OsamaBinLikhon commited on
Commit
268bce3
·
verified ·
1 Parent(s): 5bb1b25

Initial upload of SheikhMiniMoE medical model (2026-01-18)

Browse files
Files changed (6) hide show
  1. README.md +82 -0
  2. config.json +9 -0
  3. generation_config.json +9 -0
  4. pytorch_model.bin +3 -0
  5. tokenizer.json +394 -0
  6. vocab.txt +0 -0
README.md ADDED
@@ -0,0 +1,82 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ - bn
5
+ tags:
6
+ - medical
7
+ - healthcare
8
+ - bilingual
9
+ - mixture-of-experts
10
+ - humanitarian
11
+ license: apache-2.0
12
+ datasets:
13
+ - medical_qa
14
+ ---
15
+
16
+ # SheikhMiniMoE-Medical
17
+
18
+ A production-ready Mixture of Experts (MoE) model optimized for bilingual medical Q&A in English and Bengali.
19
+
20
+ ## Model Details
21
+
22
+ - **Architecture**: SheikhMiniMoE - Sparse Mixture of Experts with top-k routing
23
+ - **Parameters**: N/A
24
+ - **Vocabulary Size**: 5000
25
+ - **Embedding Dimension**: 256
26
+ - **Hidden Dimension**: 512
27
+ - **Number of Experts**: 8
28
+ - **Top-K Routing**: 2
29
+
30
+ ## Intended Use
31
+
32
+ This model is designed for humanitarian medical assistance, providing:
33
+ - Bilingual medical Q&A (English/Bengali)
34
+ - Safety-first responses with appropriate disclaimers
35
+ - Emergency keyword detection
36
+ - Knowledge base augmentation
37
+
38
+ ## Training Configuration
39
+
40
+ - **Training Date**: 2026-01-18
41
+ - **Device**: CPU-optimized for deployment in resource-constrained environments
42
+
43
+ ## Usage
44
+
45
+ ```python
46
+ from transformers import AutoModelForCausalLM, AutoTokenizer
47
+ import torch
48
+
49
+ # Load model and tokenizer
50
+ model_name = "OsamaBinLikhon/SheikhMiniMoE-Medical"
51
+ tokenizer = AutoTokenizer.from_pretrained(model_name)
52
+ model = AutoModelForCausalLM.from_pretrained(model_name)
53
+
54
+ # Medical query
55
+ prompt = "What are symptoms of fever?"
56
+ inputs = tokenizer(prompt, return_tensors="pt")
57
+
58
+ # Generate response
59
+ outputs = model.generate(**inputs, max_new_tokens=100, temperature=0.7)
60
+ response = tokenizer.decode(outputs[0], skip_special_tokens=True)
61
+ print(response)
62
+ ```
63
+
64
+ ## Safety Notice
65
+
66
+ ⚠️ This model provides medical information for educational purposes only.
67
+ It is not a substitute for professional medical advice, diagnosis, or treatment.
68
+ Always consult qualified healthcare providers for medical concerns.
69
+
70
+ ## License
71
+
72
+ Apache 2.0 - See LICENSE file for details.
73
+
74
+ ## Citation
75
+
76
+ ```bibtex
77
+ @misc{SheikhMiniMoE,
78
+ title={SheikhMiniMoE: Bilingual Medical Assistant},
79
+ author={MiniMax Agent},
80
+ year={2025},
81
+ url={https://huggingface.co/OsamaBinLikhon/SheikhMiniMoE-Medical}
82
+ }
config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 5000,
3
+ "embedding_dim": 256,
4
+ "hidden_dim": 512,
5
+ "num_layers": 4,
6
+ "num_experts": 8,
7
+ "top_k": 2,
8
+ "dropout": 0.1
9
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "eos_token_id": 3,
4
+ "pad_token_id": 0,
5
+ "max_new_tokens": 100,
6
+ "temperature": 0.7,
7
+ "top_k": 40,
8
+ "do_sample": true
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2a491bd3d1ec8c16ac21127183339c450fa75ec436102c6ec27d9554699f875
3
+ size 33803400
tokenizer.json ADDED
@@ -0,0 +1,394 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "vocab_size": 5000,
3
+ "special_tokens": {
4
+ "<PAD>": 0,
5
+ "<UNK>": 1,
6
+ "<BOS>": 2,
7
+ "<EOS>": 3
8
+ },
9
+ "char_vocab": {
10
+ "\u0000": 4,
11
+ "\u0001": 5,
12
+ "\u0002": 6,
13
+ "\u0003": 7,
14
+ "\u0004": 8,
15
+ "\u0005": 9,
16
+ "\u0006": 10,
17
+ "\u0007": 11,
18
+ "\b": 12,
19
+ "\t": 13,
20
+ "\n": 14,
21
+ "\u000b": 15,
22
+ "\f": 16,
23
+ "\r": 17,
24
+ "\u000e": 18,
25
+ "\u000f": 19,
26
+ "\u0010": 20,
27
+ "\u0011": 21,
28
+ "\u0012": 22,
29
+ "\u0013": 23,
30
+ "\u0014": 24,
31
+ "\u0015": 25,
32
+ "\u0016": 26,
33
+ "\u0017": 27,
34
+ "\u0018": 28,
35
+ "\u0019": 29,
36
+ "\u001a": 30,
37
+ "\u001b": 31,
38
+ "\u001c": 32,
39
+ "\u001d": 33,
40
+ "\u001e": 34,
41
+ "\u001f": 35,
42
+ " ": 36,
43
+ "!": 37,
44
+ "\"": 38,
45
+ "#": 39,
46
+ "$": 40,
47
+ "%": 41,
48
+ "&": 42,
49
+ "'": 43,
50
+ "(": 44,
51
+ ")": 45,
52
+ "*": 46,
53
+ "+": 47,
54
+ ",": 48,
55
+ "-": 49,
56
+ ".": 50,
57
+ "/": 51,
58
+ "0": 52,
59
+ "1": 53,
60
+ "2": 54,
61
+ "3": 55,
62
+ "4": 56,
63
+ "5": 57,
64
+ "6": 58,
65
+ "7": 59,
66
+ "8": 60,
67
+ "9": 61,
68
+ ":": 62,
69
+ ";": 63,
70
+ "<": 64,
71
+ "=": 65,
72
+ ">": 66,
73
+ "?": 67,
74
+ "@": 68,
75
+ "A": 69,
76
+ "B": 70,
77
+ "C": 71,
78
+ "D": 72,
79
+ "E": 73,
80
+ "F": 74,
81
+ "G": 75,
82
+ "H": 76,
83
+ "I": 77,
84
+ "J": 78,
85
+ "K": 79,
86
+ "L": 80,
87
+ "M": 81,
88
+ "N": 82,
89
+ "O": 83,
90
+ "P": 84,
91
+ "Q": 85,
92
+ "R": 86,
93
+ "S": 87,
94
+ "T": 88,
95
+ "U": 89,
96
+ "V": 90,
97
+ "W": 91,
98
+ "X": 92,
99
+ "Y": 93,
100
+ "Z": 94,
101
+ "[": 95,
102
+ "\\": 96,
103
+ "]": 97,
104
+ "^": 98,
105
+ "_": 99,
106
+ "`": 100,
107
+ "a": 101,
108
+ "b": 102,
109
+ "c": 103,
110
+ "d": 104,
111
+ "e": 105,
112
+ "f": 106,
113
+ "g": 107,
114
+ "h": 108,
115
+ "i": 109,
116
+ "j": 110,
117
+ "k": 111,
118
+ "l": 112,
119
+ "m": 113,
120
+ "n": 114,
121
+ "o": 115,
122
+ "p": 116,
123
+ "q": 117,
124
+ "r": 118,
125
+ "s": 119,
126
+ "t": 120,
127
+ "u": 121,
128
+ "v": 122,
129
+ "w": 123,
130
+ "x": 124,
131
+ "y": 125,
132
+ "z": 126,
133
+ "{": 127,
134
+ "|": 128,
135
+ "}": 129,
136
+ "~": 130,
137
+ "\u007f": 131,
138
+ "\u0080": 132,
139
+ "\u0081": 133,
140
+ "\u0082": 134,
141
+ "\u0083": 135,
142
+ "\u0084": 136,
143
+ "\u0085": 137,
144
+ "\u0086": 138,
145
+ "\u0087": 139,
146
+ "\u0088": 140,
147
+ "\u0089": 141,
148
+ "\u008a": 142,
149
+ "\u008b": 143,
150
+ "\u008c": 144,
151
+ "\u008d": 145,
152
+ "\u008e": 146,
153
+ "\u008f": 147,
154
+ "\u0090": 148,
155
+ "\u0091": 149,
156
+ "\u0092": 150,
157
+ "\u0093": 151,
158
+ "\u0094": 152,
159
+ "\u0095": 153,
160
+ "\u0096": 154,
161
+ "\u0097": 155,
162
+ "\u0098": 156,
163
+ "\u0099": 157,
164
+ "\u009a": 158,
165
+ "\u009b": 159,
166
+ "\u009c": 160,
167
+ "\u009d": 161,
168
+ "\u009e": 162,
169
+ "\u009f": 163,
170
+ "\u00a0": 164,
171
+ "\u00a1": 165,
172
+ "\u00a2": 166,
173
+ "\u00a3": 167,
174
+ "\u00a4": 168,
175
+ "\u00a5": 169,
176
+ "\u00a6": 170,
177
+ "\u00a7": 171,
178
+ "\u00a8": 172,
179
+ "\u00a9": 173,
180
+ "\u00aa": 174,
181
+ "\u00ab": 175,
182
+ "\u00ac": 176,
183
+ "\u00ad": 177,
184
+ "\u00ae": 178,
185
+ "\u00af": 179,
186
+ "\u00b0": 180,
187
+ "\u00b1": 181,
188
+ "\u00b2": 182,
189
+ "\u00b3": 183,
190
+ "\u00b4": 184,
191
+ "\u00b5": 185,
192
+ "\u00b6": 186,
193
+ "\u00b7": 187,
194
+ "\u00b8": 188,
195
+ "\u00b9": 189,
196
+ "\u00ba": 190,
197
+ "\u00bb": 191,
198
+ "\u00bc": 192,
199
+ "\u00bd": 193,
200
+ "\u00be": 194,
201
+ "\u00bf": 195,
202
+ "\u00c0": 196,
203
+ "\u00c1": 197,
204
+ "\u00c2": 198,
205
+ "\u00c3": 199,
206
+ "\u00c4": 200,
207
+ "\u00c5": 201,
208
+ "\u00c6": 202,
209
+ "\u00c7": 203,
210
+ "\u00c8": 204,
211
+ "\u00c9": 205,
212
+ "\u00ca": 206,
213
+ "\u00cb": 207,
214
+ "\u00cc": 208,
215
+ "\u00cd": 209,
216
+ "\u00ce": 210,
217
+ "\u00cf": 211,
218
+ "\u00d0": 212,
219
+ "\u00d1": 213,
220
+ "\u00d2": 214,
221
+ "\u00d3": 215,
222
+ "\u00d4": 216,
223
+ "\u00d5": 217,
224
+ "\u00d6": 218,
225
+ "\u00d7": 219,
226
+ "\u00d8": 220,
227
+ "\u00d9": 221,
228
+ "\u00da": 222,
229
+ "\u00db": 223,
230
+ "\u00dc": 224,
231
+ "\u00dd": 225,
232
+ "\u00de": 226,
233
+ "\u00df": 227,
234
+ "\u00e0": 228,
235
+ "\u00e1": 229,
236
+ "\u00e2": 230,
237
+ "\u00e3": 231,
238
+ "\u00e4": 232,
239
+ "\u00e5": 233,
240
+ "\u00e6": 234,
241
+ "\u00e7": 235,
242
+ "\u00e8": 236,
243
+ "\u00e9": 237,
244
+ "\u00ea": 238,
245
+ "\u00eb": 239,
246
+ "\u00ec": 240,
247
+ "\u00ed": 241,
248
+ "\u00ee": 242,
249
+ "\u00ef": 243,
250
+ "\u00f0": 244,
251
+ "\u00f1": 245,
252
+ "\u00f2": 246,
253
+ "\u00f3": 247,
254
+ "\u00f4": 248,
255
+ "\u00f5": 249,
256
+ "\u00f6": 250,
257
+ "\u00f7": 251,
258
+ "\u00f8": 252,
259
+ "\u00f9": 253,
260
+ "\u00fa": 254,
261
+ "\u00fb": 255,
262
+ "\u00fc": 256,
263
+ "\u00fd": 257,
264
+ "\u00fe": 258,
265
+ "\u00ff": 259,
266
+ "\u0980": 256,
267
+ "\u0981": 257,
268
+ "\u0982": 258,
269
+ "\u0983": 259,
270
+ "\u0984": 260,
271
+ "\u0985": 261,
272
+ "\u0986": 262,
273
+ "\u0987": 263,
274
+ "\u0988": 264,
275
+ "\u0989": 265,
276
+ "\u098a": 266,
277
+ "\u098b": 267,
278
+ "\u098c": 268,
279
+ "\u098d": 269,
280
+ "\u098e": 270,
281
+ "\u098f": 271,
282
+ "\u0990": 272,
283
+ "\u0991": 273,
284
+ "\u0992": 274,
285
+ "\u0993": 275,
286
+ "\u0994": 276,
287
+ "\u0995": 277,
288
+ "\u0996": 278,
289
+ "\u0997": 279,
290
+ "\u0998": 280,
291
+ "\u0999": 281,
292
+ "\u099a": 282,
293
+ "\u099b": 283,
294
+ "\u099c": 284,
295
+ "\u099d": 285,
296
+ "\u099e": 286,
297
+ "\u099f": 287,
298
+ "\u09a0": 288,
299
+ "\u09a1": 289,
300
+ "\u09a2": 290,
301
+ "\u09a3": 291,
302
+ "\u09a4": 292,
303
+ "\u09a5": 293,
304
+ "\u09a6": 294,
305
+ "\u09a7": 295,
306
+ "\u09a8": 296,
307
+ "\u09a9": 297,
308
+ "\u09aa": 298,
309
+ "\u09ab": 299,
310
+ "\u09ac": 300,
311
+ "\u09ad": 301,
312
+ "\u09ae": 302,
313
+ "\u09af": 303,
314
+ "\u09b0": 304,
315
+ "\u09b1": 305,
316
+ "\u09b2": 306,
317
+ "\u09b3": 307,
318
+ "\u09b4": 308,
319
+ "\u09b5": 309,
320
+ "\u09b6": 310,
321
+ "\u09b7": 311,
322
+ "\u09b8": 312,
323
+ "\u09b9": 313,
324
+ "\u09ba": 314,
325
+ "\u09bb": 315,
326
+ "\u09bc": 316,
327
+ "\u09bd": 317,
328
+ "\u09be": 318,
329
+ "\u09bf": 319,
330
+ "\u09c0": 320,
331
+ "\u09c1": 321,
332
+ "\u09c2": 322,
333
+ "\u09c3": 323,
334
+ "\u09c4": 324,
335
+ "\u09c5": 325,
336
+ "\u09c6": 326,
337
+ "\u09c7": 327,
338
+ "\u09c8": 328,
339
+ "\u09c9": 329,
340
+ "\u09ca": 330,
341
+ "\u09cb": 331,
342
+ "\u09cc": 332,
343
+ "\u09cd": 333,
344
+ "\u09ce": 334,
345
+ "\u09cf": 335,
346
+ "\u09d0": 336,
347
+ "\u09d1": 337,
348
+ "\u09d2": 338,
349
+ "\u09d3": 339,
350
+ "\u09d4": 340,
351
+ "\u09d5": 341,
352
+ "\u09d6": 342,
353
+ "\u09d7": 343,
354
+ "\u09d8": 344,
355
+ "\u09d9": 345,
356
+ "\u09da": 346,
357
+ "\u09db": 347,
358
+ "\u09dc": 348,
359
+ "\u09dd": 349,
360
+ "\u09de": 350,
361
+ "\u09df": 351,
362
+ "\u09e0": 352,
363
+ "\u09e1": 353,
364
+ "\u09e2": 354,
365
+ "\u09e3": 355,
366
+ "\u09e4": 356,
367
+ "\u09e5": 357,
368
+ "\u09e6": 358,
369
+ "\u09e7": 359,
370
+ "\u09e8": 360,
371
+ "\u09e9": 361,
372
+ "\u09ea": 362,
373
+ "\u09eb": 363,
374
+ "\u09ec": 364,
375
+ "\u09ed": 365,
376
+ "\u09ee": 366,
377
+ "\u09ef": 367,
378
+ "\u09f0": 368,
379
+ "\u09f1": 369,
380
+ "\u09f2": 370,
381
+ "\u09f3": 371,
382
+ "\u09f4": 372,
383
+ "\u09f5": 373,
384
+ "\u09f6": 374,
385
+ "\u09f7": 375,
386
+ "\u09f8": 376,
387
+ "\u09f9": 377,
388
+ "\u09fa": 378,
389
+ "\u09fb": 379,
390
+ "\u09fc": 380,
391
+ "\u09fd": 381,
392
+ "\u09fe": 382
393
+ }
394
+ }
vocab.txt ADDED
Binary file (2.58 kB). View file