Add jpeg format to qwenvl2 (#684)Small PR to enable jpeg encoded image communication when using qwenvl2. main(#684)· v0.3.110.3.8-rc 2 parents e2b469b + 4dcf2cf commit 2d8b9ee File tree node-hub dora-qwenvl dora_qwenvl main.py pyproject.toml llama-factory-recorder llama_factory_re...
llamafactory-cli train examples/train_lora/qwen2vl_lora_sft.yaml ``` You can then ### Qwenvl2 ```bash # Clone llama factory cd .. git clone --depth 1 https://github.com/hiyouga/LLaMA-Factory.git cd ../dora_mickrobot dora build qwenvl2_recorder.yml ...
ChildFailedError for qwenvl2#7032 New issue ClosedDescription xiayq1 opened on Feb 21, 2025 Reminder I have read the above rules and searched the existing issues. System Info python: 3.10 accelerate 0.33.0 torch 2.4.0 torchvision 0.19.0 transformers 4.45.0 llamafactory 0.9.0 Reproduction ...
dora-qwenvl-recorder/text env: IMAGE_WIDTH: 640 IMAGE_HEIGHT: 480 - id: dora-qwenvl-recorder build: pip install -e llama-factory-recorder path: llama-factory-recorder inputs: image: source: camera/image queue_size: 1 ground_truth: key-interpolation/text text: text-interpolation...
Motivation 能否支持一下qwenvl2 Related resources No response Additional context No responseCollaborator irexyc commented Sep 10, 2024 正在做,应该会先在 pytorch backend 进行支持,然后在 turobmind backend 优化与加速。 👍 1 BlueBlueFF commented Sep 10, 2024 正在做,应该会先在 pytorch backend ...
dora-qwenvl/text env: IMAGE_WIDTH: 640 IMAGE_HEIGHT: 480 - id: dora-qwenvl build: pip install -e llama-factory path: llama-factory inputs: image: source: camera/image queue_size: 1 ground_truth: key-interpolation/text text: text-interpolation/question outputs: - text env: DEFAUL...