0.0497 - mean_absolute_error: 0.0497 - val_loss: 0.0523 - val_mean_absolute_error: 0.0523
Epoch 8/30
1/19 [>.............................] - ETA: 0s - loss: 0.0628 - mean_absolute_error: 0.0628 19/19 [==============================] - 0s 2ms/step - loss: 0.0531 - mean_absolute_error: 0.0531 - val_loss: 0.0551 - val_mean_absolute_error: 0.0551
Epoch 9/30
1/19 [>.............................] - ETA: 0s - loss: 0.0425 - mean_absolute_error: 0.0425 19/19 [==============================] - 0s 2ms/step - loss: 0.0543 - mean_absolute_error: 0.0543 - val_loss: 0.0527 - val_mean_absolute_error: 0.0527
Epoch 10/30
1/19 [>.............................] - ETA: 0s - loss: 0.0560 - mean_absolute_error: 0.0560 19/19 [==============================] - 0s 2ms/step - loss: 0.0549 - mean_absolute_error: 0.0549 - val_loss: 0.0525 - val_mean_absolute_error: 0.0525
Epoch 11/30
1/19 [>.............................] - ETA: 0s - loss: 0.0391 - mean_absolute_error: 0.0391 19/19 [==============================] - 0s 2ms/step - loss: 0.0520 - mean_absolute_error: 0.0520 - val_loss: 0.0556 - val_mean_absolute_error: 0.0556
Epoch 12/30
1/19 [>.............................] - ETA: 0s - loss: 0.0417 - mean_absolute_error: 0.0417 19/19 [==============================] - 0s 2ms/step - loss: 0.0578 - mean_absolute_error: 0.0578 - val_loss: 0.0522 - val_mean_absolute_error: 0.0522
Epoch 13/30
1/19 [>.............................] - ETA: 0s - loss: 0.0834 - mean_absolute_error: 0.0834 19/19 [==============================] - 0s 2ms/step - loss: 0.0605 - mean_absolute_error: 0.0605 - val_loss: 0.0532 - val_mean_absolute_error: 0.0532
Epoch 14/30
1/19 [>.............................] - ETA: 0s - loss: 0.0430 - mean_absolute_error: 0.0430 19/19 [==============================] - 0s 2ms/step - loss: 0.0582 - mean_absolute_error: 0.0582 - val_loss: 0.0526 - val_mean_absolute_error: 0.0526
Epoch 15/30
1/19 [>.............................] - ETA: 0s - loss: 0.0506 - mean_absolute_error: 0.0506 19/19 [==============================] - 0s 2ms/step - loss: 0.0512 - mean_absolute_error: 0.0512 - val_loss: 0.0533 - val_mean_absolute_error: 0.0533
Epoch 16/30
1/19 [>.............................] - ETA: 0s - loss: 0.0402 - mean_absolute_error: 0.0402 19/19 [==============================] - 0s 2ms/step - loss: 0.0514 - mean_absolute_error: 0.0514 - val_loss: 0.0537 - val_mean_absolute_error: 0.0537
Epoch 17/30
1/19 [>.............................] - ETA: 0s - loss: 0.0247 - mean_absolute_error: 0.0247 19/19 [==============================] - 0s 2ms/step - loss: 0.0463 - mean_absolute_error: 0.0463 - val_loss: 0.0519 - val_mean_absolute_error: 0.0519
Epoch 18/30
1/19 [>.............................] - ETA: 0s - loss: 0.0401 - mean_absolute_error: 0.0401 19/19 [==============================] - 0s 2ms/step - loss: 0.0537 - mean_absolute_error: 0.0537 - val_loss: 0.0568 - val_mean_absolute_error: 0.0568
Epoch 19/30
1/19 [>.............................] - ETA: 0s - loss: 0.0930 - mean_absolute_error: 0.0930 19/19 [==============================] - 0s 2ms/step - loss: 0.0534 - mean_absolute_error: 0.0534 - val_loss: 0.0523 - val_mean_absolute_error: 0.0523
Epoch 20/30
1/19 [>.............................] - ETA: 0s - loss: 0.0631 - mean_absolute_error: 0.0631 19/19 [==============================] - 0s 2ms/step - loss: 0.0577 - mean_absolute_error: 0.0577 - val_loss: 0.0532 - val_mean_absolute_error: 0.0532
Epoch 21/30
1/19 [>.............................] - ETA: 0s - loss: 0.0524 - mean_absolute_error: 0.0524 19/19 [==============================] - 0s 2ms/step - loss: 0.0538 - mean_absolute_error: 0.0538 - val_loss: 0.0540 - val_mean_absolute_error: 0.0540
Epoch 22/30
1/19 [>.............................] - ETA: 0s - loss: 0.0435 - mean_absolute_error: 0.0435 19/19 [==============================] - 0s 2ms/step - loss: 0.0510 - mean_absolute_error: 0.0510 - val_loss: 0.0594 - val_mean_absolute_error: 0.0594
Epoch 23/30
1/19 [>.............................] - ETA: 0s - loss: 0.0324 - mean_absolute_error: 0.0324 19/19 [==============================] - 0s 2ms/step - loss: 0.0573 - mean_absolute_error: 0.0573 - val_loss: 0.0537 - val_mean_absolute_error: 0.0537
Epoch 24/30
1/19 [>.............................] - ETA: 0s - loss: 0.0354 - mean_absolute_error: 0.0354 19/19 [==============================] - 0s 2ms/step - loss: 0.0510 - mean_absolute_error: 0.0510 - val_loss: 0.0546 - val_mean_absolute_error: 0.0546
Epoch 25/30
1/19 [>.............................] - ETA: 0s - loss: 0.0474 - mean_absolute_error: 0.0474 19/19 [==============================] - 0s 2ms/step - loss: 0.0539 - mean_absolute_error: 0.0539 - val_loss: 0.0525 - val_mean_absolute_error: 0.0525
Epoch 26/30
1/19 [>.............................] - ETA: 0s - loss: 0.0928 - mean_absolute_error: 0.0928 19/19 [==============================] - 0s 2ms/step - loss: 0.0612 - mean_absolute_error: 0.0612 - val_loss: 0.0540 - val_mean_absolute_error: 0.0540
Epoch 27/30
1/19 [>.............................] - ETA: 0s - loss: 0.0582 - mean_absolute_error: 0.0582 19/19 [==============================] - 0s 2ms/step - loss: 0.0535 - mean_absolute_error: 0.0535 - val_loss: 0.0548 - val_mean_absolute_error: 0.0548
Epoch 28/30
1/19 [>.............................] - ETA: 0s - loss: 0.0415 - mean_absolute_error: 0.0415 19/19 [==============================] - 0s 2ms/step - loss: 0.0511 - mean_absolute_error: 0.0511 - val_loss: 0.0533 - val_mean_absolute_error: 0.0533
Epoch 29/30
1/19 [>.............................] - ETA: 0s - loss: 0.0491 - mean_absolute_error: 0.0491 19/19 [==============================] - 0s 3ms/step - loss: 0.0532 - mean_absolute_error: 0.0532 - val_loss: 0.0528 - val_mean_absolute_error: 0.0528
Epoch 30/30
1/19 [>.............................] - ETA: 0s - loss: 0.0475 - mean_absolute_error: 0.0475 19/19 [==============================] - 0s 2ms/step - loss: 0.0529 - mean_absolute_error: 0.0529 - val_loss: 0.0529 - val_mean_absolute_error: 0.0529
views 1
dtype: int32
views 488
dtype: int32
likes 1
dtype: int32
likes 3345
dtype: int32
114882.99377127373
114882.99377127373
114882.99377127373

3
my_runs/1/info.json Normal file
View File

@ -0,0 +1,3 @@
{
"prepare_message_ts": "2021-05-20 21:59:18.264490"
}

1
my_runs/1/metrics.json Normal file
View File

@ -0,0 +1 @@
{}

87
my_runs/1/run.json Normal file
View File

@ -0,0 +1,87 @@
{
"artifacts": [],
"command": "my_main",
"experiment": {
"base_dir": "C:\\Users\\karol\\PycharmProjects\\ium_434765",
"dependencies": [
"numpy==1.19.5",
"pandas==1.2.4",
"sacred==0.8.2",
"tensorflow==2.5.0rc1"
],
"mainfile": "neural_network.py",
"name": "sacred_scopes",
"repositories": [
{
"commit": "07479089e2d0bd86c8b0dd3bb005f7178078cc34",
"dirty": true,
"url": "https://git.wmi.amu.edu.pl/s434765/ium_434765.git"
},
{
"commit": "07479089e2d0bd86c8b0dd3bb005f7178078cc34",
"dirty": true,
"url": "https://git.wmi.amu.edu.pl/s434765/ium_434765.git"
}
],
"sources": [
[
"evaluate_network.py",
"_sources\\evaluate_network_6bc39a6cabbc78720ddbbd5b23f51cc3.py"
],
[
"neural_network.py",
"_sources\\neural_network_cdaa9eab635a60c87899a6eaac9e398e.py"
]
]
},
"heartbeat": "2021-05-20T19:59:22.263859",
"host": {
"ENV": {},
"cpu": "Unknown",
"gpus": {
"driver_version": "452.06",
"gpus": [
{
"model": "GeForce GTX 1650 Ti",
"persistence_mode": false,
"total_memory": 4096
}
]
},
"hostname": "DESKTOP-5PRPHO6",
"os": [
"Windows",
"Windows-10-10.0.19041-SP0"
],
"python_version": "3.9.2"
},
"meta": {
"command": "my_main",
"options": {
"--beat-interval": null,
"--capture": null,
"--comment": null,
"--debug": false,
"--enforce_clean": false,
"--file_storage": null,
"--force": false,
"--help": false,
"--loglevel": null,
"--mongo_db": null,
"--name": null,
"--pdb": false,
"--print-config": false,
"--priority": null,
"--queue": false,
"--s3": null,
"--sql": null,
"--tiny_db": null,
"--unobserved": false
}
},
"resources": [],
"result": null,
"start_time": "2021-05-20T19:59:18.258489",
"status": "COMPLETED",
"stop_time": "2021-05-20T19:59:22.263859"
}

4
my_runs/2/config.json Normal file
View File

@ -0,0 +1,4 @@
{
"epochs_amount": 30,
"seed": 535480662
}

78
my_runs/2/cout.txt Normal file
View File

@ -0,0 +1,78 @@
views 0
dtype: int32
views 488
dtype: int32
likes 1
dtype: int32
likes 3345
dtype: int32
Epoch 1/30
1/19 [>.............................] - ETA: 6s - loss: 0.1168 - mean_absolute_error: 0.1168 19/19 [==============================] - 1s 10ms/step - loss: 0.0788 - mean_absolute_error: 0.0788 - val_loss: 0.0639 - val_mean_absolute_error: 0.0639
Epoch 2/30
1/19 [>.............................] - ETA: 0s - loss: 0.0699 - mean_absolute_error: 0.0699 19/19 [==============================] - 0s 2ms/step - loss: 0.0622 - mean_absolute_error: 0.0622 - val_loss: 0.0589 - val_mean_absolute_error: 0.0589
Epoch 3/30
1/19 [>.............................] - ETA: 0s - loss: 0.0547 - mean_absolute_error: 0.0547 19/19 [==============================] - 0s 2ms/step - loss: 0.0566 - mean_absolute_error: 0.0566 - val_loss: 0.0542 - val_mean_absolute_error: 0.0542
Epoch 4/30
1/19 [>.............................] - ETA: 0s - loss: 0.0351 - mean_absolute_error: 0.0351 19/19 [==============================] - 0s 2ms/step - loss: 0.0534 - mean_absolute_error: 0.0534 - val_loss: 0.0524 - val_mean_absolute_error: 0.0524
Epoch 5/30
1/19 [>.............................] - ETA: 0s - loss: 0.0436 - mean_absolute_error: 0.0436 19/19 [==============================] - 0s 2ms/step - loss: 0.0562 - mean_absolute_error: 0.0562 - val_loss: 0.0560 - val_mean_absolute_error: 0.0560
Epoch 6/30
1/19 [>.............................] - ETA: 0s - loss: 0.0474 - mean_absolute_error: 0.0474 19/19 [==============================] - 0s 2ms/step - loss: 0.0513 - mean_absolute_error: 0.0513 - val_loss: 0.0533 - val_mean_absolute_error: 0.0533
Epoch 7/30
1/19 [>.............................] - ETA: 0s - loss: 0.0714 - mean_absolute_error: 0.0714 19/19 [==============================] - 0s 2ms/step - loss: 0.0562 - mean_absolute_error: 0.0562 - val_loss: 0.0519 - val_mean_absolute_error: 0.0519
Epoch 8/30
1/19 [>.............................] - ETA: 0s - loss: 0.0567 - mean_absolute_error: 0.0567 19/19 [==============================] - 0s 2ms/step - loss: 0.0535 - mean_absolute_error: 0.0535 - val_loss: 0.0526 - val_mean_absolute_error: 0.0526
Epoch 9/30
1/19 [>.............................] - ETA: 0s - loss: 0.0472 - mean_absolute_error: 0.0472 19/19 [==============================] - 0s 2ms/step - loss: 0.0571 - mean_absolute_error: 0.0571 - val_loss: 0.0559 - val_mean_absolute_error: 0.0559
Epoch 10/30
1/19 [>.............................] - ETA: 0s - loss: 0.0634 - mean_absolute_error: 0.0634 19/19 [==============================] - 0s 2ms/step - loss: 0.0528 - mean_absolute_error: 0.0528 - val_loss: 0.0527 - val_mean_absolute_error: 0.0527
Epoch 11/30
1/19 [>.............................] - ETA: 0s - loss: 0.0412 - mean_absolute_error: 0.0412 19/19 [==============================] - 0s 2ms/step - loss: 0.0529 - mean_absolute_error: 0.0529 - val_loss: 0.0542 - val_mean_absolute_error: 0.0542
Epoch 12/30
1/19 [>.............................] - ETA: 0s - loss: 0.0390 - mean_absolute_error: 0.0390 19/19 [==============================] - 0s 2ms/step - loss: 0.0496 - mean_absolute_error: 0.0496 - val_loss: 0.0596 - val_mean_absolute_error: 0.0596
Epoch 13/30
1/19 [>.............................] - ETA: 0s - loss: 0.0625 - mean_absolute_error: 0.0625 19/19 [==============================] - 0s 2ms/step - loss: 0.0545 - mean_absolute_error: 0.0545 - val_loss: 0.0560 - val_mean_absolute_error: 0.0560
Epoch 14/30
1/19 [>.............................] - ETA: 0s - loss: 0.0206 - mean_absolute_error: 0.0206 19/19 [==============================] - 0s 2ms/step - loss: 0.0542 - mean_absolute_error: 0.0542 - val_loss: 0.0542 - val_mean_absolute_error: 0.0542
Epoch 15/30
1/19 [>.............................] - ETA: 0s - loss: 0.0311 - mean_absolute_error: 0.0311 19/19 [==============================] - 0s 3ms/step - loss: 0.0486 - mean_absolute_error: 0.0486 - val_loss: 0.0527 - val_mean_absolute_error: 0.0527
Epoch 16/30
1/19 [>.............................] - ETA: 0s - loss: 0.0270 - mean_absolute_error: 0.0270 19/19 [==============================] - 0s 2ms/step - loss: 0.0477 - mean_absolute_error: 0.0477 - val_loss: 0.0558 - val_mean_absolute_error: 0.0558
Epoch 17/30
1/19 [>.............................] - ETA: 0s - loss: 0.0808 - mean_absolute_error: 0.0808 19/19 [==============================] - 0s 2ms/step - loss: 0.0563 - mean_absolute_error: 0.0563 - val_loss: 0.0546 - val_mean_absolute_error: 0.0546
Epoch 18/30
1/19 [>.............................] - ETA: 0s - loss: 0.0433 - mean_absolute_error: 0.0433 19/19 [==============================] - 0s 2ms/step - loss: 0.0499 - mean_absolute_error: 0.0499 - val_loss: 0.0551 - val_mean_absolute_error: 0.0551
Epoch 19/30
1/19 [>.............................] - ETA: 0s - loss: 0.0431 - mean_absolute_error: 0.0431 19/19 [==============================] - 0s 2ms/step - loss: 0.0524 - mean_absolute_error: 0.0524 - val_loss: 0.0530 - val_mean_absolute_error: 0.0530
Epoch 20/30
1/19 [>.............................] - ETA: 0s - loss: 0.0298 - mean_absolute_error: 0.0298 19/19 [==============================] - 0s 2ms/step - loss: 0.0490 - mean_absolute_error: 0.0490 - val_loss: 0.0517 - val_mean_absolute_error: 0.0517
Epoch 21/30
1/19 [>.............................] - ETA: 0s - loss: 0.0499 - mean_absolute_error: 0.0499 19/19 [==============================] - 0s 2ms/step - loss: 0.0555 - mean_absolute_error: 0.0555 - val_loss: 0.0557 - val_mean_absolute_error: 0.0557
Epoch 22/30
1/19 [>.............................] - ETA: 0s - loss: 0.0401 - mean_absolute_error: 0.0401 19/19 [==============================] - 0s 2ms/step - loss: 0.0524 - mean_absolute_error: 0.0524 - val_loss: 0.0602 - val_mean_absolute_error: 0.0602
Epoch 23/30
1/19 [>.............................] - ETA: 0s - loss: 0.0652 - mean_absolute_error: 0.0652 19/19 [==============================] - 0s 2ms/step - loss: 0.0596 - mean_absolute_error: 0.0596 - val_loss: 0.0567 - val_mean_absolute_error: 0.0567
Epoch 24/30
1/19 [>.............................] - ETA: 0s - loss: 0.0275 - mean_absolute_error: 0.0275 19/19 [==============================] - 0s 2ms/step - loss: 0.0562 - mean_absolute_error: 0.0562 - val_loss: 0.0531 - val_mean_absolute_error: 0.0531
Epoch 25/30
1/19 [>.............................] - ETA: 0s - loss: 0.0602 - mean_absolute_error: 0.0602 19/19 [==============================] - 0s 2ms/step - loss: 0.0576 - mean_absolute_error: 0.0576 - val_loss: 0.0540 - val_mean_absolute_error: 0.0540
Epoch 26/30
1/19 [>.............................] - ETA: 0s - loss: 0.0388 - mean_absolute_error: 0.0388 19/19 [==============================] - 0s 2ms/step - loss: 0.0542 - mean_absolute_error: 0.0542 - val_loss: 0.0555 - val_mean_absolute_error: 0.0555
Epoch 27/30
1/19 [>.............................] - ETA: 0s - loss: 0.0711 - mean_absolute_error: 0.0711 19/19 [==============================] - 0s 2ms/step - loss: 0.0560 - mean_absolute_error: 0.0560 - val_loss: 0.0538 - val_mean_absolute_error: 0.0538
Epoch 28/30
1/19 [>.............................] - ETA: 0s - loss: 0.0875 - mean_absolute_error: 0.0875 19/19 [==============================] - 0s 2ms/step - loss: 0.0614 - mean_absolute_error: 0.0614 - val_loss: 0.0535 - val_mean_absolute_error: 0.0535
Epoch 29/30
1/19 [>.............................] - ETA: 0s - loss: 0.0462 - mean_absolute_error: 0.0462 19/19 [==============================] - 0s 2ms/step - loss: 0.0544 - mean_absolute_error: 0.0544 - val_loss: 0.0562 - val_mean_absolute_error: 0.0562
Epoch 30/30
1/19 [>.............................] - ETA: 0s - loss: 0.0588 - mean_absolute_error: 0.0588 19/19 [==============================] - 0s 2ms/step - loss: 0.0582 - mean_absolute_error: 0.0582 - val_loss: 0.0593 - val_mean_absolute_error: 0.0593
views 1
dtype: int32
views 488
dtype: int32
likes 1
dtype: int32
likes 3345
dtype: int32
129787.96004765884
129787.96004765884

3
my_runs/2/info.json Normal file
View File

@ -0,0 +1,3 @@
{
"prepare_message_ts": "2021-05-20 22:01:49.105722"
}

13
my_runs/2/metrics.json Normal file
View File

@ -0,0 +1,13 @@
{
"training.metrics": {
"steps": [
0
],
"timestamps": [
"2021-05-20T20:01:53.071700"
],
"values": [
129787.96004765884
]
}
}

87
my_runs/2/run.json Normal file
View File

@ -0,0 +1,87 @@
{
"artifacts": [],
"command": "my_main",
"experiment": {
"base_dir": "C:\\Users\\karol\\PycharmProjects\\ium_434765",
"dependencies": [
"numpy==1.19.5",
"pandas==1.2.4",
"sacred==0.8.2",
"tensorflow==2.5.0rc1"
],
"mainfile": "neural_network.py",
"name": "sacred_scopes",
"repositories": [
{
"commit": "07479089e2d0bd86c8b0dd3bb005f7178078cc34",
"dirty": true,
"url": "https://git.wmi.amu.edu.pl/s434765/ium_434765.git"
},
{
"commit": "07479089e2d0bd86c8b0dd3bb005f7178078cc34",
"dirty": true,
"url": "https://git.wmi.amu.edu.pl/s434765/ium_434765.git"
}
],
"sources": [
[
"evaluate_network.py",
"_sources\\evaluate_network_6bc39a6cabbc78720ddbbd5b23f51cc3.py"
],
[
"neural_network.py",
"_sources\\neural_network_eca667942d0304c50d970a67f9012302.py"
]
]
},
"heartbeat": "2021-05-20T20:01:53.071700",
"host": {
"ENV": {},
"cpu": "Unknown",
"gpus": {
"driver_version": "452.06",
"gpus": [
{
"model": "GeForce GTX 1650 Ti",
"persistence_mode": false,
"total_memory": 4096
}
]
},
"hostname": "DESKTOP-5PRPHO6",
"os": [
"Windows",
"Windows-10-10.0.19041-SP0"
],
"python_version": "3.9.2"
},
"meta": {
"command": "my_main",
"options": {
"--beat-interval": null,
"--capture": null,
"--comment": null,
"--debug": false,
"--enforce_clean": false,
"--file_storage": null,
"--force": false,
"--help": false,
"--loglevel": null,
"--mongo_db": null,
"--name": null,
"--pdb": false,
"--print-config": false,
"--priority": null,
"--queue": false,
"--s3": null,
"--sql": null,
"--tiny_db": null,
"--unobserved": false
}
},
"resources": [],
"result": null,
"start_time": "2021-05-20T20:01:49.099728",
"status": "COMPLETED",
"stop_time": "2021-05-20T20:01:53.071700"
}

4
my_runs/3/config.json Normal file
View File

@ -0,0 +1,4 @@
{
"epochs_amount": 30,
"seed": 981983024
}

78
my_runs/3/cout.txt Normal file
View File

@ -0,0 +1,78 @@
views 0
dtype: int32
views 488
dtype: int32
likes 1
dtype: int32
likes 3345
dtype: int32
Epoch 1/30
1/19 [>.............................] - ETA: 7s - loss: 0.1234 - mean_absolute_error: 0.1234 19/19 [==============================] - 1s 10ms/step - loss: 0.0687 - mean_absolute_error: 0.0687 - val_loss: 0.0587 - val_mean_absolute_error: 0.0587
Epoch 2/30
1/19 [>.............................] - ETA: 0s - loss: 0.0764 - mean_absolute_error: 0.0764 19/19 [==============================] - 0s 2ms/step - loss: 0.0583 - mean_absolute_error: 0.0583 - val_loss: 0.0541 - val_mean_absolute_error: 0.0541
Epoch 3/30
1/19 [>.............................] - ETA: 0s - loss: 0.0781 - mean_absolute_error: 0.0781 19/19 [==============================] - 0s 2ms/step - loss: 0.0595 - mean_absolute_error: 0.0595 - val_loss: 0.0572 - val_mean_absolute_error: 0.0572
Epoch 4/30
1/19 [>.............................] - ETA: 0s - loss: 0.0564 - mean_absolute_error: 0.0564 19/19 [==============================] - 0s 2ms/step - loss: 0.0592 - mean_absolute_error: 0.0592 - val_loss: 0.0541 - val_mean_absolute_error: 0.0541
Epoch 5/30
1/19 [>.............................] - ETA: 0s - loss: 0.0608 - mean_absolute_error: 0.0608 19/19 [==============================] - 0s 2ms/step - loss: 0.0552 - mean_absolute_error: 0.0552 - val_loss: 0.0524 - val_mean_absolute_error: 0.0524
Epoch 6/30
1/19 [>.............................] - ETA: 0s - loss: 0.0346 - mean_absolute_error: 0.0346 19/19 [==============================] - 0s 2ms/step - loss: 0.0510 - mean_absolute_error: 0.0510 - val_loss: 0.0544 - val_mean_absolute_error: 0.0544
Epoch 7/30
1/19 [>.............................] - ETA: 0s - loss: 0.0569 - mean_absolute_error: 0.0569 19/19 [==============================] - 0s 2ms/step - loss: 0.0570 - mean_absolute_error: 0.0570 - val_loss: 0.0521 - val_mean_absolute_error: 0.0521
Epoch 8/30
1/19 [>.............................] - ETA: 0s - loss: 0.0565 - mean_absolute_error: 0.0565 19/19 [==============================] - 0s 2ms/step - loss: 0.0558 - mean_absolute_error: 0.0558 - val_loss: 0.0542 - val_mean_absolute_error: 0.0542
Epoch 9/30
1/19 [>.............................] - ETA: 0s - loss: 0.0829 - mean_absolute_error: 0.0829 19/19 [==============================] - 0s 2ms/step - loss: 0.0563 - mean_absolute_error: 0.0563 - val_loss: 0.0535 - val_mean_absolute_error: 0.0535
Epoch 10/30
1/19 [>.............................] - ETA: 0s - loss: 0.0298 - mean_absolute_error: 0.0298 19/19 [==============================] - 0s 2ms/step - loss: 0.0509 - mean_absolute_error: 0.0509 - val_loss: 0.0520 - val_mean_absolute_error: 0.0520
Epoch 11/30
1/19 [>.............................] - ETA: 0s - loss: 0.0376 - mean_absolute_error: 0.0376 19/19 [==============================] - 0s 2ms/step - loss: 0.0546 - mean_absolute_error: 0.0546 - val_loss: 0.0557 - val_mean_absolute_error: 0.0557
Epoch 12/30
1/19 [>.............................] - ETA: 0s - loss: 0.0577 - mean_absolute_error: 0.0577 19/19 [==============================] - 0s 2ms/step - loss: 0.0567 - mean_absolute_error: 0.0567 - val_loss: 0.0521 - val_mean_absolute_error: 0.0521
Epoch 13/30
1/19 [>.............................] - ETA: 0s - loss: 0.0537 - mean_absolute_error: 0.0537 19/19 [==============================] - 0s 2ms/step - loss: 0.0552 - mean_absolute_error: 0.0552 - val_loss: 0.0556 - val_mean_absolute_error: 0.0556
Epoch 14/30
1/19 [>.............................] - ETA: 0s - loss: 0.0696 - mean_absolute_error: 0.0696 19/19 [==============================] - 0s 2ms/step - loss: 0.0616 - mean_absolute_error: 0.0616 - val_loss: 0.0571 - val_mean_absolute_error: 0.0571
Epoch 15/30
1/19 [>.............................] - ETA: 0s - loss: 0.0726 - mean_absolute_error: 0.0726 19/19 [==============================] - 0s 2ms/step - loss: 0.0556 - mean_absolute_error: 0.0556 - val_loss: 0.0531 - val_mean_absolute_error: 0.0531
Epoch 16/30
1/19 [>.............................] - ETA: 0s - loss: 0.0448 - mean_absolute_error: 0.0448 19/19 [==============================] - 0s 2ms/step - loss: 0.0533 - mean_absolute_error: 0.0533 - val_loss: 0.0562 - val_mean_absolute_error: 0.0562
Epoch 17/30
1/19 [>.............................] - ETA: 0s - loss: 0.0458 - mean_absolute_error: 0.0458 19/19 [==============================] - 0s 2ms/step - loss: 0.0553 - mean_absolute_error: 0.0553 - val_loss: 0.0558 - val_mean_absolute_error: 0.0558
Epoch 18/30
1/19 [>.............................] - ETA: 0s - loss: 0.0547 - mean_absolute_error: 0.0547 19/19 [==============================] - 0s 2ms/step - loss: 0.0590 - mean_absolute_error: 0.0590 - val_loss: 0.0561 - val_mean_absolute_error: 0.0561
Epoch 19/30
1/19 [>.............................] - ETA: 0s - loss: 0.0402 - mean_absolute_error: 0.0402 19/19 [==============================] - 0s 2ms/step - loss: 0.0579 - mean_absolute_error: 0.0579 - val_loss: 0.0554 - val_mean_absolute_error: 0.0554
Epoch 20/30
1/19 [>.............................] - ETA: 0s - loss: 0.0614 - mean_absolute_error: 0.0614 19/19 [==============================] - 0s 2ms/step - loss: 0.0558 - mean_absolute_error: 0.0558 - val_loss: 0.0539 - val_mean_absolute_error: 0.0539
Epoch 21/30
1/19 [>.............................] - ETA: 0s - loss: 0.0492 - mean_absolute_error: 0.0492 19/19 [==============================] - 0s 2ms/step - loss: 0.0525 - mean_absolute_error: 0.0525 - val_loss: 0.0540 - val_mean_absolute_error: 0.0540
Epoch 22/30
1/19 [>.............................] - ETA: 0s - loss: 0.0554 - mean_absolute_error: 0.0554 19/19 [==============================] - 0s 2ms/step - loss: 0.0595 - mean_absolute_error: 0.0595 - val_loss: 0.0533 - val_mean_absolute_error: 0.0533
Epoch 23/30
1/19 [>.............................] - ETA: 0s - loss: 0.0664 - mean_absolute_error: 0.0664 19/19 [==============================] - 0s 2ms/step - loss: 0.0533 - mean_absolute_error: 0.0533 - val_loss: 0.0518 - val_mean_absolute_error: 0.0518
Epoch 24/30
1/19 [>.............................] - ETA: 0s - loss: 0.0282 - mean_absolute_error: 0.0282 19/19 [==============================] - 0s 2ms/step - loss: 0.0471 - mean_absolute_error: 0.0471 - val_loss: 0.0517 - val_mean_absolute_error: 0.0517
Epoch 25/30
1/19 [>.............................] - ETA: 0s - loss: 0.0456 - mean_absolute_error: 0.0456 19/19 [==============================] - 0s 2ms/step - loss: 0.0473 - mean_absolute_error: 0.0473 - val_loss: 0.0536 - val_mean_absolute_error: 0.0536
Epoch 26/30
1/19 [>.............................] - ETA: 0s - loss: 0.0668 - mean_absolute_error: 0.0668 19/19 [==============================] - 0s 2ms/step - loss: 0.0571 - mean_absolute_error: 0.0571 - val_loss: 0.0532 - val_mean_absolute_error: 0.0532
Epoch 27/30
1/19 [>.............................] - ETA: 0s - loss: 0.0602 - mean_absolute_error: 0.0602 19/19 [==============================] - 0s 2ms/step - loss: 0.0558 - mean_absolute_error: 0.0558 - val_loss: 0.0520 - val_mean_absolute_error: 0.0520
Epoch 28/30
1/19 [>.............................] - ETA: 0s - loss: 0.0631 - mean_absolute_error: 0.0631 19/19 [==============================] - 0s 2ms/step - loss: 0.0557 - mean_absolute_error: 0.0557 - val_loss: 0.0528 - val_mean_absolute_error: 0.0528
Epoch 29/30
1/19 [>.............................] - ETA: 0s - loss: 0.0601 - mean_absolute_error: 0.0601 19/19 [==============================] - 0s 2ms/step - loss: 0.0526 - mean_absolute_error: 0.0526 - val_loss: 0.0521 - val_mean_absolute_error: 0.0521
Epoch 30/30
1/19 [>.............................] - ETA: 0s - loss: 0.0508 - mean_absolute_error: 0.0508 19/19 [==============================] - 0s 2ms/step - loss: 0.0542 - mean_absolute_error: 0.0542 - val_loss: 0.0534 - val_mean_absolute_error: 0.0534
views 1
dtype: int32
views 488
dtype: int32
likes 1
dtype: int32
likes 3345
dtype: int32
114831.63920784603
114831.63920784603

3
my_runs/3/info.json Normal file
View File

@ -0,0 +1,3 @@
{
"prepare_message_ts": "2021-05-20 22:06:00.289863"
}

13
my_runs/3/metrics.json Normal file
View File

@ -0,0 +1,13 @@
{
"training.metrics": {
"steps": [
0
],
"timestamps": [
"2021-05-20T20:06:03.338305"
],
"values": [
114831.63920784603
]
}
}

79
my_runs/3/run.json Normal file
View File

@ -0,0 +1,79 @@
{
"artifacts": [],
"command": "my_main",
"experiment": {
"base_dir": "C:\\Users\\karol\\PycharmProjects\\ium_434765",
"dependencies": [
"numpy==1.19.5",
"pandas==1.2.4",
"sacred==0.8.2",
"scikit-learn==0.24.1",
"tensorflow==2.5.0rc1"
],
"mainfile": "neural_network.py",
"name": "sacred_scopes",
"repositories": [
{
"commit": "b0346d0b62846839e512344b20a566135e07a4b2",
"dirty": true,
"url": "https://git.wmi.amu.edu.pl/s434765/ium_434765.git"
}
],
"sources": [
[
"neural_network.py",
"_sources\\neural_network_33e5177d0655bf5fef22fcd226db36b1.py"
]
]
},
"heartbeat": "2021-05-20T20:06:03.339305",
"host": {
"ENV": {},
"cpu": "Unknown",
"gpus": {
"driver_version": "452.06",
"gpus": [
{
"model": "GeForce GTX 1650 Ti",
"persistence_mode": false,
"total_memory": 4096
}
]
},
"hostname": "DESKTOP-5PRPHO6",
"os": [
"Windows",
"Windows-10-10.0.19041-SP0"
],
"python_version": "3.9.2"
},
"meta": {
"command": "my_main",
"options": {
"--beat-interval": null,
"--capture": null,
"--comment": null,
"--debug": false,
"--enforce_clean": false,
"--file_storage": null,
"--force": false,
"--help": false,
"--loglevel": null,
"--mongo_db": null,
"--name": null,
"--pdb": false,
"--print-config": false,
"--priority": null,
"--queue": false,
"--s3": null,
"--sql": null,
"--tiny_db": null,
"--unobserved": false
}
},
"resources": [],
"result": null,
"start_time": "2021-05-20T20:06:00.285864",
"status": "COMPLETED",
"stop_time": "2021-05-20T20:06:03.339305"
}

53
my_runs/_sources/evaluate_network_6bc39a6cabbc78720ddbbd5b23f51cc3.py Normal file
View File

@ -0,0 +1,53 @@
import pandas as pd
import numpy as np
from sklearn.metrics import mean_squared_error
from tensorflow import keras
import matplotlib.pyplot as plt
def evaluate_model():
model = keras.models.load_model('model')
data = pd.read_csv("data_dev", sep=',', error_bad_lines=False,
skip_blank_lines=True, nrows=527, names=["video_id", "last_trending_date",
"publish_date", "publish_hour", "category_id",
"channel_title", "views", "likes", "dislikes",
"comment_count"]).dropna()
X_test = data.loc[:, data.columns == "views"].astype(int)
y_test = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X_test)
max_val_sub = np.max(X_test)
X_test = (X_test - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y_test)
max_val_like = np.max(y_test)
print(min_val_like)
print(max_val_like)
prediction = model.predict(X_test)
prediction_denormalized = []
for pred in prediction:
denorm = pred[0] * (max_val_like[0] - min_val_like[0]) + min_val_like[0]
prediction_denormalized.append(denorm)
f = open("predictions.txt", "w")
for (pred, test) in zip(prediction_denormalized, y_test.values):
f.write("predicted: %s expected: %s\n" % (str(pred), str(test[0])))
error = mean_squared_error(y_test, prediction_denormalized)
print(error)
with open("rmse.txt", "a") as file:
file.write(str(error) + "\n")
with open("rmse.txt", "r") as file:
lines = file.readlines()
plt.plot(range(len(lines)), [line[:-2] for line in lines])
plt.tight_layout()
plt.ylabel('RMSE')
plt.xlabel('evaluation no')
plt.savefig('evaluation.png')
return error

108
my_runs/_sources/neural_network_33e5177d0655bf5fef22fcd226db36b1.py Normal file
View File

@ -0,0 +1,108 @@
from datetime import datetime
import pandas as pd
import numpy as np
from sacred.observers import FileStorageObserver, MongoObserver
from sacred import Experiment
from sklearn.metrics import mean_squared_error
from tensorflow import keras
ex = Experiment("sacred_scopes", interactive=True)
# ex.observers.append(MongoObserver(url='mongodb://mongo_user:mongo_password_IUM_2021@172.17.0.1:27017',
# db_name='sacred'))
ex.observers.append(FileStorageObserver('my_runs'))
@ex.config
def my_config():
epochs_amount = 30
def normalize_data(data):
return (data - np.min(data)) / (np.max(data) - np.min(data))
@ex.capture
def prepare_model(epochs_amount, _run):
_run.info["prepare_message_ts"] = str(datetime.now())
data = pd.read_csv("data_train", sep=',', skip_blank_lines=True, nrows=1087, error_bad_lines=False,
names=["vipip install sacreddeo_id", "last_trending_date", "publish_date", "publish_hour",
"category_id",
"channel_title", "views", "likes", "dislikes", "comment_count"]).dropna()
X = data.loc[:, data.columns == "views"].astype(int)
y = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X)
max_val_sub = np.max(X)
X = (X - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y)
max_val_like = np.max(y)
y = (y - min_val_like) / (max_val_like - min_val_like)
print(min_val_like)
print(max_val_like)
model = keras.Sequential([
keras.layers.Dense(512, input_dim=X.shape[1], activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(128, activation='relu'),
keras.layers.Dense(1, activation='linear'),
])
model.compile(loss='mean_absolute_error', optimizer="Adam", metrics=['mean_absolute_error'])
model.fit(X, y, epochs=int(epochs_amount), validation_split=0.3)
data = pd.read_csv("data_dev", sep=',', error_bad_lines=False,
skip_blank_lines=True, nrows=527, names=["video_id", "last_trending_date",
"publish_date", "publish_hour", "category_id",
"channel_title", "views", "likes", "dislikes",
"comment_count"]).dropna()
X_test = data.loc[:, data.columns == "views"].astype(int)
y_test = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X_test)
max_val_sub = np.max(X_test)
X_test = (X_test - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y_test)
max_val_like = np.max(y_test)
print(min_val_like)
print(max_val_like)
prediction = model.predict(X_test)
prediction_denormalized = []
for pred in prediction:
denorm = pred[0] * (max_val_like[0] - min_val_like[0]) + min_val_like[0]
prediction_denormalized.append(denorm)
f = open("predictions.txt", "w")
for (pred, test) in zip(prediction_denormalized, y_test.values):
f.write("predicted: %s expected: %s\n" % (str(pred), str(test[0])))
error = mean_squared_error(y_test, prediction_denormalized)
print(error)
model.save('model')
_run.log_scalar("training.metrics", error)
return error
@ex.main
def my_main(epochs_amount):
print(prepare_model())
ex.run()
ex.add_artifact("model.pb")

79
my_runs/_sources/neural_network_cdaa9eab635a60c87899a6eaac9e398e.py Normal file
View File

@ -0,0 +1,79 @@
from datetime import datetime
import pandas as pd
import numpy as np
from sacred.observers import FileStorageObserver, MongoObserver
from sacred import Experiment
from tensorflow import keras
import sys
from evaluate_network import evaluate_model
ex = Experiment("sacred_scopes", interactive=True)
# ex.observers.append(MongoObserver(url='mongodb://mongo_user:mongo_password_IUM_2021@172.17.0.1:27017',
# db_name='sacred'))
ex.observers.append(FileStorageObserver('my_runs'))
@ex.config
def my_config():
epochs_amount = 30
def normalize_data(data):
return (data - np.min(data)) / (np.max(data) - np.min(data))
@ex.capture
def prepare_model(epochs_amount, _run):
_run.info["prepare_message_ts"] = str(datetime.now())
data = pd.read_csv("data_train", sep=',', skip_blank_lines=True, nrows=1087, error_bad_lines=False,
names=["vipip install sacreddeo_id", "last_trending_date", "publish_date", "publish_hour",
"category_id",
"channel_title", "views", "likes", "dislikes", "comment_count"]).dropna()
X = data.loc[:, data.columns == "views"].astype(int)
y = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X)
max_val_sub = np.max(X)
X = (X - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y)
max_val_like = np.max(y)
y = (y - min_val_like) / (max_val_like - min_val_like)
print(min_val_like)
print(max_val_like)
model = keras.Sequential([
keras.layers.Dense(512, input_dim=X.shape[1], activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(128, activation='relu'),
keras.layers.Dense(1, activation='linear'),
])
model.compile(loss='mean_absolute_error', optimizer="Adam", metrics=['mean_absolute_error'])
model.fit(X, y, epochs=int(epochs_amount), validation_split=0.3)
model.save('model')
metrics = evaluate_model()
print(metrics)
return metrics
@ex.main
def my_main(epochs_amount):
print(prepare_model())
ex.run()
ex.add_artifact("model.pb")

78
my_runs/_sources/neural_network_eca667942d0304c50d970a67f9012302.py Normal file
View File

@ -0,0 +1,78 @@
from datetime import datetime
import pandas as pd
import numpy as np
from sacred.observers import FileStorageObserver, MongoObserver
from sacred import Experiment
from tensorflow import keras
import sys
from evaluate_network import evaluate_model
ex = Experiment("sacred_scopes", interactive=True)
# ex.observers.append(MongoObserver(url='mongodb://mongo_user:mongo_password_IUM_2021@172.17.0.1:27017',
# db_name='sacred'))
ex.observers.append(FileStorageObserver('my_runs'))
@ex.config
def my_config():
epochs_amount = 30
def normalize_data(data):
return (data - np.min(data)) / (np.max(data) - np.min(data))
@ex.capture
def prepare_model(epochs_amount, _run):
_run.info["prepare_message_ts"] = str(datetime.now())
data = pd.read_csv("data_train", sep=',', skip_blank_lines=True, nrows=1087, error_bad_lines=False,
names=["vipip install sacreddeo_id", "last_trending_date", "publish_date", "publish_hour",
"category_id",
"channel_title", "views", "likes", "dislikes", "comment_count"]).dropna()
X = data.loc[:, data.columns == "views"].astype(int)
y = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X)
max_val_sub = np.max(X)
X = (X - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y)
max_val_like = np.max(y)
y = (y - min_val_like) / (max_val_like - min_val_like)
print(min_val_like)
print(max_val_like)
model = keras.Sequential([
keras.layers.Dense(512, input_dim=X.shape[1], activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(128, activation='relu'),
keras.layers.Dense(1, activation='linear'),
])
model.compile(loss='mean_absolute_error', optimizer="Adam", metrics=['mean_absolute_error'])
model.fit(X, y, epochs=int(epochs_amount), validation_split=0.3)
model.save('model')
metrics = evaluate_model()
_run.log_scalar("training.metrics", metrics)
return metrics
@ex.main
def my_main(epochs_amount):
print(prepare_model())
ex.run()
ex.add_artifact("model.pb")

164
neural_network.py
View File

@ -1,94 +1,108 @@
from datetime import datetime
import pandas as pd
import numpy as np
from sacred.observers import FileStorageObserver, MongoObserver
from sacred import Experiment
from sklearn.metrics import mean_squared_error
from tensorflow import keras
import sys
ex = Experiment("sacred_scopes", interactive=True)
ex.observers.append(MongoObserver(url='mongodb://mongo_user:mongo_password_IUM_2021@172.17.0.1:27017',
db_name='sacred'))
ex.observers.append(FileStorageObserver('my_runs'))
@ex.config
def my_config():
epochs_amount = 30
def normalize_data(data):
return (data - np.min(data)) / (np.max(data) - np.min(data))
data = pd.read_csv("data_train", sep=',', skip_blank_lines=True, nrows=1087, error_bad_lines=False,
names=["video_id", "last_trending_date", "publish_date", "publish_hour", "category_id",
"channel_title", "views", "likes", "dislikes", "comment_count", "comments_disabled",
"ratings_disabled", "tag_appeared_in_title_count", "tag_appeared_in_title", "title",
"tags", "description", "trend_day_count", "trend_publish_diff", "trend_tag_highest",
"trend_tag_total", "tags_count", "subscriber"]).dropna()
@ex.capture
def prepare_model(epochs_amount, _run):
_run.info["prepare_message_ts"] = str(datetime.now())
X = []
for datum in data[["views"]].values:
try:
X.append(int(datum))
except:
print(datum)
X = pd.DataFrame(X)
data = pd.read_csv("data_train", sep=',', skip_blank_lines=True, nrows=1087, error_bad_lines=False,
names=["vipip install sacreddeo_id", "last_trending_date", "publish_date", "publish_hour",
"category_id",
"channel_title", "views", "likes", "dislikes", "comment_count"]).dropna()
y = []
for datum in data[["likes"]].values:
try:
y.append(int(datum))
except:
print(datum)
y = pd.DataFrame(y)
X = data.loc[:, data.columns == "views"].astype(int)
y = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X)
max_val_sub = np.max(X)
X = (X - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_sub = np.min(X)
max_val_sub = np.max(X)
X = (X - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y)
max_val_like = np.max(y)
y = (y - min_val_like) / (max_val_like - min_val_like)
min_val_like = np.min(y)
max_val_like = np.max(y)
y = (y - min_val_like) / (max_val_like - min_val_like)
print(min_val_like)
print(max_val_like)
print(min_val_like)
print(max_val_like)
model = keras.Sequential([
keras.layers.Dense(512, input_dim=X.shape[1], activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(128, activation='relu'),
keras.layers.Dense(1, activation='linear'),
])
model.compile(loss='mean_absolute_error', optimizer="Adam", metrics=['mean_absolute_error'])
model.fit(X, y, epochs=int(epochs_amount), validation_split=0.3)
data = pd.read_csv("data_dev", sep=',', error_bad_lines=False,
skip_blank_lines=True, nrows=527, names=["video_id", "last_trending_date",
"publish_date", "publish_hour", "category_id",
"channel_title", "views", "likes", "dislikes",
"comment_count"]).dropna()
X_test = data.loc[:, data.columns == "views"].astype(int)
y_test = data.loc[:, data.columns == "likes"].astype(int)
min_val_sub = np.min(X_test)
max_val_sub = np.max(X_test)
X_test = (X_test - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y_test)
max_val_like = np.max(y_test)
print(min_val_like)
print(max_val_like)
prediction = model.predict(X_test)
prediction_denormalized = []
for pred in prediction:
denorm = pred[0] * (max_val_like[0] - min_val_like[0]) + min_val_like[0]
prediction_denormalized.append(denorm)
f = open("predictions.txt", "w")
for (pred, test) in zip(prediction_denormalized, y_test.values):
f.write("predicted: %s expected: %s\n" % (str(pred), str(test[0])))
error = mean_squared_error(y_test, prediction_denormalized)
print(error)
model.save('model')
_run.log_scalar("training.metrics", error)
return error
model = keras.Sequential([
keras.layers.Dense(512,input_dim = X.shape[1], activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(256, activation='relu'),
keras.layers.Dense(128, activation='relu'),
keras.layers.Dense(1,activation='linear'),
])
@ex.main
def my_main(epochs_amount):
print(prepare_model())
model.compile(loss='mean_absolute_error', optimizer="Adam", metrics=['mean_absolute_error'])
model.fit(X, y, epochs=int(sys.argv[1]), validation_split = 0.3)
data = pd.read_csv("data_dev", sep=',', error_bad_lines=False,
skip_blank_lines=True, nrows=527, names=["video_id", "last_trending_date",
"publish_date", "publish_hour", "category_id",
"channel_title", "views", "likes", "dislikes",
"comment_count"]).dropna()
X_test = data.loc[:,data.columns == "views"].astype(int)
y_test = data.loc[:,data.columns == "likes"].astype(int)
min_val_sub = np.min(X_test)
max_val_sub = np.max(X_test)
X_test = (X_test - min_val_sub) / (max_val_sub - min_val_sub)
print(min_val_sub)
print(max_val_sub)
min_val_like = np.min(y_test)
max_val_like = np.max(y_test)
print(min_val_like)
print(max_val_like)
prediction = model.predict(X_test)
prediction_denormalized = []
for pred in prediction:
denorm = pred[0] * (max_val_like[0] - min_val_like[0]) + min_val_like[0]
prediction_denormalized.append(denorm)
f = open("predictions.txt", "w")
for (pred, test) in zip(prediction_denormalized, y_test.values):
f.write("predicted: %s expected: %s\n" % (str(pred), str(test[0])))
error = mean_squared_error(y_test, prediction_denormalized)
print(error)
model.save('model')
ex.run()
ex.add_artifact("model.pb")

844
predictions.txt
View File

@ -1,422 +1,422 @@
predicted: 600.7225935459137 expected: 617
predicted: 75.86929032206535 expected: 172
predicted: 2123.512541770935 expected: 611
predicted: 219.68049824237823 expected: 269
predicted: 2123.512541770935 expected: 1095
predicted: 219.68049824237823 expected: 68
predicted: 14.693757019937038 expected: 5
predicted: 600.7225935459137 expected: 986
predicted: 320.3024938106537 expected: 262
predicted: 545.3180384635925 expected: 817
predicted: 355.80036973953247 expected: 197
predicted: 288.7730129957199 expected: 264
predicted: 486.0374472141266 expected: 830
predicted: 2123.512541770935 expected: 1415
predicted: 482.08706545829773 expected: 134
predicted: 59.75554421544075 expected: 58
predicted: 109.09023916721344 expected: 93
predicted: 695.9524660110474 expected: 830
predicted: 2123.512541770935 expected: 1207
predicted: 493.93860936164856 expected: 269
predicted: 672.1024808883667 expected: 558
predicted: 1915.4331912994385 expected: 1558
predicted: 81.1770147383213 expected: 37
predicted: 699.9274635314941 expected: 364
predicted: 723.9729795455933 expected: 1020
predicted: 12.627800181508064 expected: 11
predicted: 537.4120926856995 expected: 225
predicted: 699.9274635314941 expected: 228
predicted: 1915.4331912994385 expected: 1184
predicted: 517.6473278999329 expected: 370
predicted: 91.83225864171982 expected: 68
predicted: 280.89013826847076 expected: 201
predicted: 2123.512541770935 expected: 1113
predicted: 517.6473278999329 expected: 496
predicted: 91.83225864171982 expected: 43
predicted: 115.16775435209274 expected: 59
predicted: 91.83225864171982 expected: 60
predicted: 121.46323615312576 expected: 78
predicted: 383.4159938097 expected: 263
predicted: 493.93860936164856 expected: 400
predicted: 2123.512541770935 expected: 1256
predicted: 91.83225864171982 expected: 23
predicted: 2123.512541770935 expected: 3345
predicted: 139.51090186834335 expected: 98
predicted: 774.2462439537048 expected: 238
predicted: 155.66449910402298 expected: 69
predicted: 537.4120926856995 expected: 170
predicted: 48.527550891041756 expected: 31
predicted: 165.96958500146866 expected: 102
predicted: 2123.512541770935 expected: 1070
predicted: 249.08315706253052 expected: 96
predicted: 644.3031105995178 expected: 387
predicted: 70.57235398888588 expected: 25
predicted: 624.4822378158569 expected: 574
predicted: 545.3180384635925 expected: 165
predicted: 596.7626445293427 expected: 765
predicted: 695.9524660110474 expected: 599
predicted: 672.1024808883667 expected: 906
predicted: 133.6499207019806 expected: 71
predicted: 612.6023907661438 expected: 433
predicted: 612.6023907661438 expected: 152
predicted: 189.14595025777817 expected: 116
predicted: 21.161466643214226 expected: 19
predicted: 17.829518601298332 expected: 24
predicted: 165.96958500146866 expected: 97
predicted: 189.14595025777817 expected: 49
predicted: 774.2462439537048 expected: 291
predicted: 2123.512541770935 expected: 2816
predicted: 537.4120926856995 expected: 152
predicted: 695.9524660110474 expected: 1033
predicted: 672.1024808883667 expected: 740
predicted: 109.09023916721344 expected: 32
predicted: 170.8152904510498 expected: 74
predicted: 774.2462439537048 expected: 453
predicted: 537.4120926856995 expected: 219
predicted: 232.32288587093353 expected: 82
predicted: 257.2054126262665 expected: 72
predicted: 86.49207654595375 expected: 109
predicted: 600.7225935459137 expected: 567
predicted: 434.70216703414917 expected: 389
predicted: 109.09023916721344 expected: 70
predicted: 2123.512541770935 expected: 987
predicted: 2123.512541770935 expected: 1812
predicted: 699.9274635314941 expected: 169
predicted: 387.36106872558594 expected: 270
predicted: 43.05705846846104 expected: 33
predicted: 211.13050639629364 expected: 75
predicted: 2123.512541770935 expected: 1424
predicted: 21.161469757556915 expected: 39
predicted: 43.05705846846104 expected: 49
predicted: 48.527550891041756 expected: 141
predicted: 28.208953201770782 expected: 24
predicted: 232.32288587093353 expected: 91
predicted: 332.129297375679 expected: 101
predicted: 723.9730293750763 expected: 401
predicted: 723.9730293750763 expected: 570
predicted: 236.52261435985565 expected: 106
predicted: 32.85735569894314 expected: 43
predicted: 774.2462439537048 expected: 439
predicted: 1915.4331912994385 expected: 1220
predicted: 91.83225864171982 expected: 82
predicted: 11.343863356858492 expected: 5
predicted: 699.9274635314941 expected: 314
predicted: 517.6473278999329 expected: 180
predicted: 97.1587750017643 expected: 88
predicted: 2123.512541770935 expected: 1642
predicted: 600.7225935459137 expected: 417
predicted: 414.9767426252365 expected: 346
predicted: 14.693757019937038 expected: 10
predicted: 249.08315706253052 expected: 93
predicted: 48.527550891041756 expected: 26
predicted: 180.16424822807312 expected: 41
predicted: 2123.512541770935 expected: 505
predicted: 11.343863356858492 expected: 11
predicted: 2123.512541770935 expected: 929
predicted: 545.3180384635925 expected: 877
predicted: 155.66449910402298 expected: 370
predicted: 12.627800181508064 expected: 28
predicted: 2123.512541770935 expected: 1085
predicted: 1915.4331912994385 expected: 654
predicted: 97.1587750017643 expected: 59
predicted: 280.89013826847076 expected: 259
predicted: 497.8892402648926 expected: 165
predicted: 11.343863356858492 expected: 1
predicted: 28.208953201770782 expected: 26
predicted: 232.32288587093353 expected: 399
predicted: 596.7625946998596 expected: 155
predicted: 273.00728845596313 expected: 158
predicted: 644.3031105995178 expected: 782
predicted: 276.9487009048462 expected: 376
predicted: 11.343863356858492 expected: 1
predicted: 336.07454669475555 expected: 116
predicted: 414.9767426252365 expected: 628
predicted: 2123.512541770935 expected: 1897
predicted: 202.41596513986588 expected: 76
predicted: 830.9600687026978 expected: 450
predicted: 387.36106872558594 expected: 272
predicted: 537.4120926856995 expected: 149
predicted: 1915.4331912994385 expected: 1069
predicted: 316.3615049123764 expected: 286
predicted: 273.00728845596313 expected: 526
predicted: 81.1770147383213 expected: 29
predicted: 414.9767426252365 expected: 373
predicted: 232.32288587093353 expected: 481
predicted: 219.68049824237823 expected: 74
predicted: 70.57235398888588 expected: 54
predicted: 240.71561586856842 expected: 102
predicted: 11.343863356858492 expected: 22
predicted: 2123.512541770935 expected: 1360
predicted: 644.3031105995178 expected: 324
predicted: 596.7626445293427 expected: 973
predicted: 600.7225935459137 expected: 407
predicted: 328.18449652194977 expected: 317
predicted: 21.161469757556915 expected: 76
predicted: 624.4822378158569 expected: 688
predicted: 65.26819238066673 expected: 44
predicted: 493.93860936164856 expected: 239
predicted: 644.3031105995178 expected: 344
predicted: 624.482337474823 expected: 688
predicted: 537.4120926856995 expected: 442
predicted: 198.0343341231346 expected: 157
predicted: 2123.512541770935 expected: 1193
predicted: 127.64895606040955 expected: 102
predicted: 596.7626445293427 expected: 768
predicted: 103.10498949885368 expected: 42
predicted: 434.70216703414917 expected: 265
predicted: 81.1770147383213 expected: 15
predicted: 91.83225864171982 expected: 44
predicted: 11.343863356858492 expected: 1
predicted: 537.4120926856995 expected: 216
predicted: 48.527550891041756 expected: 20
predicted: 774.2462439537048 expected: 308
predicted: 59.75554421544075 expected: 29
predicted: 139.51090186834335 expected: 208
predicted: 261.1651872396469 expected: 99
predicted: 127.64895606040955 expected: 48
predicted: 21.161469757556915 expected: 15
predicted: 600.7225935459137 expected: 630
predicted: 434.70216703414917 expected: 333
predicted: 54.16564789414406 expected: 55
predicted: 97.1587750017643 expected: 52
predicted: 115.16775435209274 expected: 27
predicted: 65.26819238066673 expected: 16
predicted: 774.2462439537048 expected: 472
predicted: 537.4120926856995 expected: 162
predicted: 2123.512541770935 expected: 1054
predicted: 240.71561586856842 expected: 223
predicted: 28.208953201770782 expected: 22
predicted: 2123.512541770935 expected: 3345
predicted: 202.41596513986588 expected: 71
predicted: 699.9274635314941 expected: 862
predicted: 86.49207654595375 expected: 26
predicted: 1915.4331912994385 expected: 624
predicted: 103.10498949885368 expected: 24
predicted: 48.527550891041756 expected: 115
predicted: 2123.512541770935 expected: 3131
predicted: 115.16775435209274 expected: 27
predicted: 2123.512541770935 expected: 1116
predicted: 387.36106872558594 expected: 501
predicted: 1915.4331912994385 expected: 1380
predicted: 644.3031105995178 expected: 538
predicted: 65.26819238066673 expected: 77
predicted: 411.03159296512604 expected: 270
predicted: 1915.4331912994385 expected: 618
predicted: 774.2462439537048 expected: 335
predicted: 600.7225935459137 expected: 550
predicted: 219.68049824237823 expected: 169
predicted: 600.7225935459137 expected: 653
predicted: 70.57235398888588 expected: 21
predicted: 332.129297375679 expected: 225
predicted: 2123.512541770935 expected: 2192
predicted: 545.3180384635925 expected: 213
predicted: 612.6023907661438 expected: 695
predicted: 37.927471339702606 expected: 23
predicted: 198.0343341231346 expected: 148
predicted: 257.2054126262665 expected: 57
predicted: 86.49207654595375 expected: 42
predicted: 257.2054126262665 expected: 195
predicted: 276.9487009048462 expected: 172
predicted: 723.9729795455933 expected: 220
predicted: 304.5386129617691 expected: 112
predicted: 54.16564789414406 expected: 14
predicted: 537.4120926856995 expected: 314
predicted: 115.16775435209274 expected: 47
predicted: 672.1024808883667 expected: 836
predicted: 612.6023907661438 expected: 375
predicted: 600.7225935459137 expected: 501
predicted: 549.2710113525391 expected: 392
predicted: 612.6023907661438 expected: 824
predicted: 184.68381971120834 expected: 220
predicted: 14.693757019937038 expected: 3
predicted: 375.52579414844513 expected: 307
predicted: 12.627800181508064 expected: 18
predicted: 109.09023916721344 expected: 93
predicted: 493.93860936164856 expected: 180
predicted: 14.693757019937038 expected: 3
predicted: 774.2462439537048 expected: 297
predicted: 723.9729795455933 expected: 576
predicted: 774.2462439537048 expected: 314
predicted: 336.07454669475555 expected: 139
predicted: 109.09023916721344 expected: 105
predicted: 493.93860936164856 expected: 231
predicted: 54.16564789414406 expected: 12
predicted: 1915.4331912994385 expected: 1026
predicted: 493.93860936164856 expected: 304
predicted: 11.343863356858492 expected: 3
predicted: 545.3180384635925 expected: 335
predicted: 184.68381971120834 expected: 110
predicted: 48.527550891041756 expected: 43
predicted: 165.96958500146866 expected: 113
predicted: 600.7225935459137 expected: 487
predicted: 545.3180384635925 expected: 541
predicted: 359.7454446554184 expected: 114
predicted: 54.16564789414406 expected: 101
predicted: 320.3024938106537 expected: 251
predicted: 2123.512541770935 expected: 1358
predicted: 549.2710113525391 expected: 1031
predicted: 2123.512541770935 expected: 1788
predicted: 328.18449652194977 expected: 137
predicted: 21.161469757556915 expected: 29
predicted: 115.16775435209274 expected: 68
predicted: 699.9274635314941 expected: 442
predicted: 65.26819238066673 expected: 24
predicted: 672.1024808883667 expected: 1129
predicted: 24.54372066259384 expected: 35
predicted: 133.6499207019806 expected: 181
predicted: 774.2462439537048 expected: 894
predicted: 24.54372066259384 expected: 49
predicted: 198.0343341231346 expected: 170
predicted: 257.2054126262665 expected: 196
predicted: 2123.512541770935 expected: 3345
predicted: 115.16775435209274 expected: 24
predicted: 644.3031105995178 expected: 629
predicted: 774.2462439537048 expected: 290
predicted: 644.3031105995178 expected: 342
predicted: 240.71561586856842 expected: 177
predicted: 65.26819238066673 expected: 57
predicted: 1915.4331912994385 expected: 707
predicted: 379.47086906433105 expected: 289
predicted: 86.49207654595375 expected: 78
predicted: 545.3180384635925 expected: 530
predicted: 320.3024938106537 expected: 276
predicted: 600.7225935459137 expected: 389
predicted: 91.83225864171982 expected: 173
predicted: 774.2462439537048 expected: 717
predicted: 699.9274635314941 expected: 707
predicted: 774.2462937831879 expected: 440
predicted: 37.927471339702606 expected: 36
predicted: 139.51090186834335 expected: 115
predicted: 232.32288587093353 expected: 437
predicted: 115.16775435209274 expected: 75
predicted: 2123.512541770935 expected: 611
predicted: 70.57235398888588 expected: 17
predicted: 28.208953201770782 expected: 52
predicted: 699.9274635314941 expected: 849
predicted: 596.7626445293427 expected: 230
predicted: 545.3180384635925 expected: 537
predicted: 2123.512541770935 expected: 1645
predicted: 545.3180384635925 expected: 221
predicted: 236.52261435985565 expected: 167
predicted: 596.7626445293427 expected: 274
predicted: 273.00728845596313 expected: 141
predicted: 774.2462439537048 expected: 414
predicted: 28.208953201770782 expected: 32
predicted: 545.3180384635925 expected: 203
predicted: 21.161469757556915 expected: 18
predicted: 320.3024938106537 expected: 212
predicted: 86.49207654595375 expected: 29
predicted: 2123.512541770935 expected: 1665
predicted: 537.4120926856995 expected: 192
predicted: 65.26819238066673 expected: 24
predicted: 328.18449652194977 expected: 175
predicted: 2123.512541770935 expected: 1329
predicted: 723.9729795455933 expected: 261
predicted: 545.3180384635925 expected: 712
predicted: 189.14595025777817 expected: 52
predicted: 537.4120926856995 expected: 157
predicted: 367.6356443166733 expected: 285
predicted: 486.0374970436096 expected: 405
predicted: 493.93860936164856 expected: 452
predicted: 2123.512541770935 expected: 1267
predicted: 28.208953201770782 expected: 50
predicted: 211.13050639629364 expected: 150
predicted: 316.3615049123764 expected: 255
predicted: 43.05705846846104 expected: 18
predicted: 11.343863356858492 expected: 4
predicted: 640.3356373310089 expected: 437
predicted: 28.208953201770782 expected: 24
predicted: 65.26819238066673 expected: 71
predicted: 774.2462439537048 expected: 532
predicted: 699.9274635314941 expected: 729
predicted: 54.16564789414406 expected: 35
predicted: 672.1024808883667 expected: 368
predicted: 11.343863356858492 expected: 12
predicted: 2123.512541770935 expected: 2034
predicted: 644.3031105995178 expected: 391
predicted: 545.3180384635925 expected: 560
predicted: 774.2462439537048 expected: 1011
predicted: 672.1024808883667 expected: 600
predicted: 328.18449652194977 expected: 167
predicted: 28.208953201770782 expected: 34
predicted: 70.57235398888588 expected: 47
predicted: 2123.512541770935 expected: 1148
predicted: 434.70216703414917 expected: 326
predicted: 2123.512541770935 expected: 876
predicted: 12.627800181508064 expected: 10
predicted: 2123.512541770935 expected: 3345
predicted: 612.6023907661438 expected: 993
predicted: 155.66449910402298 expected: 49
predicted: 244.90425729751587 expected: 230
predicted: 644.3031105995178 expected: 679
predicted: 2123.512541770935 expected: 2201
predicted: 276.9487009048462 expected: 202
predicted: 830.9600687026978 expected: 663
predicted: 184.68381971120834 expected: 79
predicted: 482.08706545829773 expected: 214
predicted: 612.6023907661438 expected: 829
predicted: 115.16775435209274 expected: 149
predicted: 545.3180384635925 expected: 729
predicted: 91.83225864171982 expected: 19
predicted: 383.4159938097 expected: 173
predicted: 596.7626445293427 expected: 240
predicted: 211.13050639629364 expected: 89
predicted: 54.16564789414406 expected: 49
predicted: 600.7225935459137 expected: 228
predicted: 2123.512541770935 expected: 651
predicted: 32.85735569894314 expected: 15
predicted: 165.96958500146866 expected: 61
predicted: 81.1770147383213 expected: 84
predicted: 54.16564789414406 expected: 36
predicted: 198.0343341231346 expected: 101
predicted: 462.3333628177643 expected: 184
predicted: 486.0374472141266 expected: 268
predicted: 2123.512541770935 expected: 2910
predicted: 81.1770147383213 expected: 106
predicted: 434.70216703414917 expected: 433
predicted: 2123.512541770935 expected: 1700
predicted: 75.86929032206535 expected: 41
predicted: 11.343863356858492 expected: 1
predicted: 600.7225935459137 expected: 520
predicted: 17.82952171564102 expected: 50
predicted: 644.3031105995178 expected: 734
predicted: 37.927471339702606 expected: 45
predicted: 2123.512541770935 expected: 2837
predicted: 81.1770147383213 expected: 23
predicted: 219.68049824237823 expected: 145
predicted: 774.2462439537048 expected: 185
predicted: 24.54372066259384 expected: 42
predicted: 355.80036973953247 expected: 410
predicted: 2123.512541770935 expected: 1622
predicted: 612.6023907661438 expected: 661
predicted: 14.693757019937038 expected: 4
predicted: 462.3333628177643 expected: 369
predicted: 411.03159296512604 expected: 221
predicted: 462.3333628177643 expected: 234
predicted: 236.52261435985565 expected: 380
predicted: 545.3180384635925 expected: 249
predicted: 12.627800181508064 expected: 25
predicted: 2123.512541770935 expected: 1876
predicted: 411.03159296512604 expected: 241
predicted: 343.96504533290863 expected: 334
predicted: 407.0865429639816 expected: 303
predicted: 17.82952171564102 expected: 19
predicted: 2123.512541770935 expected: 1248
predicted: 232.32288587093353 expected: 501
predicted: 644.3031105995178 expected: 328
predicted: 414.9767426252365 expected: 406
predicted: 273.00728845596313 expected: 141
predicted: 232.32288587093353 expected: 408
predicted: 11.343863356858492 expected: 4
predicted: 2123.512541770935 expected: 3147
predicted: 109.09023916721344 expected: 99
predicted: 320.3024938106537 expected: 89
predicted: 103.10498949885368 expected: 61
predicted: 65.26819238066673 expected: 27
predicted: 2123.512541770935 expected: 1088
predicted: 109.09023916721344 expected: 105
predicted: 215.4403829574585 expected: 173
predicted: 1915.4327926635742 expected: 1496
predicted: 774.2463436126709 expected: 866
predicted: 723.9729795455933 expected: 399
predicted: 407.0865429639816 expected: 317
predicted: 401.773549079895 expected: 617
predicted: 73.26794216036797 expected: 172
predicted: 1314.783121585846 expected: 611
predicted: 144.73656958341599 expected: 269
predicted: 1314.783121585846 expected: 1095
predicted: 144.73656958341599 expected: 68
predicted: 42.01345157623291 expected: 5
predicted: 401.773549079895 expected: 986
predicted: 212.28029680252075 expected: 262
predicted: 364.7861204147339 expected: 817
predicted: 237.52211904525757 expected: 197
predicted: 190.2229397892952 expected: 264
predicted: 325.382360458374 expected: 830
predicted: 1314.783121585846 expected: 1415
predicted: 322.7557487487793 expected: 134
predicted: 65.99762749671936 expected: 58
predicted: 87.60356676578522 expected: 93
predicted: 464.96899366378784 expected: 830
predicted: 1314.783121585846 expected: 1207
predicted: 330.63169717788696 expected: 269
predicted: 449.28351950645447 expected: 558
predicted: 1190.0794587135315 expected: 1558
predicted: 75.64841490983963 expected: 37
predicted: 467.5880811214447 expected: 364
predicted: 483.3023567199707 expected: 1020
predicted: 40.28947292268276 expected: 11
predicted: 359.5212368965149 expected: 225
predicted: 467.5880811214447 expected: 228
predicted: 1190.0794587135315 expected: 1184
predicted: 346.38180017471313 expected: 370
predicted: 80.41028225421906 expected: 68
predicted: 184.85256725549698 expected: 201
predicted: 1314.783121585846 expected: 1113
predicted: 346.38180017471313 expected: 496
predicted: 80.41028225421906 expected: 43
predicted: 90.07705247402191 expected: 59
predicted: 80.41028225421906 expected: 60
predicted: 92.55052572488785 expected: 78
predicted: 256.8926827907562 expected: 263
predicted: 330.63169717788696 expected: 400
predicted: 1314.783121585846 expected: 1256
predicted: 80.41028225421906 expected: 23
predicted: 1314.783121585846 expected: 3345
predicted: 99.96638607978821 expected: 98
predicted: 514.7448103427887 expected: 238
predicted: 107.66191440820694 expected: 69
predicted: 359.5212368965149 expected: 170
predicted: 61.403461277484894 expected: 31
predicted: 112.91360557079315 expected: 102
predicted: 1314.783121585846 expected: 1070
predicted: 163.47358882427216 expected: 96
predicted: 430.82638001441956 expected: 387
predicted: 70.90167081356049 expected: 25
predicted: 417.6245069503784 expected: 574
predicted: 364.7861204147339 expected: 165
predicted: 399.1275038719177 expected: 765
predicted: 464.96899366378784 expected: 599
predicted: 449.28351950645447 expected: 906
predicted: 97.49749714136124 expected: 71
predicted: 409.7033631801605 expected: 433
predicted: 409.7033631801605 expected: 152
predicted: 126.12042421102524 expected: 116
predicted: 46.30681589245796 expected: 19
predicted: 43.969118639826775 expected: 24
predicted: 112.91360557079315 expected: 97
predicted: 126.12042421102524 expected: 49
predicted: 514.7448103427887 expected: 291
predicted: 1314.783121585846 expected: 2816
predicted: 359.5212368965149 expected: 152
predicted: 464.96899366378784 expected: 1033
predicted: 449.28351950645447 expected: 740
predicted: 87.60356676578522 expected: 32
predicted: 115.54896235466003 expected: 74
predicted: 514.7448103427887 expected: 453
predicted: 359.5212368965149 expected: 219
predicted: 152.77325546741486 expected: 82
predicted: 168.81658005714417 expected: 72
predicted: 78.01309171319008 expected: 109
predicted: 401.773549079895 expected: 567
predicted: 291.23790311813354 expected: 389
predicted: 87.60356676578522 expected: 70
predicted: 1314.783121585846 expected: 987
predicted: 1314.783121585846 expected: 1812
predicted: 467.5880811214447 expected: 169
predicted: 259.63664293289185 expected: 270
predicted: 58.91799157857895 expected: 33
predicted: 139.40237325429916 expected: 75
predicted: 1314.783121585846 expected: 1424
predicted: 46.30680966377258 expected: 39
predicted: 58.91799157857895 expected: 49
predicted: 61.403461277484894 expected: 141
predicted: 51.25085982680321 expected: 24
predicted: 152.77325546741486 expected: 91
predicted: 220.80836367607117 expected: 101
predicted: 483.30240654945374 expected: 401
predicted: 483.30240654945374 expected: 570
predicted: 155.45215076208115 expected: 106
predicted: 53.80667006969452 expected: 43
predicted: 514.7448103427887 expected: 439
predicted: 1190.0794587135315 expected: 1220
predicted: 80.41028225421906 expected: 82
predicted: 38.92591091990471 expected: 5
predicted: 467.5880811214447 expected: 314
predicted: 346.38180017471313 expected: 180
predicted: 82.76570945978165 expected: 88
predicted: 1314.783121585846 expected: 1642
predicted: 401.773549079895 expected: 417
predicted: 278.1119203567505 expected: 346
predicted: 42.01345157623291 expected: 10
predicted: 163.47358882427216 expected: 93
predicted: 61.403461277484894 expected: 26
predicted: 120.82244145870209 expected: 41
predicted: 1314.783121585846 expected: 505
predicted: 38.92591091990471 expected: 11
predicted: 1314.783121585846 expected: 929
predicted: 364.7861204147339 expected: 877
predicted: 107.66191440820694 expected: 370
predicted: 40.28947292268276 expected: 28
predicted: 1314.783121585846 expected: 1085
predicted: 1190.0794587135315 expected: 654
predicted: 82.76570945978165 expected: 59
predicted: 184.85256725549698 expected: 259
predicted: 333.25636553764343 expected: 165
predicted: 38.92591091990471 expected: 1
predicted: 51.25085982680321 expected: 26
predicted: 152.77325546741486 expected: 399
predicted: 399.1275038719177 expected: 155
predicted: 179.49257171154022 expected: 158
predicted: 430.82638001441956 expected: 782
predicted: 182.16745573282242 expected: 376
predicted: 38.92591091990471 expected: 1
predicted: 223.59136521816254 expected: 116
predicted: 278.1119203567505 expected: 628
predicted: 1314.783121585846 expected: 1897
predicted: 134.08489471673965 expected: 76
predicted: 548.8336579799652 expected: 450
predicted: 259.63664293289185 expected: 272
predicted: 359.5212368965149 expected: 149
predicted: 1190.0794587135315 expected: 1069
predicted: 209.44163572788239 expected: 286
predicted: 179.49255925416946 expected: 526
predicted: 75.64841490983963 expected: 29
predicted: 278.1119203567505 expected: 373
predicted: 152.77325546741486 expected: 481
predicted: 144.73656958341599 expected: 74
predicted: 70.90167081356049 expected: 54
predicted: 158.13044810295105 expected: 102
predicted: 38.92591091990471 expected: 22
predicted: 1314.783121585846 expected: 1360
predicted: 430.82638001441956 expected: 324
predicted: 399.1275038719177 expected: 973
predicted: 401.773549079895 expected: 407
predicted: 218.01178359985352 expected: 317
predicted: 46.30680966377258 expected: 76
predicted: 417.6245069503784 expected: 688
predicted: 68.44694590568542 expected: 44
predicted: 330.63169717788696 expected: 239
predicted: 430.82638001441956 expected: 344
predicted: 417.6245069503784 expected: 688
predicted: 359.5212867259979 expected: 442
predicted: 131.4261554479599 expected: 157
predicted: 1314.783121585846 expected: 1193
predicted: 95.02401143312454 expected: 102
predicted: 399.1275038719177 expected: 768
predicted: 85.1340674161911 expected: 42
predicted: 291.23790311813354 expected: 265
predicted: 75.64841490983963 expected: 15
predicted: 80.41028225421906 expected: 44
predicted: 38.92591091990471 expected: 1
predicted: 359.5212368965149 expected: 216
predicted: 61.403461277484894 expected: 20
predicted: 514.7448103427887 expected: 308
predicted: 65.99762749671936 expected: 29
predicted: 99.96638607978821 expected: 208
predicted: 171.48555928468704 expected: 99
predicted: 95.02401143312454 expected: 48
predicted: 46.30680966377258 expected: 15
predicted: 401.773549079895 expected: 630
predicted: 291.23790311813354 expected: 333
predicted: 63.73411077260971 expected: 55
predicted: 82.76570945978165 expected: 52
predicted: 90.07705247402191 expected: 27
predicted: 68.44694590568542 expected: 16
predicted: 514.7448103427887 expected: 472
predicted: 359.5212368965149 expected: 162
predicted: 1314.783121585846 expected: 1054
predicted: 158.13044810295105 expected: 223
predicted: 51.25085982680321 expected: 22
predicted: 1314.783121585846 expected: 3345
predicted: 134.08489471673965 expected: 71
predicted: 467.5880811214447 expected: 862
predicted: 78.0130854845047 expected: 26
predicted: 1190.0794587135315 expected: 624
predicted: 85.1340674161911 expected: 24
predicted: 61.403461277484894 expected: 115
predicted: 1314.783121585846 expected: 3131
predicted: 90.07705247402191 expected: 27
predicted: 1314.783121585846 expected: 1116
predicted: 259.63664293289185 expected: 501
predicted: 1190.0794587135315 expected: 1380
predicted: 430.82638001441956 expected: 538
predicted: 68.44694590568542 expected: 77
predicted: 275.48745131492615 expected: 270
predicted: 1190.0794587135315 expected: 618
predicted: 514.7448103427887 expected: 335
predicted: 401.773549079895 expected: 550
predicted: 144.73656958341599 expected: 169
predicted: 401.773549079895 expected: 653
predicted: 70.90167081356049 expected: 21
predicted: 220.80836367607117 expected: 225
predicted: 1314.783121585846 expected: 2192
predicted: 364.7861204147339 expected: 213
predicted: 409.7033631801605 expected: 695
predicted: 56.28768002986908 expected: 23
predicted: 131.4261554479599 expected: 148
predicted: 168.81658005714417 expected: 57
predicted: 78.01309171319008 expected: 42
predicted: 168.81658005714417 expected: 195
predicted: 182.16745573282242 expected: 172
predicted: 483.3023567199707 expected: 220
predicted: 201.05135983228683 expected: 112
predicted: 63.73411077260971 expected: 14
predicted: 359.5212867259979 expected: 314
predicted: 90.07705247402191 expected: 47
predicted: 449.28351950645447 expected: 836
predicted: 409.7033631801605 expected: 375
predicted: 401.773549079895 expected: 501
predicted: 367.4239935874939 expected: 392
predicted: 409.7033631801605 expected: 824
predicted: 123.46761465072632 expected: 220
predicted: 42.01345157623291 expected: 3
predicted: 251.4055597782135 expected: 307
predicted: 40.28947292268276 expected: 18
predicted: 87.60356676578522 expected: 93
predicted: 330.63169717788696 expected: 180
predicted: 42.01345157623291 expected: 3
predicted: 514.7448103427887 expected: 297
predicted: 483.3023567199707 expected: 576
predicted: 514.7448103427887 expected: 314
predicted: 223.59136521816254 expected: 139
predicted: 87.60356676578522 expected: 105
predicted: 330.63169717788696 expected: 231
predicted: 63.73411077260971 expected: 12
predicted: 1190.0794587135315 expected: 1026
predicted: 330.63169717788696 expected: 304
predicted: 38.92591091990471 expected: 3
predicted: 364.7861204147339 expected: 335
predicted: 123.46761465072632 expected: 110
predicted: 61.403461277484894 expected: 43
predicted: 112.91360557079315 expected: 113
predicted: 401.773549079895 expected: 487
predicted: 364.7861204147339 expected: 541
predicted: 240.3073878288269 expected: 114
predicted: 63.73411077260971 expected: 101
predicted: 212.28029680252075 expected: 251
predicted: 1314.7829222679138 expected: 1358
predicted: 367.4239935874939 expected: 1031
predicted: 1314.783121585846 expected: 1788
predicted: 218.01178359985352 expected: 137
predicted: 46.30680966377258 expected: 29
predicted: 90.07705247402191 expected: 68
predicted: 467.5880811214447 expected: 442
predicted: 68.44694590568542 expected: 24
predicted: 449.28351950645447 expected: 1129
predicted: 48.75834237039089 expected: 35
predicted: 97.49749714136124 expected: 181
predicted: 514.7448103427887 expected: 894
predicted: 48.75834237039089 expected: 49
predicted: 131.4261554479599 expected: 170
predicted: 168.81658005714417 expected: 196
predicted: 1314.783121585846 expected: 3345
predicted: 90.07705247402191 expected: 24
predicted: 430.82638001441956 expected: 629
predicted: 514.7448103427887 expected: 290
predicted: 430.82638001441956 expected: 342
predicted: 158.13044810295105 expected: 177
predicted: 68.44694590568542 expected: 57
predicted: 1190.0794587135315 expected: 707
predicted: 254.1435902118683 expected: 289
predicted: 78.01309171319008 expected: 78
predicted: 364.7861204147339 expected: 530
predicted: 212.28029680252075 expected: 276
predicted: 401.773549079895 expected: 389
predicted: 80.41028225421906 expected: 173
predicted: 514.7448103427887 expected: 717
predicted: 467.5880811214447 expected: 707
predicted: 514.7448103427887 expected: 440
predicted: 56.28769248723984 expected: 36
predicted: 99.96638607978821 expected: 115
predicted: 152.77325546741486 expected: 437
predicted: 90.07705247402191 expected: 75
predicted: 1314.783121585846 expected: 611
predicted: 70.90167081356049 expected: 17
predicted: 51.25085982680321 expected: 52
predicted: 467.5880811214447 expected: 849
predicted: 399.1275038719177 expected: 230
predicted: 364.7861204147339 expected: 537
predicted: 1314.783121585846 expected: 1645
predicted: 364.7861204147339 expected: 221
predicted: 155.45215076208115 expected: 167
predicted: 399.1275038719177 expected: 274
predicted: 179.49255925416946 expected: 141
predicted: 514.7448103427887 expected: 414
predicted: 51.25085982680321 expected: 32
predicted: 364.7861204147339 expected: 203
predicted: 46.30680966377258 expected: 18
predicted: 212.28029680252075 expected: 212
predicted: 78.01309171319008 expected: 29
predicted: 1314.783121585846 expected: 1665
predicted: 359.5212368965149 expected: 192
predicted: 68.44694590568542 expected: 24
predicted: 218.01178359985352 expected: 175
predicted: 1314.783121585846 expected: 1329
predicted: 483.3023567199707 expected: 261
predicted: 364.7861204147339 expected: 712
predicted: 126.12042421102524 expected: 52
predicted: 359.5212368965149 expected: 157
predicted: 245.8841540813446 expected: 285
predicted: 325.382360458374 expected: 405
predicted: 330.63169717788696 expected: 452
predicted: 1314.783121585846 expected: 1267
predicted: 51.25085982680321 expected: 50
predicted: 139.40237325429916 expected: 150
predicted: 209.44163572788239 expected: 255
predicted: 58.91799157857895 expected: 18
predicted: 38.92591091990471 expected: 4
predicted: 428.1859655380249 expected: 437
predicted: 51.25085982680321 expected: 24
predicted: 68.44694590568542 expected: 71
predicted: 514.7448103427887 expected: 532
predicted: 467.5880811214447 expected: 729
predicted: 63.73411077260971 expected: 35
predicted: 449.28351950645447 expected: 368
predicted: 38.92591091990471 expected: 12
predicted: 1314.783121585846 expected: 2034
predicted: 430.82638001441956 expected: 391
predicted: 364.7861204147339 expected: 560
predicted: 514.7448103427887 expected: 1011
predicted: 449.28351950645447 expected: 600
predicted: 218.01178359985352 expected: 167
predicted: 51.25085982680321 expected: 34
predicted: 70.90167081356049 expected: 47
predicted: 1314.783121585846 expected: 1148
predicted: 291.23790311813354 expected: 326
predicted: 1314.783121585846 expected: 876
predicted: 40.28947292268276 expected: 10
predicted: 1314.783121585846 expected: 3345
predicted: 409.7033631801605 expected: 993
predicted: 107.66191440820694 expected: 49
predicted: 160.80200600624084 expected: 230
predicted: 430.82638001441956 expected: 679
predicted: 1314.7829222679138 expected: 2201
predicted: 182.16745573282242 expected: 202
predicted: 548.8336579799652 expected: 663
predicted: 123.46761465072632 expected: 79
predicted: 322.7557487487793 expected: 214
predicted: 409.7033631801605 expected: 829
predicted: 90.07705247402191 expected: 149
predicted: 364.7861204147339 expected: 729
predicted: 80.41028225421906 expected: 19
predicted: 256.8926827907562 expected: 173
predicted: 399.1275038719177 expected: 240
predicted: 139.40237325429916 expected: 89
predicted: 63.73411077260971 expected: 49
predicted: 401.773549079895 expected: 228
predicted: 1314.783121585846 expected: 651
predicted: 53.80667006969452 expected: 15
predicted: 112.91360557079315 expected: 61
predicted: 75.64841490983963 expected: 84
predicted: 63.73411077260971 expected: 36
predicted: 131.4261554479599 expected: 101
predicted: 309.62283968925476 expected: 184
predicted: 325.382360458374 expected: 268
predicted: 1314.783121585846 expected: 2910
predicted: 75.64841490983963 expected: 106
predicted: 291.23790311813354 expected: 433
predicted: 1314.783121585846 expected: 1700
predicted: 73.26794216036797 expected: 41
predicted: 38.92591091990471 expected: 1
predicted: 401.773549079895 expected: 520
predicted: 43.969118639826775 expected: 50
predicted: 430.82638001441956 expected: 734
predicted: 56.28769248723984 expected: 45
predicted: 1314.7829222679138 expected: 2837
predicted: 75.64841490983963 expected: 23
predicted: 144.73656958341599 expected: 145
predicted: 514.7448103427887 expected: 185
predicted: 48.75834237039089 expected: 42
predicted: 237.52211904525757 expected: 410
predicted: 1314.783121585846 expected: 1622
predicted: 409.7033631801605 expected: 661
predicted: 42.01345157623291 expected: 4
predicted: 309.62283968925476 expected: 369
predicted: 275.48745131492615 expected: 221
predicted: 309.62283968925476 expected: 234
predicted: 155.45215076208115 expected: 380
predicted: 364.7861204147339 expected: 249
predicted: 40.28947292268276 expected: 25
predicted: 1314.783121585846 expected: 1876
predicted: 275.48745131492615 expected: 241
predicted: 229.16215193271637 expected: 334
predicted: 272.8629822731018 expected: 303
predicted: 43.969118639826775 expected: 19
predicted: 1314.783121585846 expected: 1248
predicted: 152.77325546741486 expected: 501
predicted: 430.82638001441956 expected: 328
predicted: 278.1119203567505 expected: 406
predicted: 179.49255925416946 expected: 141
predicted: 152.77325546741486 expected: 408
predicted: 38.92591091990471 expected: 4
predicted: 1314.783121585846 expected: 3147
predicted: 87.60356676578522 expected: 99
predicted: 212.28029680252075 expected: 89
predicted: 85.1340674161911 expected: 61
predicted: 68.44694590568542 expected: 27
predicted: 1314.7829222679138 expected: 1088
predicted: 87.6035729944706 expected: 105
predicted: 142.0643888115883 expected: 173
predicted: 1190.0794587135315 expected: 1496
predicted: 514.7448103427887 expected: 866
predicted: 483.30240654945374 expected: 399
predicted: 272.8629324436188 expected: 317

25
rmse.txt Normal file
View File

@ -0,0 +1,25 @@
109845.55756236914
104845.55756236914
109845.55756236914
109845.55756236914
104845.55756236914
19845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
109845.55756236914
114882.99377127373
129787.96004765884