Downloads · 30 days
0
anyflow-biomegacorr-stage2-tracks-camera-v3-50k
yimingshi/anyflow-biomegacorr-stage2-tracks-camera-v3-50k
anyflow-biomegacorr-stage2-tracks-camera-v3-50k is a machine learning model from yimingshi. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This repository publishes the accepted Stage 2 checkpoint from the AnyFlow / Bi-OmegaCorr curriculum after 50,000 track-supervision training steps.
Access
Public
Updated Jul 24, 2026
Repo size
4.8 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt4.8 GB · 100%
At a glance
- Access
- Public
- Created
- Jul 24, 2026
- Updated
- Jul 24, 2026
- SHA
- 40fe1518
Model card
From the Hugging Face model README
AnyFlow Bi-OmegaCorr — Stage 2 Tracks / Camera-v3 (50k)
This repository publishes the accepted Stage 2 checkpoint from the AnyFlow / Bi-OmegaCorr curriculum after 50,000 track-supervision training steps.
Artifacts
| File | Purpose | Size | SHA256 |
|---|---|---|---|
anyflow_biomegacorr_stage2_tracks_camera_v3_50k.pt | Exact native PyTorch checkpoint, including model, optimizer, and scaler state | 4,731,466,401 bytes | 61872b31b0b0dfdd19390aa17663b05d34cdfd6476eb9cf3b3c5724b28f41b9f |
anyflow_biomegacorr_stage2_tracks_camera_v3_head_only.pt | Collaborator package containing only the 86 trained omega_corr_head.* tensors and training metadata | 51,552,489 bytes | d71d1584c2e358c68f75972bc5e7ecbd38697abcd44db30b0a5b47285dd7834c |
Use the full checkpoint for exact continuation into Stage 3. Use the head-only artifact when the consumer already has the compatible VGGT backbone.
Important scope
Stage 2 trained only the Bi-OmegaCorr correspondence head. The VGGT backbone was frozen:
unfreeze_last_blocks = 0joint_vggt_loss = false- all 1,411 non-head model tensors remained byte-identical to accepted Stage 1
- all 86
omega_corr_head.*tensors changed - optimizer state contains only the 86 head parameters
Therefore the head-only artifact intentionally omits the unchanged backbone. The first curriculum stage that unfreezes VGGT is Stage 4, not Stage 2 or Stage 3.
Training result
| Field | Value |
|---|---|
| Step | 50,000 |
| Source DLC job | dlc8sduru74dp8nx |
| Job status | Succeeded / JobSucceeded |
| Input mode | window |
| Training manifest | 138,619 flow + track rows |
| Final logged minibatch loss | 0.02204616 |
Final logged minibatch epe_i2j | 2.5607288 |
| Final logged coordinate loss | 0.0134771 |
| Final logged visibility loss | 0.0412766 |
| Final logged cycle loss | 0.03137335 |
| Final logged VGGT geometry loss | 0 |
These are the last logged training-minibatch values, not held-out validation metrics.
Lineage
- Training and validation code:
1e6e356cfce908c6f82b65e2451749be362f1d5c - Parent Stage 1 checkpoint SHA256:
9db208657c7f18ad8392a0e7db963661c9bf08030728452db6744610b1fbe045 - Camera-v3 tracks manifest SHA256:
430c3f5ce41b3759c512e971d263178d981e7a8f92db9c8356445c459871eb40 - W&B: https://wandb.ai/xiugo/vggt-omega-bicorr/runs/anyflow-cv3-main1e6e356-stage02-tracks-cleaninit-20260723-165324
Loading
from huggingface_hub import hf_hub_download
import torch
full_path = hf_hub_download(
repo_id="Shiym/anyflow-biomegacorr-stage2-tracks-camera-v3-50k",
filename="anyflow_biomegacorr_stage2_tracks_camera_v3_50k.pt",
)
checkpoint = torch.load(full_path, map_location="cpu")
print(checkpoint["step"])
For the head-only package:
head_path = hf_hub_download(
repo_id="Shiym/anyflow-biomegacorr-stage2-tracks-camera-v3-50k",
filename="anyflow_biomegacorr_stage2_tracks_camera_v3_head_only.pt",
)
head_package = torch.load(head_path, map_location="cpu")
model.load_state_dict(head_package["model"], strict=False)
The head-only package excludes optimizer/scaler state and is not itself a resume checkpoint.
- Created
- Jul 24, 2026
- Updated
- Jul 24, 2026