File size: 13,089 Bytes
a3ffd31
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
'''
Downloads models from Hugging Face to models/username_modelname.

Example:
python download-model.py facebook/opt-1.3b

'''

import argparse
import base64
import datetime
import hashlib
import json
import os
import re
import sys
from pathlib import Path

import requests
import tqdm
from requests.adapters import HTTPAdapter
from tqdm.contrib.concurrent import thread_map

base = "https://huggingface.co"


class ModelDownloader:
    def __init__(self, max_retries=5):
        self.session = requests.Session()
        if max_retries:
            self.session.mount('https://cdn-lfs.huggingface.co', HTTPAdapter(max_retries=max_retries))
            self.session.mount('https://huggingface.co', HTTPAdapter(max_retries=max_retries))

        if os.getenv('HF_USER') is not None and os.getenv('HF_PASS') is not None:
            self.session.auth = (os.getenv('HF_USER'), os.getenv('HF_PASS'))

        try:
            from huggingface_hub import get_token
            token = get_token()
        except ImportError:
            token = os.getenv("HF_TOKEN")

        if token is not None:
            self.session.headers = {'authorization': f'Bearer {token}'}

    def sanitize_model_and_branch_names(self, model, branch):
        if model[-1] == '/':
            model = model[:-1]

        if model.startswith(base + '/'):
            model = model[len(base) + 1:]

        model_parts = model.split(":")
        model = model_parts[0] if len(model_parts) > 0 else model
        branch = model_parts[1] if len(model_parts) > 1 else branch

        if branch is None:
            branch = "main"
        else:
            pattern = re.compile(r"^[a-zA-Z0-9._-]+$")
            if not pattern.match(branch):
                raise ValueError(
                    "Invalid branch name. Only alphanumeric characters, period, underscore and dash are allowed.")

        return model, branch

    def get_download_links_from_huggingface(self, model, branch, text_only=False, specific_file=None):
        page = f"/api/models/{model}/tree/{branch}"
        cursor = b""

        links = []
        sha256 = []
        classifications = []
        has_pytorch = False
        has_pt = False
        has_gguf = False
        has_safetensors = False
        is_lora = False
        while True:
            url = f"{base}{page}" + (f"?cursor={cursor.decode()}" if cursor else "")
            r = self.session.get(url, timeout=10)
            r.raise_for_status()
            content = r.content

            dict = json.loads(content)
            if len(dict) == 0:
                break

            for i in range(len(dict)):
                fname = dict[i]['path']
                if specific_file not in [None, ''] and fname != specific_file:
                    continue

                if not is_lora and fname.endswith(('adapter_config.json', 'adapter_model.bin')):
                    is_lora = True

                is_pytorch = re.match(r"(pytorch|adapter|gptq)_model.*\.bin", fname)
                is_safetensors = re.match(r".*\.safetensors", fname)
                is_pt = re.match(r".*\.pt", fname)
                is_gguf = re.match(r'.*\.gguf', fname)
                is_tiktoken = re.match(r".*\.tiktoken", fname)
                is_tokenizer = re.match(r"(tokenizer|ice|spiece).*\.model", fname) or is_tiktoken
                is_text = re.match(r".*\.(txt|json|py|md)", fname) or is_tokenizer
                if any((is_pytorch, is_safetensors, is_pt, is_gguf, is_tokenizer, is_text)):
                    if 'lfs' in dict[i]:
                        sha256.append([fname, dict[i]['lfs']['oid']])

                    if is_text:
                        links.append(f"https://huggingface.co/{model}/resolve/{branch}/{fname}")
                        classifications.append('text')
                        continue

                    if not text_only:
                        links.append(f"https://huggingface.co/{model}/resolve/{branch}/{fname}")
                        if is_safetensors:
                            has_safetensors = True
                            classifications.append('safetensors')
                        elif is_pytorch:
                            has_pytorch = True
                            classifications.append('pytorch')
                        elif is_pt:
                            has_pt = True
                            classifications.append('pt')
                        elif is_gguf:
                            has_gguf = True
                            classifications.append('gguf')

            cursor = base64.b64encode(f'{{"file_name":"{dict[-1]["path"]}"}}'.encode()) + b':50'
            cursor = base64.b64encode(cursor)
            cursor = cursor.replace(b'=', b'%3D')

        # If both pytorch and safetensors are available, download safetensors only
        if (has_pytorch or has_pt) and has_safetensors:
            for i in range(len(classifications) - 1, -1, -1):
                if classifications[i] in ['pytorch', 'pt']:
                    links.pop(i)

        # For GGUF, try to download only the Q4_K_M if no specific file is specified.
        # If not present, exclude all GGUFs, as that's likely a repository with both
        # GGUF and fp16 files.
        if has_gguf and specific_file is None:
            has_q4km = False
            for i in range(len(classifications) - 1, -1, -1):
                if 'q4_k_m' in links[i].lower():
                    has_q4km = True

            if has_q4km:
                for i in range(len(classifications) - 1, -1, -1):
                    if 'q4_k_m' not in links[i].lower():
                        links.pop(i)
            else:
                for i in range(len(classifications) - 1, -1, -1):
                    if links[i].lower().endswith('.gguf'):
                        links.pop(i)

        is_llamacpp = has_gguf and specific_file is not None
        return links, sha256, is_lora, is_llamacpp

    def get_output_folder(self, model, branch, is_lora, is_llamacpp=False, base_folder=None):
        if base_folder is None:
            base_folder = 'models' if not is_lora else 'loras'

        # If the model is of type GGUF, save directly in the base_folder
        if is_llamacpp:
            return Path(base_folder)

        output_folder = f"{'_'.join(model.split('/')[-2:])}"
        if branch != 'main':
            output_folder += f'_{branch}'

        output_folder = Path(base_folder) / output_folder
        return output_folder

    def get_single_file(self, url, output_folder, start_from_scratch=False):
        filename = Path(url.rsplit('/', 1)[1])
        output_path = output_folder / filename
        headers = {}
        mode = 'wb'
        if output_path.exists() and not start_from_scratch:

            # Check if the file has already been downloaded completely
            r = self.session.get(url, stream=True, timeout=10)
            total_size = int(r.headers.get('content-length', 0))
            if output_path.stat().st_size >= total_size:
                return

            # Otherwise, resume the download from where it left off
            headers = {'Range': f'bytes={output_path.stat().st_size}-'}
            mode = 'ab'

        with self.session.get(url, stream=True, headers=headers, timeout=10) as r:
            r.raise_for_status()  # Do not continue the download if the request was unsuccessful
            total_size = int(r.headers.get('content-length', 0))
            block_size = 1024 * 1024  # 1MB

            tqdm_kwargs = {
                'total': total_size,
                'unit': 'iB',
                'unit_scale': True,
                'bar_format': '{l_bar}{bar}| {n_fmt:6}/{total_fmt:6} {rate_fmt:6}'
            }

            if 'COLAB_GPU' in os.environ:
                tqdm_kwargs.update({
                    'position': 0,
                    'leave': True
                })

            with open(output_path, mode) as f:
                with tqdm.tqdm(**tqdm_kwargs) as t:
                    count = 0
                    for data in r.iter_content(block_size):
                        t.update(len(data))
                        f.write(data)
                        if total_size != 0 and self.progress_bar is not None:
                            count += len(data)
                            self.progress_bar(float(count) / float(total_size), f"{filename}")

    def start_download_threads(self, file_list, output_folder, start_from_scratch=False, threads=4):
        thread_map(lambda url: self.get_single_file(url, output_folder, start_from_scratch=start_from_scratch), file_list, max_workers=threads, disable=True)

    def download_model_files(self, model, branch, links, sha256, output_folder, progress_bar=None, start_from_scratch=False, threads=4, specific_file=None, is_llamacpp=False):
        self.progress_bar = progress_bar

        # Create the folder and writing the metadata
        output_folder.mkdir(parents=True, exist_ok=True)

        if not is_llamacpp:
            metadata = f'url: https://huggingface.co/{model}\n' \
                       f'branch: {branch}\n' \
                       f'download date: {datetime.datetime.now().strftime("%Y-%m-%d %H:%M:%S")}\n'

            sha256_str = '\n'.join([f'    {item[1]} {item[0]}' for item in sha256])
            if sha256_str:
                metadata += f'sha256sum:\n{sha256_str}'

            metadata += '\n'
            (output_folder / 'huggingface-metadata.txt').write_text(metadata)

        if specific_file:
            print(f"Downloading {specific_file} to {output_folder}")
        else:
            print(f"Downloading the model to {output_folder}")

        self.start_download_threads(links, output_folder, start_from_scratch=start_from_scratch, threads=threads)

    def check_model_files(self, model, branch, links, sha256, output_folder):
        # Validate the checksums
        validated = True
        for i in range(len(sha256)):
            fpath = (output_folder / sha256[i][0])

            if not fpath.exists():
                print(f"The following file is missing: {fpath}")
                validated = False
                continue

            with open(output_folder / sha256[i][0], "rb") as f:
                file_hash = hashlib.file_digest(f, "sha256").hexdigest()
                if file_hash != sha256[i][1]:
                    print(f'Checksum failed: {sha256[i][0]}  {sha256[i][1]}')
                    validated = False
                else:
                    print(f'Checksum validated: {sha256[i][0]}  {sha256[i][1]}')

        if validated:
            print('[+] Validated checksums of all model files!')
        else:
            print('[-] Invalid checksums. Rerun download-model.py with the --clean flag.')


if __name__ == '__main__':

    parser = argparse.ArgumentParser()
    parser.add_argument('MODEL', type=str, default=None, nargs='?')
    parser.add_argument('--branch', type=str, default='main', help='Name of the Git branch to download from.')
    parser.add_argument('--threads', type=int, default=4, help='Number of files to download simultaneously.')
    parser.add_argument('--text-only', action='store_true', help='Only download text files (txt/json).')
    parser.add_argument('--specific-file', type=str, default=None, help='Name of the specific file to download (if not provided, downloads all).')
    parser.add_argument('--output', type=str, default=None, help='The folder where the model should be saved.')
    parser.add_argument('--clean', action='store_true', help='Does not resume the previous download.')
    parser.add_argument('--check', action='store_true', help='Validates the checksums of model files.')
    parser.add_argument('--max-retries', type=int, default=5, help='Max retries count when get error in download time.')
    args = parser.parse_args()

    branch = args.branch
    model = args.MODEL
    specific_file = args.specific_file

    if model is None:
        print("Error: Please specify the model you'd like to download (e.g. 'python download-model.py facebook/opt-1.3b').")
        sys.exit()

    downloader = ModelDownloader(max_retries=args.max_retries)
    # Clean up the model/branch names
    try:
        model, branch = downloader.sanitize_model_and_branch_names(model, branch)
    except ValueError as err_branch:
        print(f"Error: {err_branch}")
        sys.exit()

    # Get the download links from Hugging Face
    links, sha256, is_lora, is_llamacpp = downloader.get_download_links_from_huggingface(model, branch, text_only=args.text_only, specific_file=specific_file)

    # Get the output folder
    output_folder = downloader.get_output_folder(model, branch, is_lora, is_llamacpp=is_llamacpp, base_folder=args.output)

    if args.check:
        # Check previously downloaded files
        downloader.check_model_files(model, branch, links, sha256, output_folder)
    else:
        # Download files
        downloader.download_model_files(model, branch, links, sha256, output_folder, specific_file=specific_file, threads=args.threads, is_llamacpp=is_llamacpp)