File size: 9,128 Bytes
f961d5d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
# BirdNET Uploader Usage Guide

This guide covers how to download, install, and use the BirdNET Uploader CLI for uploading audio datasets to Hugging Face.

## Installation

### Option 1: Windows Standalone (Recommended)

1. Download the latest release from [Releases](https://huggingface.co/spaces/jrrribeiro/BirdNET-Uploader-App/discussions)
2. Extract the ZIP file to a folder
3. No Python installation required!
4. Run: `birdnet-uploader.exe --help`

### Option 2: Python + Pip

```bash
pip install birdnet-uploader
birdnet-uploader --help
```

### Option 3: Docker

```bash
docker pull birdnet-uploader:latest
docker run --rm \
  -e HF_TOKEN="hf_your_token" \
  -v C:\segments:C:\data\segments \
  birdnet-uploader:latest \
  upload --repo-id user/dataset --segments C:\data\segments
```

### Option 4: Build from Source

```bash
git clone https://github.com/jrrribeiro/BirdNET-Uploader-App
cd BirdNET-Uploader-App
pip install -r requirements.txt
python -m src.uploader_cli.main --help
```

## Quick Start (5 Minutes)

### Step 1: Authenticate

Store your Hugging Face token securely:

```bash
birdnet-uploader login
# Follow the prompt to enter your token
# Token is stored securely in your OS credential manager
```

### Step 2: Create/Validate Dataset

```bash
birdnet-uploader init-repo --repo-id username/my-dataset --private
# Creates or validates the dataset exists
```

### Step 3: Prepare Audio Files

Organize your audio files in a folder:

```
C:\my-segments\
β”œβ”€β”€ species-A\
β”‚   β”œβ”€β”€ recording-001.wav
β”‚   β”œβ”€β”€ recording-002.wav
β”‚   └── ...
β”œβ”€β”€ species-B\
β”‚   β”œβ”€β”€ recording-003.wav
β”‚   └── ...
└── detections.csv  (optional)
```

### Step 4: Dry Run (Recommended)

Preview what will be uploaded without actually uploading:

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\my-segments \
  --dry-run
```

### Step 5: Upload

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\my-segments \
  --csv detections.csv
```

Done! Audio files are now in `audio/` and CSV is in `index/detections.csv`.

## Common Workflows

### Upload with Progress Tracking

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\my-segments \
  --workers 4
```

The `--workers` option enables parallel uploads (faster on multi-core systems).

### Resume Interrupted Upload

```bash
# Check existing sessions
ls ~/.birdnet-uploader/sessions/

# Resume a specific session
birdnet-uploader resume upload-20260429T120000Z

# Or start fresh
birdnet-uploader upload --repo-id username/my-dataset --segments C:\my-segments
```

### Upload with Custom Token (CI/CD)

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\my-segments \
  --token hf_your_token
```

### Batch Upload Multiple Folders

```powershell
$folders = @("C:\data\2024", "C:\data\2025")
foreach ($folder in $folders) {
  birdnet-uploader upload `
    --repo-id username/my-dataset `
    --segments $folder
}
```

### Upload Only CSV (Index Only)

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\dummy-empty \
  --csv detections.csv \
  --dry-run
```

## Token Management

### Method 1: Stored Token (Recommended for Local Use)

```bash
# Store token securely (OS Credential Manager)
birdnet-uploader login

# Use stored token automatically
birdnet-uploader upload --repo-id username/my-dataset --segments C:\segments
```

### Method 2: Environment Variable (Docker/CI)

```bash
# Linux/macOS
export HF_TOKEN="hf_..."
birdnet-uploader upload ...

# Windows PowerShell
$env:HF_TOKEN = "hf_..."
birdnet-uploader.exe upload ...

# Windows CMD
set HF_TOKEN=hf_...
birdnet-uploader.exe upload ...
```

### Method 3: CLI Option (One-Off)

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\segments \
  --token hf_...
```

## Advanced Usage

### Parallel Uploads

Speed up uploads on multi-core systems:

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\segments \
  --workers 8
```

Recommended values:
- 4-8 workers for residential/office internet
- 16-32 workers for datacenter/gigabit internet

### Custom Dataset Path

Upload to a subdirectory in the dataset:

```bash
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\segments \
  --remote-base audio/my-collection
```

Files uploaded to: `audio/my-collection/...` instead of `audio/...`

### Session Management

```bash
# List all sessions
ls ~/.birdnet-uploader/sessions/

# View session checkpoint
birdnet-uploader resume upload-20260429T120000Z

# Remove old session (careful!)
rm -r ~/.birdnet-uploader/sessions/upload-20260101T000000Z
```

### Clear Deduplication Cache

If you need to re-upload files that were marked as already uploaded:

```bash
# Remove cache (forces re-upload check)
rm -r ~/.birdnet-uploader/cache/

# Then retry upload
birdnet-uploader upload --repo-id username/my-dataset --segments C:\segments
```

## Troubleshooting

### "No stored token found"

**Problem**: You haven't logged in yet.

**Solution**:
```bash
birdnet-uploader login
# Enter your Hugging Face token when prompted
```

### "Token validation failed"

**Problem**: Token is invalid or expired.

**Solution**:
1. Go to https://huggingface.co/settings/tokens
2. Create a new token
3. Run `birdnet-uploader login` and enter the new token
4. Or use `--token hf_new_token` directly

### "Dataset not found"

**Problem**: Dataset doesn't exist or you don't have access.

**Solution**:
```bash
# Create the dataset
birdnet-uploader init-repo --repo-id username/my-dataset

# Or verify access
curl -H "Authorization: Bearer hf_..." \
  https://huggingface.co/api/datasets/username/my-dataset
```

### "Upload stuck or slow"

**Problem**: Network issues or single-threaded upload.

**Solution**:
```bash
# Use parallel workers
birdnet-uploader upload \
  --repo-id username/my-dataset \
  --segments C:\segments \
  --workers 4
```

### "DLL not found" (Windows executable)

**Problem**: Missing runtime dependencies in bundled version.

**Solution**:
1. Download latest release (should include all DLLs)
2. Or run from Python: `python -m src.uploader_cli.main upload ...`
3. Or report issue on GitHub

### Out of Disk Space

**Problem**: Destination dataset is full.

**Solution**:
1. Check dataset size: `huggingface_hub`
2. Delete old uploads if not needed
3. Create a new dataset: `birdnet-uploader init-repo --repo-id username/my-dataset-2`

## Dataset Structure

After upload, your Hugging Face dataset will look like:

```text
my-dataset/
β”œβ”€β”€ audio/
β”‚   β”œβ”€β”€ species-A/
β”‚   β”‚   β”œβ”€β”€ recording-001.wav
β”‚   β”‚   β”œβ”€β”€ recording-002.wav
β”‚   β”‚   └── ...
β”‚   └── species-B/
β”‚       β”œβ”€β”€ recording-003.wav
β”‚       └── ...
β”œβ”€β”€ index/
β”‚   β”œβ”€β”€ manifest.json          # Index metadata
β”‚   β”œβ”€β”€ detections.csv         # Your CSV (if provided)
β”‚   └── shards/
β”‚       β”œβ”€β”€ detections_0.parquet
β”‚       └── detections_1.parquet
└── README.md
```

## Performance Tips

1. **Use SSD for local segments**: Faster scanning and hashing
2. **Enable parallel uploads**: `--workers 4` or higher
3. **Stable network**: Wired internet > WiFi
4. **Large batches**: Consider `--batch-size 100` for many files
5. **Monitor system resources**: CPU and memory usage during upload

## Security

- **Never hardcode tokens**: Use `birdnet-uploader login` or `HF_TOKEN` env var
- **Rotate tokens**: Generate new token every 3-6 months at https://huggingface.co/settings/tokens
- **Limit token scope**: Use fine-grained tokens with minimal permissions
- **Don't share tokens**: Each user should have their own token
- **Use dataset privacy**: Set `--private` when creating repos with sensitive data

## Getting Help

- Check [TOKEN_SECURITY.md](docs/TOKEN_SECURITY.md) for token security best practices
- Check [TOKEN_USAGE_EXAMPLES.md](docs/TOKEN_USAGE_EXAMPLES.md) for Docker/CI/CD examples
- Check [CI_CD_PIPELINE.md](docs/CI_CD_PIPELINE.md) for build and release info
- Report issues: Create a GitHub issue with error logs and reproduction steps

## FAQ

**Q: Can I cancel an upload?**
A: Yes, press Ctrl+C. The session checkpoint saves progress, so you can resume later with the same command.

**Q: How large can the audio collection be?**
A: Hugging Face datasets support large collections (100GB+). Upload speed depends on your internet.

**Q: Can multiple users upload to the same dataset?**
A: Yes, but they need write access. Use separate tokens for each user to track uploads separately.

**Q: Does it support audio formats other than WAV?**
A: Yes: WAV, MP3, FLAC, OGG, M4A. Extend `AUDIO_EXTENSIONS` in `src/uploader_cli/config.py` if needed.

**Q: Can I upload without audio files (CSV only)?**
A: Yes, but you need to provide a dummy `--segments` directory (can be empty).

**Q: How do I verify upload integrity?**
A: The uploader computes SHA-256 hashes and deduplicates against remote index. Files are verified on upload.