hypersphere

Hypersphere's group workspace

Group: retinanet-nms-hyperparam-search

187

1-1

of 1

Tags

best-model

dynamic-tiling

eval-voc2012

iou-thres-0.5

mob-median

nms-hyperparam-search

nms-mode-argmax

nms-thres-0.25

resnet152

score-thres-0.03

test-set

Author

hypersphere

State

Finished

Start time

December 28th, 2020 9:25:00 PM

Sweep

m28nj07u

Runtime

7m 57s

Tracked hours

7m 45s

Run path

hypersphere/masters-thesis/etsafshh

Linux-3.10.0-1160.6.1.el7.x86_64-x86_64-with-Ubuntu-18.04-bionic

Python version

3.6.8

Git repository

git clone git@version.aalto.fi:pasi.pyrro/masters-thesis.git

Git state

git checkout -b "kind-sweep-16" 947fcddb1f3f90addbfaa429daf127600802921c

Command

/usr/local/bin/wandb agent --count 1 hypersphere/masters-thesis/m28nj07u

System Hardware

CPU count	16
GPU count	1
GPU type	Tesla V100-PCIE-32GB

W&B CLI Version

0.9.5

Group

retinanet-nms-hyperparam-search

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 30 keys
- anchor_scale:
  0.965
- backbone:
  "infer"
- config:
  "config.ini"
- convert_model:
  true
- dataset_checksum:
  "71cdbcdfeb56eeacf811239f08468dd2d4886766"
- dataset_name:
  "heridal_keras_retinanet_voc"
- dataset_num_files:
  3,338
- dataset_size:
  "8378.9M"
- dataset_type:
  "pascal"
- eval_mode:
  "voc2012"
- gpu:
  0
- group:
  "retinanet-nms-hyperparam-search"
- image_extension:
  ".jpg"
- image_max_side:
  2,025
- image_min_side:
  1,525
- image_tiling:
  true
- iou_threshold:
  0.5
- max_detections:
  1,000
- model:
  "/scratch/cs/sar-uav-cv/masters-thesis/computer_vision/keras-retinanet/snapshots/dauntless-sweep-2_resnet152_pascal.h5"
- name:
  null
- nms_mode:
  "argmax"
- nms_threshold:
  0.25
- pascal_path:
  "/tmp/UNIQUE_JOB_FOLDER/data/datasets/heridal_keras_retinanet_voc"
- save_path:
  null
- score_threshold:
  0.025
- seed:
  null
- set_name:
  "test"
- ▶
  tags:[] 3 items
  - 0:
    "best-model"
  - 1:
    "nms-hyperparam-search"
  - 2:
    "mob-median"
- top_k:
  10
- wandb_version:
  1

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 18 keys
- iou_threshold:
  0.5
- score_threshold:
  0.025
- ▶
  test_image_1:{} 8 keys
- ▶
  test_image_15:{} 8 keys
- ▶
  test_image_29:{} 8 keys
- ▶
  test_image_43:{} 8 keys
- ▶
  test_image_57:{} 8 keys
- ▶
  test_image_71:{} 8 keys
- ▶
  test_image_85:{} 8 keys
- ▶
  test_image_99:{} 8 keys
- test_mAP_person:
  0.8692215723006789
- test_mean_inference_time:
  0.3016588510853229
- test_num_anns_person:
  337
- test_num_fp_person:
  1,397
- test_num_tp_person:
  300
- ▶
  test_pr_curve_person:{} 4 keys
  - _type:
    "plotly-file"
  - path:
    "media/plotly/test_pr_curve_person_9_48fa29f1.plotly.json"
  - sha256:
    "48fa29f16c227b86bebebb4b1b25364947afd642ffc890fe402e506a79bbb439"
  - size:
    74,915
- test_precision_person:
  0.17678255745433116
- test_recall_person:
  0.8902077151335311

This run produced these artifacts as outputs. Total: 1. Learn more

wandb-history

run-etsafshh-history:v0