clipinvariance

Lisadunlap's group workspace

Group: incorrect-correct

false

incorrect

false

incorrect

false

incorrect

["matter"]

incorrect

["dog"]

incorrect

["carnivore"]

incorrect

["wearable"]

incorrect

["implement"]

incorrect

["covering"]

incorrect

["equipment"]

incorrect

["bird"]

incorrect

["insect"]

incorrect

["reptile"]

incorrect

["self-propelled vehicle"]

incorrect

["structure"]

incorrect

["instrument"]

incorrect

["transport"]

incorrect

["wearable"]

incorrect

false

incorrect

false

incorrect

1-20

of 24

Tags

Notes

Author

lisadunlap

State

Finished

Start time

November 15th, 2023 6:26:28 AM

Runtime

3m 56s

Tracked hours

3m 42s

Run path

clipinvariance/VisDiff-ImageNetV2/0y1m4wgz

Linux-5.15.0-25-generic-x86_64-with-glibc2.35

Python version

3.9.16

Git repository

git clone git@github.com:lisadunlap/SummarizingDatasets.git

Git state

git checkout -b "CLIPValidator" 1b52ad429b731ec8bab688516db753ce8178ae19

Command

/shared/lisabdunlap/SummarizingDatasets/main.py --config configs/imagenet_v2/incorrect.yaml summarize.rounds=10 summarize.class_col=set summarize.class_ixds=[imagenet_v2]

System Hardware

CPU count	128
Logical CPU count	256
GPU count	8
GPU type	NVIDIA A100-SXM4-80GB

W&B CLI Version

0.15.10

Group

incorrect-correct

Config parameters are your model's inputs. Learn more

▶
Config parameters:{} 78 keys
- base_config:
  "configs/imagenet_v2/base.yaml"
- caption hash:
  "7e1dde16301cbb191e9e013bbb2e35e4"
- caption.anonomyze:
  false
- caption.img_features:
  "CLIPImg"
- caption.k:
  10
- caption.load:
  true
- caption.max_length:
  200
- caption.method:
  "blip"
- caption.min_length:
  10
- caption.model:
  "Salesforce/blip2-opt-2.7b"
- caption.prompt:
  "caption this image."
- caption.resume:
  false
- caption.text_features:
  "SBERT"
- checkpoint_name:
  false
- compute_summary:
  false
- cpu_offloading:
  false
- data:
  "diffusion_plates"
- dataset:
  "imagenet_v2_ensamble_preds"
- dataset_name:
  "ImageNetV2"
- debug:
  false
- device:
  "cuda"
- gptq_act_order:
  false
- gptq_ckpt:
  "None"
- gptq_groupsize:
  -1
- gptq_wbits:
  16
- gpus:
  "None"
- group A:
  "incorrect"
- group B:
  "correct"
- ▶
  group names:[] 2 items
  - 0:
    "incorrect"
  - 1:
    "correct"
- hash:
  "8eb8966f26c1701a05778b7fa9265648"
- load_8bit:
  false
- max_gpu_memory:
  "40GB"
- max_new_tokens:
  64
- ▶
  message:[] 2 items
  - 0:
    "howdy there, can you tell me a joke?"
  - 1:
    "what's the deadliest natural disaster in recorded history?"
- model_path:
  "lmsys/vicuna-13b-v1.3"
- name:
  "VisDiff-ImageNetV2"
- num_gpus:
  2
- openai_key:
  "/shared/lisabdunlap/SummarizingDatasets/openai_key.txt"
- repetition_penalty:
  1
- revision:
  "main"
- save_dir:
  "./data"
- seed:
  0
- summarize.cache:
  true
- summarize.caption_df:
  "/shared/lisabdunlap/SummarizingDatasets/data/imagenet_v2_ensamble_preds.csv"
- summarize.class_col:
  "set"
- summarize.class_idxs:
  false
- train.loss_weights:
  false
- verbose:
  false
- wandb:
  true
- yaml:
  "configs/imagenet_v2/incorrect.yaml"

Summary metrics are your model's outputs. Learn more

▶
Summary metrics:{} 6 keys
- original_run_id:
  "0y1m4wgz"
- results:
  "table-file"
- scored groundtruth:
  "table-file"
- scored hypotheses:
  "table-file"
- top_hypotheses:
  "complex scenes with multiple subjects"
- top_hypotheses_score:
  0.002336918671282339

This run consumed these artifacts as inputs. Learn more

job

job-git_github.com_lisadunlap_SummarizingDatasets.git_main.py:v29

This run produced these artifacts as outputs. Total: 4. Learn more

run_table

run-0y1m4wgz-results:v0

run_table

run-0y1m4wgz-scoredhypotheses:v0

run_table

run-0y1m4wgz-scoredgroundtruth:v0

wandb-history

run-0y1m4wgz-history:v0