caption-test/README.md

# Caption Quality Test for Cross-View Geo-Localization

Validate whether generated text captions improve retrieval R@1 in cross-view
geo-localization (drone-to-satellite).

## Architecture

### Overview

Shared DINOv3 WEB encoder with MONA adapters (last 12 blocks, bfloat16),
DGTRS-CLIP text fusion, and projection to 512-dim retrieval space.

```
┌─────────────────────────────── QUERY BRANCH ─────────────────────────────────┐
│                                                                               │
│  drone_img ──► DINOv3 ViT-L/16 WEB (frozen, 303M) ──► CLS [B,1024]         │
│  [B,3,256,256]     + MONA adapters (3.5M, bf16)              │                │
│                    (2 per block × last 12 of 24, bn=64)      │                │
│                    + gradient checkpointing              Projection           │
│                                                     Linear(1024→512)         │
│                                                         d_img [B,512]        │
│                                                              │                │
│  L1 (overview)    ──► DGTRS-CLIP ViT-L-14 ──► z₁ [B,768] ─┐                 │
│  L2 (full desc)   ──►   (frozen, 124M)    ──► z₂ [B,768] ─┼─ cat ──[B,2304] │
│  L3 (fingerprint) ──►   + LoRA r=4 (147K) ──► z₃ [B,768] ─┘    │            │
│                        (248 tok, KPS pos.)      TextFusionMLP (shared, 1.5M) │
│                        + gradient checkpointing Linear(2304→512)→GELU→       │
│                                                 Linear(512→512)              │
│                                                       │                      │
│                                                  d_txt [B,512]               │
│                                                       │                      │
│              q = σ(α_q)·d_img + (1−σ(α_q))·d_txt    GatedFusion_q           │
│                            │                                                  │
│                      q̂ = q / ‖q‖₂  ──► query [B,512]                         │
└───────────────────────────────────────────────────────────────────────────────┘

┌────────────────────────────── GALLERY BRANCH ────────────────────────────────┐
│                                                                               │
│  sat_img ──► DINOv3 WEB (shared encoder) ──► CLS ──► Projection              │
│  [B,3,256,256]  + MONA (shared adapters)        s_img [B,512]                │
│                                                      │                        │
│  sat_L1 ──► DGTRS-CLIP ──► z₁ ─┐                                             │
│  sat_L2 ──► (shared)    ──► z₂ ─┼─ cat ──► TextFusionMLP ──► s_txt [B,512]  │
│  sat_L3 ──► + LoRA      ──► z₃ ─┘                 │                          │
│                                                    │                          │
│         g = σ(α_g)·s_img + (1−σ(α_g))·s_txt    GatedFusion_g                │
│                            │                                                  │
│                     ĝ = g / ‖g‖₂  ──► gallery [B,512]                        │
└───────────────────────────────────────────────────────────────────────────────┘

┌────────────────────────────── RETRIEVAL ──────────────────────────────────────┐
│                                                                               │
│  similarity = q̂ · ĝᵀ / τ      (τ learnable, init=0.07, clamp [0.01, 0.5])   │
│  loss = 0.6·CE(q→g) + 0.4·CE(g→q)     (label_smoothing=0.1)                 │
│                                                                               │
│  Retrieval space: 512-dim (DINOv3 1024 → projection → 512)                   │
│  All shared: one DINOv3, one MONA set, one DGTRS-CLIP, one TextFusionMLP     │
│  For sat images without captions: s_txt=None → g = s_img (gate passthrough)   │
│  BASELINE: σ(α) = 1.0 for both branches (text disabled, DGTRS not loaded)    │
└───────────────────────────────────────────────────────────────────────────────┘

┌──────────────────────────── DIAGNOSTICS PIPELINE ────────────────────────────┐
│                                                                               │
│  Per-batch ──► CSV (train_batches.csv)  ──► TensorBoard / W&B scalars        │
│  Per-epoch ──► CSV (train.csv, val.csv) ──► Seaborn plots (PNG)              │
│  Every N epochs ──► Grad-CAM heatmaps (drone + satellite DINOv3 last block)  │
│  Epoch 0 (opt) ──► PyTorch Profiler (Chrome trace + CUDA timeline)           │
│  Per-50-batch ──► Gradient norms per group (MONA, LoRA, MLP, gates, τ)       │
│  On init ──► torchinfo model summary (model_summary.txt)                     │
└───────────────────────────────────────────────────────────────────────────────┘
```

### Text hierarchy (L1 / L2 / L3)

Each drone image has a VLM-generated caption (Qwen3-VL) split into 3 levels:

| Level | Name | Content | Typical length |
|-------|------|---------|----------------|
| **L1** | Overview | First sentence of P1: land-cover summary with class percentages | 15–30 tokens |
| **L2** | Full description | Complete P1 (inventory) + P2 (spatial layout with 5+ zones) | 100–200 tokens |
| **L3** | Fingerprint | P3: unique landmarks and spatial signature for matching | 20–50 tokens |

All three levels are encoded by a **single DGTRS-CLIP ViT-L-14** text encoder
(248-token context via KPS positional embedding, 768-dim output)
adapted with **LoRA** (rank=4 on Q/V in all 12 blocks).

### Text fusion (shared MLP for both branches)

```math
\mathbf{z}_{\text{text}} = \text{MLP}\bigl([\mathbf{z}_1 \;;\; \mathbf{z}_2 \;;\; \mathbf{z}_3]\bigr)
```

where `[z₁ ; z₂ ; z₃] ∈ ℝ^{B×2304}` is the concatenation of three 768-dim DGTRS-CLIP embeddings, and

```math
\text{MLP}: \text{Linear}(2304, 512) \to \text{GELU} \to \text{Linear}(512, 512), \quad \mathbf{z}_{\text{text}} \in \mathbb{R}^{B \times 512}
```

### Gated fusion (separate gates for query and gallery)

```math
\mathbf{q} = \sigma(\alpha_q) \cdot \mathbf{d}_{\text{img}} + \bigl(1 - \sigma(\alpha_q)\bigr) \cdot \mathbf{d}_{\text{txt}} \qquad \text{(query branch)}
```

```math
\mathbf{g} = \sigma(\alpha_g) \cdot \mathbf{s}_{\text{img}} + \bigl(1 - \sigma(\alpha_g)\bigr) \cdot \mathbf{s}_{\text{txt}} \qquad \text{(gallery branch)}
```

- `α_q, α_g` — separate learnable scalars in logit-space, init `σ(α) ≈ 0.7`
- `σ` — sigmoid function
- `d_img, s_img ∈ ℝ^{B×512}` — DINOv3+MONA → projection(1024→512)
- `d_txt, s_txt ∈ ℝ^{B×512}` — fused text embeddings (TextFusionMLP → 512)
- For satellite images without captions: `s_txt = None → g = s_img`

### Adaptation methods

| Method | Applied to | Where | Params |
|--------|-----------|-------|--------|
| **MONA** (CVPR 2025) | DINOv3 ViT-L/16 (shared) | After MSA and MLP in each of 24 blocks | 6.85M |
| **LoRA** (rank=4) | DGTRS-CLIP text encoder | Q and V projections in all 12 blocks | 147K |
| **Projection** | After DINOv3 CLS output | Linear(1024→512) | 525K |

**MONA adapter** (per block):

```math
\mathbf{x} \leftarrow \mathbf{x} + \text{Up}_{64 \to 1024}\!\Bigl(\text{GELU}\bigl(\text{MonaOp}\bigl(\text{Down}_{1024 \to 64}(\hat{\mathbf{x}})\bigr)\bigr)\Bigr)
```

where `x̂ = γ · LN(x) + γₓ · x` (scaled LayerNorm, `γ` init `10⁻⁶`, `γₓ` init `1`)

```math
\text{MonaOp}(\mathbf{x}) = \frac{\text{DWConv}_{3 \times 3}(\mathbf{x}) + \text{DWConv}_{5 \times 5}(\mathbf{x}) + \text{DWConv}_{7 \times 7}(\mathbf{x})}{3} + \mathbf{x}
```

MONA runs in **bfloat16** with gradient checkpointing to save VRAM.
Applied only to the **last 12 blocks** (out of 24) — early blocks extract low-level
features (edges, textures) that are domain-agnostic and don't need spatial adaptation.

**Why bfloat16, not fp16:**

MONA's scaled LayerNorm uses `gamma` initialized at `1e-6` for near-identity output at start.
fp16 has min normal ~6.1e-5, so `1e-6` falls into the subnormal range where precision collapses,
causing NaN after a few blocks. bfloat16 has the same exponent range as fp32 (min ~1.2e-38),
so `1e-6` is safely representable. RTX 4090 supports bf16 natively with comparable throughput.

| Precision | `1e-6` representable | MONA stable | VRAM (bs=48) |
|-----------|:---:|:---:|:---:|
| fp32 | yes | yes | 21.4 GB |
| **bfloat16** | **yes** | **yes** | **21.8 GB** |
| fp16 | subnormal (lossy) | **NaN** | — |

**Why MONA over LoRA for DINOv3:**

MONA uses multi-scale depthwise convolutions (3×3, 5×5, 7×7) that provide **spatial inductive bias**
critical for cross-view geo-localization. Drone images (oblique, 100-600m altitude) and satellite
images (nadir) exhibit a strong **geometric domain gap** — the same building looks spatially different
from each viewpoint. MONA's multi-scale spatial filters learn scale-invariant features to bridge
this gap, while LoRA (pure linear low-rank correction) would only handle style/distribution shifts.

| | MONA | LoRA (on DINOv3) |
|---|---|---|
| Inductive bias | 2D spatial (knows about pixel neighbors) | None (linear correction) |
| Best for | Geometric domain gap (aerial↔satellite) | Style/distribution shift |
| Params | 6.85M (bottleneck=64) | ~0.3M (rank=4) |
| Compute | Heavier (192 conv ops per forward) | Light |
| CVGL fit | Strong (multi-scale spatial adaptation) | Weak (no spatial awareness) |

**LoRA** (per DGTRS-CLIP attention layer):

```math
\mathbf{Q}' = \mathbf{Q} + \frac{\alpha}{r} \cdot \mathbf{x} \mathbf{A}_Q^T \mathbf{B}_Q^T, \qquad \mathbf{V}' = \mathbf{V} + \frac{\alpha}{r} \cdot \mathbf{x} \mathbf{A}_V^T \mathbf{B}_V^T
```

where `A ∈ ℝ^{r×d}`, `B ∈ ℝ^{d×r}`, `r = 4`. LoRA is appropriate for the text encoder since
text has no spatial structure — the adaptation needed is purely semantic/distributional.

### Projection head (DINOv3 1024 → 512)

```math
\mathbf{h} = \text{Linear}_{1024 \to 512}\!\bigl(\text{CLS}_{\text{DINOv3}}\bigr)
```

Reduces the retrieval space from DINOv3 native 1024-dim to 512-dim. Benefits:
- Smaller similarity matrix in InfoNCE (`B×B` at 512 vs 1024)
- TextFusionMLP outputs 512 instead of 1024 (fewer params: 1.5M vs 3.4M)
- Shared projection for both drone and satellite branches

### Pair formation and negative sampling

The dataset provides only **positive pairs** — each entry maps one drone image to its
matching satellite crop(s). Negative pairs are **not** stored explicitly; instead, they
are constructed automatically inside each training batch via the InfoNCE loss:

```
Batch (B = 8):
  drone_0 ↔ sat_0  ← positive (from dataset)
  drone_1 ↔ sat_1  ← positive
  ...
  drone_7 ↔ sat_7  ← positive

Similarity matrix B×B:
       sat_0  sat_1  ...  sat_7
  q_0 [ pos    neg   ...   neg ]   ← CE target = 0
  q_1 [ neg    pos   ...   neg ]   ← CE target = 1
  ...
  q_7 [ neg    neg   ...   pos ]   ← CE target = 7
```

- **Positives:** diagonal `sim[i, i]` — correct drone-satellite pair
- **Negatives:** off-diagonal `sim[i, j≠i]` — other satellite images in the batch (in-batch negatives)
- At `batch_size=8`: 1 positive + 7 in-batch negatives per query
- No hard-negative mining — negatives are random within the batch
- Larger batch → more negatives → harder contrastive task

Each drone image may have multiple satellite candidates (with distance-based weights).
At training time, one satellite crop is sampled per drone (weighted if semi-positive).

### Loss function

Symmetric InfoNCE with learnable temperature (CLIP-style `logit_scale`):

```math
\mathcal{L} = w_{q \to g} \cdot \mathcal{L}_{q \to g} + w_{g \to q} \cdot \mathcal{L}_{g \to q}
```

```math
\mathcal{L}_{q \to g} = \text{CrossEntropy}\!\left(\frac{\hat{\mathbf{q}} \cdot \hat{\mathbf{g}}^T}{\tau},\; \text{targets}\right), \qquad \mathcal{L}_{g \to q} = \text{CrossEntropy}\!\left(\frac{\hat{\mathbf{g}} \cdot \hat{\mathbf{q}}^T}{\tau},\; \text{targets}\right)
```

- `τ = 1 / exp(logit_scale)` — learnable scalar, clamped `τ ∈ [0.01, 0.5]`, init `τ₀ = 0.07`
- `w_{q→g} = 0.6`, `w_{g→q} = 0.4`
- `targets = [0, 1, 2, ..., B−1]` — positives on diagonal
- label smoothing `= 0.1`

Loss and adapters run in **fp32** (AMP autocast disabled) to prevent gradient overflow.

### Gradient checkpointing

Gradient checkpointing trades compute for VRAM by recomputing activations during
backward instead of storing them. Enabled by default (`gradient_checkpointing=True`).

| Component | Without checkpointing | With checkpointing |
|-----------|:---:|:---:|
| DINOv3 (24 blocks) | stores all 24 block activations | recomputes on backward |
| DGTRS-CLIP (12 blocks) | stores all 12 block activations | recomputes on backward |
| **Max batch_size** (RTX 4090, shared encoder) | **8** | **24** |
| Speed penalty | — | ~20-30% slower per step |

VRAM tested on RTX 4090 (24 GB) with shared DINOv3 WEB + MONA top-12 bf16 + DGTRS-CLIP + text:

| `batch_size` | Peak VRAM | Status |
|:---:|:---:|:---:|
| 32 | 16.1 GB | OK |
| 40 | 19.1 GB | OK |
| **48** | **21.8 GB** | **OK (recommended)** |
| 56 | >24 GB | OOM |

### Gradient accumulation

Gradient accumulation emulates a larger effective batch without extra memory:

```
effective_batch_size = batch_size × grad_accum_steps
```

| Setting | `batch_size` | `grad_accum_steps` | Effective batch | In-batch negatives |
|---------|:---:|:---:|:---:|:---:|
| Default | 48 | 1 | 48 | 47 |
| Large effective batch | 48 | 4 | 192 | 47 per micro-batch |

**Note:** gradient accumulation averages gradients across micro-batches, but each
micro-batch still only sees `batch_size` in-batch negatives. To increase the number
of negatives per forward pass, increase `batch_size` directly (requires more VRAM).

```bash
# Example: effective batch of 192 with gradient accumulation
python -m src.training.train_gtauav --config conf/gtauav_balanced.gin \
    --filter-meta meta/seg_filter.json --batch-size 48 --grad-accum 4
```

### Metrics

| Metric | Formula | Direction | Computed on |
|--------|---------|-----------|:-----------:|
| **R@K** (Recall at K) | fraction of queries where correct gallery is in top-K | q→g and g→q | train + val |
| **AP** (Average Precision) | mean of 1/(rank+1) across all queries (MRR) | q→g and g→q | train + val |
| **Loss** (InfoNCE) | symmetric cross-entropy on similarity matrix | — | train + val |
| **Delta R@1** | R@1(with_text) − R@1(baseline) | q→g | val only |

R@1, R@5, R@10, AP, and loss are computed on both **train** (subset matching test size, clean
transforms) and **val** (full test set) every epoch. Train vs val comparison enables overfitting
detection: if train R@1/AP rises while val stagnates → overfitting.

**Output CSVs:**

| File | Content | Updated |
|------|---------|---------|
| `logs/train.csv` | Epoch-level train loss, temperature, gates, lr | Every epoch |
| `logs/val.csv` | Val R@K, AP, loss, gates | Every eval epoch |
| `logs/train_recall.csv` | Train R@K, AP, loss (subset) | Every eval epoch |
| `logs/train_batches.csv` | Per-batch loss, temperature, gates, lr | Every batch |

**Plots** (auto-generated in `logs/`):

| Plot | Panels |
|------|--------|
| `train_metrics.png` | Loss, temperature (τ), gates (σ(α)), learning rate |
| `val_metrics.png` | R@K q→g (train vs val), R@K g→q (train vs val), AP (train vs val) |
| `overview.png` | Train+val loss, val R@1, gates + temperature |

### Optimizer & scheduler

```
Optimizer: AdamW
  - MONA adapters, projection, TextFusionMLP, gate α_q, gate α_g, logit_scale:
      lr = 1e-4,  weight_decay = 1e-4
  - LoRA adapters (DGTRS-CLIP text encoder):
      lr = 1e-5  (10× lower, --text-lr-factor 0.1)

Scheduler: Linear warmup (2 epochs) + cosine annealing
  - Per optimizer step (accounts for gradient accumulation)
  - warmup: lr linearly ramps from 0 to lr_max over warmup_steps
  - cosine: lr decays from lr_max to 0 over remaining steps

Gradient clipping: max_norm = 1.0
Gradient accumulation: configurable (default 1, --grad-accum N)
Mixed precision: AMP fp16 for DINOv3/DGTRS forward, bf16 for MONA, fp32 for loss
Gradient checkpointing: DINOv3 (24 blocks) + DGTRS-CLIP (12 blocks)
```

### Augmentations

| Transform | Drone (train) | Satellite (train) | Eval |
|-----------|:---:|:---:|:---:|
| RandomResizedCrop(256, scale=0.7–1.0) | ✓ | ✓ | — |
| Resize(256) + CenterCrop(256) | — | — | ✓ |
| RandomHorizontalFlip(0.5) | ✓ | ✓ | — |
| RandomRotation(15°) | ✓ | — | — |
| ColorJitter(0.3, 0.3, 0.2, 0.1) | ✓ | ✓ | — |
| RandomGrayscale(0.05) | ✓ | ✓ | — |
| GaussianBlur(k=3, σ=0.1–2.0) | ✓ | — | — |
| ImageNet Normalize | ✓ | ✓ | ✓ |

### Model summary

| Component | Params | Trainable | Notes |
|-----------|--------|-----------|-------|
| DINOv3 ViT-L/16 WEB (shared) | 303M | frozen | single encoder for drone + satellite |
| MONA adapters (shared, bf16) | 3.5M | 3.5M | 2 per block × last 12 blocks, bottleneck=64 |
| Image projection | 525K | 525K | Linear(1024→512) after DINOv3 CLS |
| DGTRS-CLIP ViT-L-14 (text) | 124M | frozen | backbone weights frozen |
| LoRA adapters (text) | 147K | 147K | Q+V, rank=4, 12 blocks |
| TextFusionMLP (shared) | 1.5M | 1.5M | Linear(2304,512) + GELU + Linear(512,512) |
| GatedFusion α_q + α_g | 2 | 2 | separate gate scalars |
| logit_scale | 1 | 1 | learnable temperature |
| **Total (shared)** | **432M** | **5.6M (1.30%)** | retrieval dim = 512 |

> **Asymmetric mode** (`shared_encoder=False`): separate DINOv3 WEB (drone) + DINOv3 SAT
> (satellite) encoders with independent MONA adapters. Requires ~4-5 GB more VRAM.
> Use `conf/gtauav_balanced_asym.gin` / `conf/gtauav_baseline_asym.gin` — these set
> `shared_encoder=False` and `mona_last_n_blocks=24` for the full-capacity setup.

### Eval directions

`_evaluate` computes R@1/5/10 and AP (MRR) for both retrieval directions:

| Key | Direction | Notes |
|-----|-----------|-------|
| `r@K_q2g`, `ap_q2g` | drone → satellite (query → gallery) | Denominator: all queries (q2g convention) |
| `r@K_g2q`, `ap_g2q` | satellite → drone (gallery → query) | Denominator: only sat-tiles with ≥1 positive drone in subsample |

g2q is computed via inverted ground truth: for each sat-tile, collect the drone indices
that list it as a valid candidate. `n_scored_g2q` is reported in metrics for transparency.

## Experiments

### V3 — GTA-UAV + DINOv3 + DGTRS-CLIP (active)

**Dataset:** GTA-UAV-LR (33K drone + 14K satellite, GTA V synthetic)
- RGB: `/home/servml/Документы/datasets/GTA-UAV-LR/`
- Captions: `/home/servml/Документы/datasets/GTA-UAV-LR-captions/` (40K JSON, 3-paragraph VLM)
- Segmentation: `/home/servml/Документы/datasets/GTA-UAV-LR-aug/` (17 classes)
- Seg filter: 37,498 passed / 10,905 excluded (>=90% background+water)
- Split: 80/20 random (26,966 train / 6,742 test → 24,891/6,252 after seg filter)

### V2 — UAV-GeoLoc + GeoRSCLIP (legacy)

Single backbone with template captions.

```
Query:   drone_img + caption -> GeoRSCLIP -> GatedFusion -> query
Gallery: sat_img -> GeoRSCLIP -> gallery
```

**Dataset:** UAV-GeoLoc Terrain split (206K train queries)

## Structure

```
caption-test/
├── conf/                         # Gin configs
│   ├── gtauav_balanced.gin       # GTA-UAV with text, shared encoder, MONA 12/24 (v3)
│   ├── gtauav_baseline.gin       # GTA-UAV baseline, shared, MONA 12/24, no text (v3)
│   ├── gtauav_balanced_asym.gin  # GTA-UAV with text, asymmetric WEB+SAT, MONA 24/24
│   ├── gtauav_baseline_asym.gin  # GTA-UAV baseline, asymmetric, MONA 24/24, no text
│   ├── gtauav_text_heavy.gin     # GTA-UAV text-heavy gate=0.3 (v3)
│   ├── gtauav_image_heavy.gin    # GTA-UAV image-heavy gate=0.9 (v3)
│   ├── balanced.gin              # UAV-GeoLoc with text (v2)
│   ├── baseline_no_text.gin      # UAV-GeoLoc baseline (v2)
│   └── text_heavy.gin            # UAV-GeoLoc text-heavy (v2)
├── nn_models/                    # Pre-trained checkpoints (v3, gitignored)
│   ├── DINO_WEB/                 # DINOv3 ViT-L/16 LVD-1689M (.pth)
│   ├── DINO_SAT/                 # DINOv3 ViT-L/16 SAT-493M (.safetensors)
│   └── LRSCLIP/                  # DGTRS-CLIP ViT-L-14 (.pt)
├── meta/                         # Generated metadata
│   ├── train_80.json             # 80% train split (26,966 pairs)
│   ├── test_20.json              # 20% test split (6,742 pairs)
│   └── seg_filter.json           # Segmentation filter results
├── scripts/
│   ├── make_split.py             # Create 80/20 train/test split
│   ├── filter_segmentation.py    # Exclude 90%+ background/water images
│   ├── compare_runs.py           # Delta R@1 comparison report
│   └── generate_captions.py      # Offline caption generation
├── src/
│   ├── datasets/
│   │   ├── gtauav_dataset.py     # GTA-UAV-LR loader + L1/L2/L3 parsing (v3)
│   │   └── visloc_with_captions.py  # UAV-GeoLoc loader (v2)
│   ├── models/
│   │   ├── dgtrs/                # Official DGTRS-CLIP text encoder (Apache-2.0)
│   │   │   ├── model.py          # DGTRSTextEncoder, build_model, tokenize
│   │   │   ├── simple_tokenizer.py  # BPE tokenizer (248 tokens)
│   │   │   └── bpe_simple_vocab_16e6.txt.gz
│   │   │   ├── adapters.py          # MONA (bf16) + LoRA adapters
│   │   ├── asymmetric_encoder.py # DINOv3 + projection + DGTRS + GatedFusion (v3)
│   │   └── dual_encoder.py       # GeoRSCLIP + GatedFusion (v2)
│   ├── losses/
│   │   └── multi_infonce.py      # InfoNCE with learnable temperature
│   ├── training/
│   │   ├── train_gtauav.py       # Training loop GTA-UAV (v3)
│   │   ├── train.py              # Training loop UAV-GeoLoc (v2)
│   │   ├── trackers.py           # Unified tracker: W&B + TensorBoard
│   │   ├── grad_monitor.py       # Gradient norm monitoring per group
│   │   ├── gradcam.py            # Grad-CAM visualization for DINOv3
│   │   ├── profiling.py          # PyTorch Profiler + torchinfo summary
│   │   └── plot_metrics.py       # Seaborn/matplotlib metric plots
│   └── eval/
│       └── evaluate.py           # R@K metrics, Delta R@1
└── checkpoints/                  # GeoRSCLIP RS5M_ViT-B-32.pt (v2)
```

## Prerequisites

```
torch>=2.0
safetensors
coloredlogs
tqdm
ftfy
regex
gin-config
Pillow
numpy
pandas
matplotlib
seaborn
```

### Optional (for extended diagnostics)

```
wandb          # Weights & Biases experiment tracking
torchinfo      # Model summary tables
tensorboard    # TensorBoard logging (included with torch)
```

## Workflow (V3 — GTA-UAV)

### 1. Create 80/20 split and filter segmentation

```bash
python -m scripts.make_split --output-dir meta
python -m scripts.filter_segmentation --output meta/seg_filter.json
```

### 2. Train with gin configs (recommended)

```bash
# With captions (L1/L2/L3, bs=48, 30 epochs, eval every epoch)
python -m src.training.train_gtauav --config conf/gtauav_balanced.gin \
    --filter-meta meta/seg_filter.json --batch-size 48 --epochs 30 \
    --gin-param 'TrainConfigGTAUAV.eval_every=1'

# Baseline (no text)
python -m src.training.train_gtauav --config conf/gtauav_baseline.gin \
    --filter-meta meta/seg_filter.json --batch-size 48 --epochs 30

# Text-heavy (gate=0.3, 70% text weight)
python -m src.training.train_gtauav --config conf/gtauav_text_heavy.gin \
    --filter-meta meta/seg_filter.json --batch-size 48 --epochs 30

# Image-heavy (gate=0.9, 10% text weight)
python -m src.training.train_gtauav --config conf/gtauav_image_heavy.gin \
    --filter-meta meta/seg_filter.json --batch-size 48 --epochs 30

# Asymmetric variants: separate WEB (drone) + SAT (satellite) encoders, MONA in all 24 blocks
# Higher capacity (~733M total / ~17.6M trainable), larger VRAM footprint.
python -m src.training.train_gtauav --config conf/gtauav_balanced_asym.gin \
    --filter-meta meta/seg_filter.json
python -m src.training.train_gtauav --config conf/gtauav_baseline_asym.gin \
    --filter-meta meta/seg_filter.json
```

### 3. Train without gin (CLI-only)

```bash
python -m src.training.train_gtauav --baseline --filter-meta meta/seg_filter.json --batch-size 48
python -m src.training.train_gtauav --filter-meta meta/seg_filter.json --batch-size 48
```

### 4. Enable diagnostics

```bash
# W&B + Grad-CAM + PyTorch Profiler
python -m src.training.train_gtauav --config conf/gtauav_balanced.gin \
    --filter-meta meta/seg_filter.json --batch-size 48 --wandb --gradcam --profile

# Gin parameter overrides from CLI
python -m src.training.train_gtauav --config conf/gtauav_balanced.gin \
    --filter-meta meta/seg_filter.json \
    --gin-param 'TrainConfigGTAUAV.eval_every=1' 'TrainConfigGTAUAV.epochs=30'
```

CLI flags (`--wandb`, `--gradcam`, `--profile`, `--epochs`, `--batch-size`, etc.) take priority over gin config.

### 5. Resume from checkpoint

```bash
python -m src.training.train_gtauav --resume out/gtauav/with_text/ckpt_epoch004.pt \
    --filter-meta meta/seg_filter.json
```

### 6. Compare and get verdict

```bash
python -m scripts.compare_runs \
    --baseline_report out/gtauav/baseline/eval_report.json \
    --full_report out/gtauav/with_text/eval_report.json \
    --output out/gtauav/comparison.md
```

### 7. View TensorBoard

```bash
tensorboard --logdir out/gtauav/with_text/tb_logs
```

## Diagnostics & Visualization

| Tool | Flag | Output | Description |
|------|------|--------|-------------|
| **TensorBoard** | `--use-tb` (default on) | `{out}/tb_logs/` | Scalars, histograms, images |
| **W&B** | `--wandb` | cloud | Full experiment tracking, Grad-CAM images |
| **Grad-CAM** | `--gradcam` | `{out}/gradcam/` | DINOv3 attention heatmaps (drone + satellite) |
| **PyTorch Profiler** | `--profile` | `{out}/profiler/` | Chrome trace, CUDA timeline, memory |
| **torchinfo** | auto | `{out}/model_summary.txt` | Layer-by-layer parameter table |
| **Gradient norms** | `--log-grad-norms` (default on) | TB/W&B | Per-group: MONA, LoRA, MLP, gates, tau |
| **CSV (per-batch)** | auto | `{out}/logs/train_batches.csv` | Loss, tau, gates, lr for every batch |
| **CSV (per-epoch)** | auto | `{out}/logs/train.csv, val.csv` | Epoch loss averages + seaborn plots |
| **CSV (recall)** | auto | `{out}/logs/train_recall.csv` | Train R@K, AP, loss (subset, clean transforms) |
| **Plots** | auto | `{out}/logs/*.png` | train/val loss, R@K, AP, gates, temperature |

## Decision rule

| Delta R@1 (drone→satellite) | Verdict |
|---|---|
| >= +3% | **PASS** — captions informative, proceed to NADEZHDA teacher |
| +1% to +3% | MARGINAL — add VLM refinement, re-run |
| 0 to +1% | WEAK — redesign caption pipeline |
| < 0 | HARMFUL — critical bug |

## Code style

- `from __future__ import annotations` everywhere
- Type hints on all signatures
- Google-style docstrings
- English-only comments