VLM models for Finals

Accuracy: 0.92500000
Speed: 0.791856667962963
Tree:

.
├── VLMManager.py
├── api_service.py
├── clip_ft_final.pt
└── highdegree60.pt

Maxed out autonomy turret speed

autonomy-v1.0.2

Max out turret rotation speed

Autonomy with faster turret rotation speed

autonomy-v1.0.1

Autonomy with faster turret rotation speed

Fix VLM after testing session

(Stats from vlm-v2.1 334a197)
Accuracy: 0.91400000
Speed: 0.7849992909259259

Tree:

.
├── Dockerfile
├── README.md
├── example.jpg
├── requirements.txt
├── results
└── src
    ├── VLMManager.py
    ├── api_service.py
    ├── augmentedbestv2.pt
    └── clip_ft_2.pt

Fix ASR after testing session

(Stats from asr-v1.2.2 11bfd9a)
Accuracy: 0.9996609403254972
Speed: 0.8045008579629629

Tree:

asr
├── Dockerfile
├── README.md
├── correction_model
│   ├── config.json
│   ├── generation_config.json
│   ├── model.safetensors
│   └── training_args.bin
├── model
│   ├── added_tokens.json
│   ├── config.json
│   ├── merges.txt
│   ├── normalizer.json
│   ├── preprocessor_config.json
│   ├── pytorch_model.bin
│   ├── special_tokens_map.json
│   ├── tokenizer_config.json
│   └── vocab.json
├── requirements.txt
├── src
│   ├── ASRManager.py
│   ├── api_service.py
│   └── correction.py
└── tokenizer
    ├── merges.txt
    ├── special_tokens_map.json
    ├── tokenizer.json
    ├── tokenizer_config.json
    └── vocab.json

4 directories, 24 files

Get assets:

(rm -rf model correction_model tokenizer) && ((gh release download asr-v1.2.3 --pattern model.zip && unzip model.zip && rm model.zip) & (gh release download asr-v1.2.3 --pattern correction_model.zip && unzip correction_model.zip && rm correction_model.zip) & (gh release download asr-v1.2.3 --pattern tokenizer.zip && unzip tokenizer.zip && rm tokenizer.zip))

Improve VLM with better training augmentation

Accuracy: 0.91400000
Speed: 0.7849992909259259
Tree:

.
├── Dockerfile
├── README.md
├── example.jpg
├── requirements.txt
├── results
└── src
    ├── VLMManager.py
    ├── api_service.py
    ├── augmentedbestv2.pt
    └── clip_ft_2.pt

Main program from base

main-v0.0

Main program from base

Autonomy with completed send_heading

autonomy-v1.0

Autonomy with completed send_heading

Change VLM to use better training dataset

Accuracy: 0.911
Speed: 0.7503025994444444
Tree:

.
├── Dockerfile
├── README.md
├── example.jpg
├── requirements.txt
├── results
└── src
    ├── 800allbest.pt
    ├── VLMManager.py
    ├── api_service.py
    └── clip_ft_2.pt

Improved VLM using OWLv2 and Clip

Accuracy: 0.598
Speed: 0.781718495
Tree:

.
├── Dockerfile
├── README.md
├── example.jpg
├── requirements.txt
├── results
└── src
    ├── VLMManager.py
    ├── api_service.py
    ├── clip_ft_2.pt
    └── yoloworldbest2.pt