Spaces:
Running
Running
v2 implemented
Browse files- .DS_Store +0 -0
- app.py +29 -14
- module_3_3.ipynb +366 -0
- v1/u_model.pth +0 -0
- {v1 β v2}/__init__.py +0 -0
- {v1 β v2}/tokenizer.json +0 -0
- v2/u_model_4000.pth +0 -0
- {v1 β v2}/usta_causal_attention.py +0 -0
- {v1 β v2}/usta_decoder_block.py +12 -6
- {v1 β v2}/usta_embedding.py +10 -9
- {v1 β v2}/usta_layer_norm.py +3 -4
- {v1 β v2}/usta_mlp.py +5 -5
- {v1 β v2}/usta_model.py +45 -8
- {v1 β v2}/usta_multi_head_attention.py +4 -4
- {v1 β v2}/usta_multi_head_attention_old.py +1 -2
- {v1 β v2}/usta_self_attention.py +0 -0
- {v1 β v2}/usta_tokenizer.py +16 -1
.DS_Store
ADDED
Binary file (6.15 kB). View file
|
|
app.py
CHANGED
@@ -3,14 +3,14 @@ import os
|
|
3 |
import gradio as gr
|
4 |
import torch
|
5 |
|
6 |
-
from
|
7 |
-
from
|
8 |
|
9 |
|
10 |
# Load the model and tokenizer
|
11 |
def load_model(custom_model_path=None):
|
12 |
try:
|
13 |
-
u_tokenizer = UstaTokenizer("
|
14 |
print("β
Tokenizer loaded successfully! vocab size:", len(u_tokenizer.vocab))
|
15 |
|
16 |
# Model parameters - adjust these to match your trained model
|
@@ -19,6 +19,7 @@ def load_model(custom_model_path=None):
|
|
19 |
embedding_dim = 12
|
20 |
num_heads = 4
|
21 |
num_layers = 8
|
|
|
22 |
|
23 |
# Load the model
|
24 |
u_model = UstaModel(
|
@@ -26,7 +27,8 @@ def load_model(custom_model_path=None):
|
|
26 |
embedding_dim=embedding_dim,
|
27 |
num_heads=num_heads,
|
28 |
context_length=context_length,
|
29 |
-
num_layers=num_layers
|
|
|
30 |
)
|
31 |
|
32 |
# Determine which model file to use
|
@@ -34,7 +36,7 @@ def load_model(custom_model_path=None):
|
|
34 |
model_path = custom_model_path
|
35 |
print(f"π― Using uploaded model: {model_path}")
|
36 |
else:
|
37 |
-
model_path = "
|
38 |
|
39 |
if not os.path.exists(model_path):
|
40 |
print("β Model file not found at", model_path)
|
@@ -58,8 +60,8 @@ def load_model(custom_model_path=None):
|
|
58 |
|
59 |
print(f"π¦ Downloaded {len(response.content)} bytes")
|
60 |
|
61 |
-
# Create
|
62 |
-
os.makedirs("
|
63 |
|
64 |
# Save the model weights to the local file system
|
65 |
with open(model_path, "wb") as f:
|
@@ -195,7 +197,7 @@ def load_model_from_file(uploaded_file):
|
|
195 |
model_status = error_msg
|
196 |
return error_msg
|
197 |
|
198 |
-
def chat_with_usta(message, history, max_tokens=20):
|
199 |
"""Simple chat function"""
|
200 |
if model is None or tokenizer is None:
|
201 |
return history + [["Error", "UstaModel is not available. Please try again later."]]
|
@@ -211,7 +213,13 @@ def chat_with_usta(message, history, max_tokens=20):
|
|
211 |
# Generate response
|
212 |
with torch.no_grad():
|
213 |
actual_max_tokens = min(max_tokens, 32 - len(tokens))
|
214 |
-
generated_tokens = model.generate(
|
|
|
|
|
|
|
|
|
|
|
|
|
215 |
|
216 |
# Decode the generated tokens
|
217 |
response = tokenizer.decode(generated_tokens)
|
@@ -249,7 +257,14 @@ with gr.Blocks(title="π€ Usta Model Chat") as demo:
|
|
249 |
clear_btn = gr.Button("Clear")
|
250 |
|
251 |
# Generation settings
|
252 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
253 |
|
254 |
# Model loading (simplified)
|
255 |
gr.Markdown("## π§ Load Custom Model (Optional)")
|
@@ -268,20 +283,20 @@ with gr.Blocks(title="π€ Usta Model Chat") as demo:
|
|
268 |
status = gr.Textbox(label="Status", value=model_status, interactive=False)
|
269 |
|
270 |
# Event handlers
|
271 |
-
def send_message(message, history, max_tok):
|
272 |
if not message.strip():
|
273 |
return history, ""
|
274 |
-
return chat_with_usta(message, history, max_tok), ""
|
275 |
|
276 |
send_btn.click(
|
277 |
send_message,
|
278 |
-
inputs=[msg, chatbot, max_tokens],
|
279 |
outputs=[chatbot, msg]
|
280 |
)
|
281 |
|
282 |
msg.submit(
|
283 |
send_message,
|
284 |
-
inputs=[msg, chatbot, max_tokens],
|
285 |
outputs=[chatbot, msg]
|
286 |
)
|
287 |
|
|
|
3 |
import gradio as gr
|
4 |
import torch
|
5 |
|
6 |
+
from v2.usta_model import UstaModel
|
7 |
+
from v2.usta_tokenizer import UstaTokenizer
|
8 |
|
9 |
|
10 |
# Load the model and tokenizer
|
11 |
def load_model(custom_model_path=None):
|
12 |
try:
|
13 |
+
u_tokenizer = UstaTokenizer("v2/tokenizer.json")
|
14 |
print("β
Tokenizer loaded successfully! vocab size:", len(u_tokenizer.vocab))
|
15 |
|
16 |
# Model parameters - adjust these to match your trained model
|
|
|
19 |
embedding_dim = 12
|
20 |
num_heads = 4
|
21 |
num_layers = 8
|
22 |
+
device = "cpu" # Use CPU for compatibility
|
23 |
|
24 |
# Load the model
|
25 |
u_model = UstaModel(
|
|
|
27 |
embedding_dim=embedding_dim,
|
28 |
num_heads=num_heads,
|
29 |
context_length=context_length,
|
30 |
+
num_layers=num_layers,
|
31 |
+
device=device
|
32 |
)
|
33 |
|
34 |
# Determine which model file to use
|
|
|
36 |
model_path = custom_model_path
|
37 |
print(f"π― Using uploaded model: {model_path}")
|
38 |
else:
|
39 |
+
model_path = "v2/u_model_4000.pth"
|
40 |
|
41 |
if not os.path.exists(model_path):
|
42 |
print("β Model file not found at", model_path)
|
|
|
60 |
|
61 |
print(f"π¦ Downloaded {len(response.content)} bytes")
|
62 |
|
63 |
+
# Create v2 directory if it doesn't exist
|
64 |
+
os.makedirs("v2", exist_ok=True)
|
65 |
|
66 |
# Save the model weights to the local file system
|
67 |
with open(model_path, "wb") as f:
|
|
|
197 |
model_status = error_msg
|
198 |
return error_msg
|
199 |
|
200 |
+
def chat_with_usta(message, history, max_tokens=20, temperature=1.0, top_k=64, top_p=1.0):
|
201 |
"""Simple chat function"""
|
202 |
if model is None or tokenizer is None:
|
203 |
return history + [["Error", "UstaModel is not available. Please try again later."]]
|
|
|
213 |
# Generate response
|
214 |
with torch.no_grad():
|
215 |
actual_max_tokens = min(max_tokens, 32 - len(tokens))
|
216 |
+
generated_tokens = model.generate(
|
217 |
+
tokens,
|
218 |
+
max_new_tokens=actual_max_tokens,
|
219 |
+
temperature=temperature,
|
220 |
+
top_k=top_k,
|
221 |
+
top_p=top_p
|
222 |
+
)
|
223 |
|
224 |
# Decode the generated tokens
|
225 |
response = tokenizer.decode(generated_tokens)
|
|
|
257 |
clear_btn = gr.Button("Clear")
|
258 |
|
259 |
# Generation settings
|
260 |
+
gr.Markdown("## βοΈ Generation Settings")
|
261 |
+
with gr.Row():
|
262 |
+
max_tokens = gr.Slider(minimum=1, maximum=30, value=20, step=1, label="Max tokens")
|
263 |
+
temperature = gr.Slider(minimum=0.1, maximum=2.0, value=1.0, step=0.1, label="Temperature")
|
264 |
+
|
265 |
+
with gr.Row():
|
266 |
+
top_k = gr.Slider(minimum=1, maximum=64, value=40, step=1, label="Top-k")
|
267 |
+
top_p = gr.Slider(minimum=0.1, maximum=1.0, value=1.0, step=0.05, label="Top-p (nucleus sampling)")
|
268 |
|
269 |
# Model loading (simplified)
|
270 |
gr.Markdown("## π§ Load Custom Model (Optional)")
|
|
|
283 |
status = gr.Textbox(label="Status", value=model_status, interactive=False)
|
284 |
|
285 |
# Event handlers
|
286 |
+
def send_message(message, history, max_tok, temp, k, p):
|
287 |
if not message.strip():
|
288 |
return history, ""
|
289 |
+
return chat_with_usta(message, history, max_tok, temp, k, p), ""
|
290 |
|
291 |
send_btn.click(
|
292 |
send_message,
|
293 |
+
inputs=[msg, chatbot, max_tokens, temperature, top_k, top_p],
|
294 |
outputs=[chatbot, msg]
|
295 |
)
|
296 |
|
297 |
msg.submit(
|
298 |
send_message,
|
299 |
+
inputs=[msg, chatbot, max_tokens, temperature, top_k, top_p],
|
300 |
outputs=[chatbot, msg]
|
301 |
)
|
302 |
|
module_3_3.ipynb
ADDED
@@ -0,0 +1,366 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cells": [
|
3 |
+
{
|
4 |
+
"cell_type": "code",
|
5 |
+
"execution_count": 1,
|
6 |
+
"metadata": {},
|
7 |
+
"outputs": [
|
8 |
+
{
|
9 |
+
"name": "stdout",
|
10 |
+
"output_type": "stream",
|
11 |
+
"text": [
|
12 |
+
"Using device: mps\n",
|
13 |
+
"tensor([ 0, 61, 1, 61, 2, 61, 0, 61, 3], device='mps:0')\n"
|
14 |
+
]
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"data": {
|
18 |
+
"text/plain": [
|
19 |
+
"torch.Size([4, 32])"
|
20 |
+
]
|
21 |
+
},
|
22 |
+
"execution_count": 1,
|
23 |
+
"metadata": {},
|
24 |
+
"output_type": "execute_result"
|
25 |
+
}
|
26 |
+
],
|
27 |
+
"source": [
|
28 |
+
"import torch\n",
|
29 |
+
"\n",
|
30 |
+
"from usta_model import UstaModel\n",
|
31 |
+
"from usta_tokenizer import UstaTokenizer\n",
|
32 |
+
"\n",
|
33 |
+
"device = \"cpu\"\n",
|
34 |
+
"\n",
|
35 |
+
"if torch.cuda.is_available():\n",
|
36 |
+
" device = \"cuda\"\n",
|
37 |
+
"elif torch.backends.mps.is_available():\n",
|
38 |
+
" device = \"mps\"\n",
|
39 |
+
" \n",
|
40 |
+
"\n",
|
41 |
+
"print(f\"Using device: {device}\")\n",
|
42 |
+
"\n",
|
43 |
+
"u_tokenizer = UstaTokenizer(\"tokenizer.json\")\n",
|
44 |
+
"\n",
|
45 |
+
"prompts = [\n",
|
46 |
+
" \"the capital of the united\",\n",
|
47 |
+
" \"madrid is in\",\n",
|
48 |
+
" \"the capital of france is\",\n",
|
49 |
+
" \"the capital of germany is\"\n",
|
50 |
+
"]\n",
|
51 |
+
"\n",
|
52 |
+
"tokens = u_tokenizer.encode(prompts[0])\n",
|
53 |
+
"tokens = tokens.to(device)\n",
|
54 |
+
"print(tokens)\n",
|
55 |
+
"batch_tokens = u_tokenizer.encode_batch(prompts, 32)\n",
|
56 |
+
"batch_tokens = batch_tokens.to(device)\n",
|
57 |
+
"batch_tokens.shape"
|
58 |
+
]
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"cell_type": "code",
|
62 |
+
"execution_count": 2,
|
63 |
+
"metadata": {},
|
64 |
+
"outputs": [
|
65 |
+
{
|
66 |
+
"data": {
|
67 |
+
"text/plain": [
|
68 |
+
"<All keys matched successfully>"
|
69 |
+
]
|
70 |
+
},
|
71 |
+
"execution_count": 2,
|
72 |
+
"metadata": {},
|
73 |
+
"output_type": "execute_result"
|
74 |
+
}
|
75 |
+
],
|
76 |
+
"source": [
|
77 |
+
"torch.manual_seed(1)\n",
|
78 |
+
"context_length = 32\n",
|
79 |
+
"\n",
|
80 |
+
"u_model = UstaModel(\n",
|
81 |
+
" vocab_size=len(u_tokenizer.vocab),\n",
|
82 |
+
" embedding_dim=12,\n",
|
83 |
+
" num_heads=4,\n",
|
84 |
+
" context_length=context_length,\n",
|
85 |
+
" num_layers=8,\n",
|
86 |
+
" device=device\n",
|
87 |
+
")\n",
|
88 |
+
"\n",
|
89 |
+
"# load model\n",
|
90 |
+
"u_model.load_state_dict(torch.load(\"../u_model_4000.pth\"))"
|
91 |
+
]
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"cell_type": "code",
|
95 |
+
"execution_count": 3,
|
96 |
+
"metadata": {},
|
97 |
+
"outputs": [
|
98 |
+
{
|
99 |
+
"data": {
|
100 |
+
"text/plain": [
|
101 |
+
"torch.Size([4, 32, 64])"
|
102 |
+
]
|
103 |
+
},
|
104 |
+
"execution_count": 3,
|
105 |
+
"metadata": {},
|
106 |
+
"output_type": "execute_result"
|
107 |
+
}
|
108 |
+
],
|
109 |
+
"source": [
|
110 |
+
"out = u_model(batch_tokens)\n",
|
111 |
+
"out.shape"
|
112 |
+
]
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"cell_type": "code",
|
116 |
+
"execution_count": 4,
|
117 |
+
"metadata": {},
|
118 |
+
"outputs": [],
|
119 |
+
"source": [
|
120 |
+
"# temperature\n",
|
121 |
+
"# top_k \n",
|
122 |
+
"# top_p\n"
|
123 |
+
]
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"cell_type": "code",
|
127 |
+
"execution_count": 5,
|
128 |
+
"metadata": {},
|
129 |
+
"outputs": [],
|
130 |
+
"source": [
|
131 |
+
"top_k = 10"
|
132 |
+
]
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"cell_type": "code",
|
136 |
+
"execution_count": 6,
|
137 |
+
"metadata": {},
|
138 |
+
"outputs": [
|
139 |
+
{
|
140 |
+
"data": {
|
141 |
+
"text/plain": [
|
142 |
+
"(tensor([17.6884, 14.0799, 9.0104, 8.4548, 7.3207, 7.2960, 6.8096, 6.6073,\n",
|
143 |
+
" 6.6009, 6.3761]),\n",
|
144 |
+
" [61, 60, 35, 58, 9, 38, 59, 4, 18, 49])"
|
145 |
+
]
|
146 |
+
},
|
147 |
+
"execution_count": 6,
|
148 |
+
"metadata": {},
|
149 |
+
"output_type": "execute_result"
|
150 |
+
}
|
151 |
+
],
|
152 |
+
"source": [
|
153 |
+
"sorted_outs = sorted(out[-1][-1].tolist(), reverse=True)\n",
|
154 |
+
"sorted_indexes = []\n",
|
155 |
+
"for so in sorted_outs[:top_k]:\n",
|
156 |
+
" so_index = out[-1][-1].tolist().index(so)\n",
|
157 |
+
" sorted_indexes.append(so_index)\n",
|
158 |
+
"sorted_outs = torch.tensor(sorted_outs[:top_k])\n",
|
159 |
+
"sorted_outs, sorted_indexes\n"
|
160 |
+
]
|
161 |
+
},
|
162 |
+
{
|
163 |
+
"cell_type": "code",
|
164 |
+
"execution_count": 7,
|
165 |
+
"metadata": {},
|
166 |
+
"outputs": [
|
167 |
+
{
|
168 |
+
"data": {
|
169 |
+
"text/plain": [
|
170 |
+
"(tensor([17.6884, 14.0799, 9.0104, 8.4548, 7.3207, 7.2960, 6.8096, 6.6073,\n",
|
171 |
+
" 6.6009, 6.3761], device='mps:0', grad_fn=<TopkBackward0>),\n",
|
172 |
+
" tensor([61, 60, 35, 58, 9, 38, 59, 4, 18, 49], device='mps:0'))"
|
173 |
+
]
|
174 |
+
},
|
175 |
+
"execution_count": 7,
|
176 |
+
"metadata": {},
|
177 |
+
"output_type": "execute_result"
|
178 |
+
}
|
179 |
+
],
|
180 |
+
"source": [
|
181 |
+
"values, indexes = torch.topk(out[-1][-1], k=10)\n",
|
182 |
+
"values, indexes"
|
183 |
+
]
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"cell_type": "code",
|
187 |
+
"execution_count": null,
|
188 |
+
"metadata": {},
|
189 |
+
"outputs": [],
|
190 |
+
"source": []
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"cell_type": "code",
|
194 |
+
"execution_count": 8,
|
195 |
+
"metadata": {},
|
196 |
+
"outputs": [
|
197 |
+
{
|
198 |
+
"name": "stderr",
|
199 |
+
"output_type": "stream",
|
200 |
+
"text": [
|
201 |
+
"/var/folders/z7/wrd0w0hn7pvb9g97kmdn17640000gn/T/ipykernel_91075/2885985782.py:2: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.detach().clone() or sourceTensor.detach().clone().requires_grad_(True), rather than torch.tensor(sourceTensor).\n",
|
202 |
+
" adjusted_outs = torch.tensor(sorted_outs) / temperature\n"
|
203 |
+
]
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"data": {
|
207 |
+
"text/plain": [
|
208 |
+
"tensor([1.6830, 1.3397, 0.8573, 0.8045, 0.6965, 0.6942, 0.6479, 0.6287, 0.6281,\n",
|
209 |
+
" 0.6067])"
|
210 |
+
]
|
211 |
+
},
|
212 |
+
"execution_count": 8,
|
213 |
+
"metadata": {},
|
214 |
+
"output_type": "execute_result"
|
215 |
+
}
|
216 |
+
],
|
217 |
+
"source": [
|
218 |
+
"temperature = 10.51\n",
|
219 |
+
"adjusted_outs = torch.tensor(sorted_outs) / temperature\n",
|
220 |
+
"adjusted_outs"
|
221 |
+
]
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"cell_type": "code",
|
225 |
+
"execution_count": 9,
|
226 |
+
"metadata": {},
|
227 |
+
"outputs": [
|
228 |
+
{
|
229 |
+
"data": {
|
230 |
+
"text/plain": [
|
231 |
+
"tensor([0.2128, 0.1509, 0.0932, 0.0884, 0.0793, 0.0791, 0.0756, 0.0741, 0.0741,\n",
|
232 |
+
" 0.0725])"
|
233 |
+
]
|
234 |
+
},
|
235 |
+
"execution_count": 9,
|
236 |
+
"metadata": {},
|
237 |
+
"output_type": "execute_result"
|
238 |
+
}
|
239 |
+
],
|
240 |
+
"source": [
|
241 |
+
"probs = torch.softmax(adjusted_outs, dim=-1)\n",
|
242 |
+
"probs"
|
243 |
+
]
|
244 |
+
},
|
245 |
+
{
|
246 |
+
"cell_type": "code",
|
247 |
+
"execution_count": 10,
|
248 |
+
"metadata": {},
|
249 |
+
"outputs": [],
|
250 |
+
"source": [
|
251 |
+
"top_p = 0.7"
|
252 |
+
]
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"cell_type": "code",
|
256 |
+
"execution_count": 11,
|
257 |
+
"metadata": {},
|
258 |
+
"outputs": [
|
259 |
+
{
|
260 |
+
"data": {
|
261 |
+
"text/plain": [
|
262 |
+
"tensor(0.5453)"
|
263 |
+
]
|
264 |
+
},
|
265 |
+
"execution_count": 11,
|
266 |
+
"metadata": {},
|
267 |
+
"output_type": "execute_result"
|
268 |
+
}
|
269 |
+
],
|
270 |
+
"source": [
|
271 |
+
"[0.2128, 0.36, 0.37, 0.38, 0.70, 0.71]\n",
|
272 |
+
"torch.sum(torch.tensor([0.2128, 0.1509, 0.0932, 0.0884]))"
|
273 |
+
]
|
274 |
+
},
|
275 |
+
{
|
276 |
+
"cell_type": "code",
|
277 |
+
"execution_count": 12,
|
278 |
+
"metadata": {},
|
279 |
+
"outputs": [
|
280 |
+
{
|
281 |
+
"data": {
|
282 |
+
"text/plain": [
|
283 |
+
"{0: 212, 4: 82, 5: 87, 9: 83, 2: 74, 6: 73, 1: 154, 3: 91, 8: 80, 7: 64}"
|
284 |
+
]
|
285 |
+
},
|
286 |
+
"execution_count": 12,
|
287 |
+
"metadata": {},
|
288 |
+
"output_type": "execute_result"
|
289 |
+
}
|
290 |
+
],
|
291 |
+
"source": [
|
292 |
+
"sample_count = {}\n",
|
293 |
+
"for _ in range(1000):\n",
|
294 |
+
" sample = torch.multinomial(probs, 1)\n",
|
295 |
+
" sample_count[sample.item()] = sample_count.get(sample.item(), 0) + 1\n",
|
296 |
+
"sample_count"
|
297 |
+
]
|
298 |
+
},
|
299 |
+
{
|
300 |
+
"cell_type": "code",
|
301 |
+
"execution_count": 14,
|
302 |
+
"metadata": {},
|
303 |
+
"outputs": [
|
304 |
+
{
|
305 |
+
"data": {
|
306 |
+
"text/plain": [
|
307 |
+
"{'the capital of the united.': 3,\n",
|
308 |
+
" 'the capital of the united the ': 22,\n",
|
309 |
+
" 'the capital of the united identity,': 1,\n",
|
310 |
+
" 'the capital of the united capitals': 5,\n",
|
311 |
+
" 'the capital of the united country ': 8,\n",
|
312 |
+
" 'the capital of the united europe ': 26,\n",
|
313 |
+
" 'the capital of the united is ': 7,\n",
|
314 |
+
" 'the capital of the united place ': 4,\n",
|
315 |
+
" 'the capital of the united europe,': 3,\n",
|
316 |
+
" 'the capital of the united united ': 6,\n",
|
317 |
+
" 'the capital of the united for ': 1,\n",
|
318 |
+
" 'the capital of the united spain,': 2,\n",
|
319 |
+
" 'the capital of the united europe.': 1,\n",
|
320 |
+
" 'the capital of the united italy,': 4,\n",
|
321 |
+
" 'the capital of the united art ': 1,\n",
|
322 |
+
" 'the capital of the united of ': 1,\n",
|
323 |
+
" 'the capital of the united united': 1,\n",
|
324 |
+
" 'the capital of the united capitaled': 1,\n",
|
325 |
+
" 'the capital of the united, country': 1,\n",
|
326 |
+
" 'the capital of the united place.': 1,\n",
|
327 |
+
" 'the capital of the united, europe': 1}"
|
328 |
+
]
|
329 |
+
},
|
330 |
+
"execution_count": 14,
|
331 |
+
"metadata": {},
|
332 |
+
"output_type": "execute_result"
|
333 |
+
}
|
334 |
+
],
|
335 |
+
"source": [
|
336 |
+
"outs = {}\n",
|
337 |
+
"for _ in range(100):\n",
|
338 |
+
" out = u_model.generate(tokens, max_new_tokens = 3, temperature = 1.7, top_k = 10, top_p = 0.7)\n",
|
339 |
+
" decoded = u_tokenizer.decode(out)\n",
|
340 |
+
" outs[decoded] = outs.get(decoded, 0) + 1\n",
|
341 |
+
"outs"
|
342 |
+
]
|
343 |
+
}
|
344 |
+
],
|
345 |
+
"metadata": {
|
346 |
+
"kernelspec": {
|
347 |
+
"display_name": "Python 3",
|
348 |
+
"language": "python",
|
349 |
+
"name": "python3"
|
350 |
+
},
|
351 |
+
"language_info": {
|
352 |
+
"codemirror_mode": {
|
353 |
+
"name": "ipython",
|
354 |
+
"version": 3
|
355 |
+
},
|
356 |
+
"file_extension": ".py",
|
357 |
+
"mimetype": "text/x-python",
|
358 |
+
"name": "python",
|
359 |
+
"nbconvert_exporter": "python",
|
360 |
+
"pygments_lexer": "ipython3",
|
361 |
+
"version": "3.13.3"
|
362 |
+
}
|
363 |
+
},
|
364 |
+
"nbformat": 4,
|
365 |
+
"nbformat_minor": 2
|
366 |
+
}
|
v1/u_model.pth
DELETED
Binary file (97.2 kB)
|
|
{v1 β v2}/__init__.py
RENAMED
File without changes
|
{v1 β v2}/tokenizer.json
RENAMED
File without changes
|
v2/u_model_4000.pth
ADDED
Binary file (96.1 kB). View file
|
|
{v1 β v2}/usta_causal_attention.py
RENAMED
File without changes
|
{v1 β v2}/usta_decoder_block.py
RENAMED
@@ -6,17 +6,23 @@ from .usta_multi_head_attention import UstaMultiHeadAttention
|
|
6 |
|
7 |
|
8 |
class UstaDecoderBlock(nn.Module):
|
9 |
-
def __init__(self, embedding_dim, num_heads, context_length):
|
10 |
super().__init__()
|
11 |
|
12 |
-
self.self_attention = UstaMultiHeadAttention(
|
13 |
-
|
14 |
-
|
15 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
|
17 |
def forward(self, x):
|
18 |
res = self.norm1(x)
|
19 |
-
|
20 |
x = self.self_attention(x)
|
21 |
x = self.norm1(x)
|
22 |
|
|
|
6 |
|
7 |
|
8 |
class UstaDecoderBlock(nn.Module):
|
9 |
+
def __init__(self, embedding_dim, num_heads, context_length, device):
|
10 |
super().__init__()
|
11 |
|
12 |
+
self.self_attention = UstaMultiHeadAttention(
|
13 |
+
embedding_dim,
|
14 |
+
embedding_dim,
|
15 |
+
context_length,
|
16 |
+
num_heads,
|
17 |
+
dropout_rate=0.5,
|
18 |
+
device=device
|
19 |
+
)
|
20 |
+
self.norm1 = UstaLayerNorm(embedding_dim, device=device)
|
21 |
+
self.mlp = UstaMLP(embedding_dim, embedding_dim, device=device)
|
22 |
+
self.norm2 = UstaLayerNorm(embedding_dim, device=device)
|
23 |
|
24 |
def forward(self, x):
|
25 |
res = self.norm1(x)
|
|
|
26 |
x = self.self_attention(x)
|
27 |
x = self.norm1(x)
|
28 |
|
{v1 β v2}/usta_embedding.py
RENAMED
@@ -3,7 +3,7 @@ import torch.nn as nn
|
|
3 |
|
4 |
|
5 |
def get_rotary_position_encoding(input: torch.Tensor, base=10000, device="cpu"):
|
6 |
-
context_length, dimension = input.shape
|
7 |
|
8 |
assert dimension % 2 == 0, "dimension must be even"
|
9 |
|
@@ -20,30 +20,31 @@ def get_rotary_position_encoding(input: torch.Tensor, base=10000, device="cpu"):
|
|
20 |
sin_angles = torch.sin(angles)
|
21 |
cos_angles = torch.cos(angles)
|
22 |
|
23 |
-
input_even = input[:, :dimension // 2] # [0, 2, 4, ..]
|
24 |
-
input_odd = input[:, dimension // 2:] # [1, 3, 5, ..]
|
25 |
|
26 |
input_even_rotated = input_even * cos_angles - input_odd * sin_angles
|
27 |
input_odd_rotated = input_even * sin_angles + input_odd * cos_angles
|
28 |
|
29 |
-
input_rotated = torch.empty_like(input)
|
30 |
|
31 |
-
input_rotated[:, :dimension // 2] = input_even_rotated
|
32 |
-
input_rotated[:, dimension // 2:] = input_odd_rotated
|
33 |
|
34 |
return input_rotated
|
35 |
|
36 |
class UstaEmbedding(nn.Module):
|
37 |
-
def __init__(self, vocab_size, embedding_dim, context_length):
|
38 |
super().__init__()
|
39 |
# position embedding but not being used in the forward pass
|
40 |
# it is just for educational purposes
|
41 |
# self.pos_embedding = nn.Embedding(context_length, embedding_dim)
|
42 |
# self.get_pos = get_rotary_position_encoding
|
43 |
-
self.embedding = nn.Embedding(vocab_size, embedding_dim)
|
44 |
self.get_pos = get_rotary_position_encoding
|
|
|
45 |
|
46 |
def forward(self, x):
|
47 |
x = self.embedding(x)
|
48 |
-
x = self.get_pos(x)
|
49 |
return x
|
|
|
3 |
|
4 |
|
5 |
def get_rotary_position_encoding(input: torch.Tensor, base=10000, device="cpu"):
|
6 |
+
batch_size, context_length, dimension = input.shape
|
7 |
|
8 |
assert dimension % 2 == 0, "dimension must be even"
|
9 |
|
|
|
20 |
sin_angles = torch.sin(angles)
|
21 |
cos_angles = torch.cos(angles)
|
22 |
|
23 |
+
input_even = input[:, :, :dimension // 2] # [0, 2, 4, ..]
|
24 |
+
input_odd = input[:, :, dimension // 2:] # [1, 3, 5, ..]
|
25 |
|
26 |
input_even_rotated = input_even * cos_angles - input_odd * sin_angles
|
27 |
input_odd_rotated = input_even * sin_angles + input_odd * cos_angles
|
28 |
|
29 |
+
input_rotated = torch.empty_like(input, device=device)
|
30 |
|
31 |
+
input_rotated[:, :, :dimension // 2] = input_even_rotated
|
32 |
+
input_rotated[:, :, dimension // 2:] = input_odd_rotated
|
33 |
|
34 |
return input_rotated
|
35 |
|
36 |
class UstaEmbedding(nn.Module):
|
37 |
+
def __init__(self, vocab_size, embedding_dim, context_length, device):
|
38 |
super().__init__()
|
39 |
# position embedding but not being used in the forward pass
|
40 |
# it is just for educational purposes
|
41 |
# self.pos_embedding = nn.Embedding(context_length, embedding_dim)
|
42 |
# self.get_pos = get_rotary_position_encoding
|
43 |
+
self.embedding = nn.Embedding(vocab_size, embedding_dim, device=device)
|
44 |
self.get_pos = get_rotary_position_encoding
|
45 |
+
self.device = device
|
46 |
|
47 |
def forward(self, x):
|
48 |
x = self.embedding(x)
|
49 |
+
x = self.get_pos(x, device=self.device)
|
50 |
return x
|
{v1 β v2}/usta_layer_norm.py
RENAMED
@@ -3,13 +3,12 @@ import torch.nn as nn
|
|
3 |
|
4 |
|
5 |
class UstaLayerNorm(nn.Module):
|
6 |
-
def __init__(self, embedding_dim, eps=1e-5):
|
7 |
super().__init__()
|
8 |
self.eps = eps
|
|
|
|
|
9 |
|
10 |
-
self.weight = nn.Parameter(torch.ones(embedding_dim))
|
11 |
-
|
12 |
-
|
13 |
def forward(self, x):
|
14 |
mean = x.mean(dim=-1, keepdim=True)
|
15 |
variance = x.var(dim=-1, keepdim=True, unbiased=False)
|
|
|
3 |
|
4 |
|
5 |
class UstaLayerNorm(nn.Module):
|
6 |
+
def __init__(self, embedding_dim, eps=1e-5, device="cpu"):
|
7 |
super().__init__()
|
8 |
self.eps = eps
|
9 |
+
self.weight = nn.Parameter(torch.ones(embedding_dim, device=device))
|
10 |
+
self.device = device
|
11 |
|
|
|
|
|
|
|
12 |
def forward(self, x):
|
13 |
mean = x.mean(dim=-1, keepdim=True)
|
14 |
variance = x.var(dim=-1, keepdim=True, unbiased=False)
|
{v1 β v2}/usta_mlp.py
RENAMED
@@ -14,13 +14,13 @@ class GELU(nn.Module):
|
|
14 |
)
|
15 |
|
16 |
class UstaMLP(nn.Module):
|
17 |
-
def __init__(self, embedding_dim, hidden_dim):
|
18 |
super().__init__()
|
19 |
|
20 |
-
self.gate_proj = nn.Linear(embedding_dim, hidden_dim)
|
21 |
-
self.up_proj = nn.Linear(embedding_dim, hidden_dim)
|
22 |
-
self.down_proj = nn.Linear(hidden_dim, embedding_dim)
|
23 |
-
self.gelu = GELU()
|
24 |
|
25 |
def forward(self, x):
|
26 |
""" gate = self.gate_proj(x)
|
|
|
14 |
)
|
15 |
|
16 |
class UstaMLP(nn.Module):
|
17 |
+
def __init__(self, embedding_dim, hidden_dim, device="cpu"):
|
18 |
super().__init__()
|
19 |
|
20 |
+
self.gate_proj = nn.Linear(embedding_dim, hidden_dim, device=device)
|
21 |
+
self.up_proj = nn.Linear(embedding_dim, hidden_dim, device=device)
|
22 |
+
self.down_proj = nn.Linear(hidden_dim, embedding_dim, device=device)
|
23 |
+
self.gelu = GELU().to(device)
|
24 |
|
25 |
def forward(self, x):
|
26 |
""" gate = self.gate_proj(x)
|
{v1 β v2}/usta_model.py
RENAMED
@@ -6,15 +6,16 @@ from .usta_embedding import UstaEmbedding
|
|
6 |
|
7 |
|
8 |
class UstaModel(nn.Module):
|
9 |
-
def __init__(self, vocab_size, embedding_dim, num_heads, context_length, num_layers):
|
10 |
super().__init__()
|
11 |
|
12 |
-
self.embedding = UstaEmbedding(vocab_size, embedding_dim, context_length)
|
13 |
self.layers = nn.Sequential(
|
14 |
-
*[UstaDecoderBlock(embedding_dim, num_heads, context_length) for _ in range(num_layers)]
|
15 |
)
|
16 |
|
17 |
-
self.lm_head = nn.Linear(embedding_dim, vocab_size)
|
|
|
18 |
|
19 |
def forward(self, x: torch.Tensor):
|
20 |
x = self.embedding(x) # dictionary meaning of the tokens (words)
|
@@ -32,13 +33,49 @@ class UstaModel(nn.Module):
|
|
32 |
max_prob, max_index, probs
|
33 |
"""
|
34 |
|
35 |
-
def
|
36 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
37 |
|
38 |
for _ in range(max_new_tokens):
|
|
|
39 |
out = self.forward(x)
|
40 |
-
|
41 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
tokens.append(max_index.item())
|
43 |
if max_index == 59 or len(tokens) > 32: # <eos> and max context length
|
44 |
break
|
|
|
6 |
|
7 |
|
8 |
class UstaModel(nn.Module):
|
9 |
+
def __init__(self, vocab_size, embedding_dim, num_heads, context_length, num_layers, device):
|
10 |
super().__init__()
|
11 |
|
12 |
+
self.embedding = UstaEmbedding(vocab_size, embedding_dim, context_length, device)
|
13 |
self.layers = nn.Sequential(
|
14 |
+
*[UstaDecoderBlock(embedding_dim, num_heads, context_length, device) for _ in range(num_layers)]
|
15 |
)
|
16 |
|
17 |
+
self.lm_head = nn.Linear(embedding_dim, vocab_size, device=device)
|
18 |
+
self.device = device
|
19 |
|
20 |
def forward(self, x: torch.Tensor):
|
21 |
x = self.embedding(x) # dictionary meaning of the tokens (words)
|
|
|
33 |
max_prob, max_index, probs
|
34 |
"""
|
35 |
|
36 |
+
def top_p_filtering(self, logits, top_p):
|
37 |
+
sorted_logits, sorted_indices = torch.sort(logits, descending=True)
|
38 |
+
cumulative_probs = torch.cumsum(torch.softmax(sorted_logits, dim=-1), dim=-1)
|
39 |
+
sorted_indices_to_remove = cumulative_probs > top_p
|
40 |
+
sorted_indices_to_remove[..., 1:] = sorted_indices_to_remove[..., :-1].clone()
|
41 |
+
sorted_indices_to_remove[..., 0] = False
|
42 |
+
|
43 |
+
sorted_logits[sorted_indices_to_remove] = -float('inf')
|
44 |
+
filtered_logits = sorted_logits.clone()
|
45 |
+
filtered_logits.scatter_(0, sorted_indices, sorted_logits)
|
46 |
+
return filtered_logits
|
47 |
+
|
48 |
+
|
49 |
+
|
50 |
+
def generate(self,
|
51 |
+
x: torch.Tensor,
|
52 |
+
max_new_tokens: int = 3,
|
53 |
+
temperature: float = 1.0,
|
54 |
+
top_k: int = 64,
|
55 |
+
top_p: float = 1.0
|
56 |
+
): # top_k, top_p, temperature
|
57 |
+
tokens = x.tolist()
|
58 |
|
59 |
for _ in range(max_new_tokens):
|
60 |
+
x = x.unsqueeze(0).to(self.device)
|
61 |
out = self.forward(x)
|
62 |
+
out = out.squeeze(0)
|
63 |
+
logits = out[-1]
|
64 |
+
if top_k > 0:
|
65 |
+
values, indexes = torch.topk(logits, k=top_k)
|
66 |
+
logits = torch.full_like(logits, -float('inf'))
|
67 |
+
logits.scatter_(0, indexes, values)
|
68 |
+
|
69 |
+
if top_p > 0 and top_p < 1:
|
70 |
+
logits = self.top_p_filtering(logits, top_p)
|
71 |
+
|
72 |
+
if temperature != 1.0 and temperature > 0:
|
73 |
+
logits = logits / temperature
|
74 |
+
|
75 |
+
probs = torch.softmax(values, dim=-1)
|
76 |
+
# _, max_index = torch.max(probs, dim=-1)
|
77 |
+
sample = torch.multinomial(probs, 1)
|
78 |
+
max_index = indexes[sample]
|
79 |
tokens.append(max_index.item())
|
80 |
if max_index == 59 or len(tokens) > 32: # <eos> and max context length
|
81 |
break
|
{v1 β v2}/usta_multi_head_attention.py
RENAMED
@@ -3,15 +3,15 @@ import torch.nn as nn
|
|
3 |
|
4 |
|
5 |
class UstaMultiHeadAttention(nn.Module):
|
6 |
-
def __init__(self, embedding_dim, output_dim, context_length, num_heads, dropout_rate = 0):
|
7 |
super().__init__()
|
8 |
|
9 |
self.context_length = context_length
|
10 |
|
11 |
-
self.multi_head_attention = nn.MultiheadAttention(embedding_dim, num_heads, dropout=dropout_rate)
|
12 |
-
self.projection = nn.Linear(embedding_dim, output_dim)
|
13 |
|
14 |
-
self.register_buffer("mask", torch.triu(torch.ones(context_length, context_length), diagonal=1).bool())
|
15 |
|
16 |
def forward(self, x):
|
17 |
number_of_tokens = x.shape[0]
|
|
|
3 |
|
4 |
|
5 |
class UstaMultiHeadAttention(nn.Module):
|
6 |
+
def __init__(self, embedding_dim, output_dim, context_length, num_heads, dropout_rate = 0, device="cpu"):
|
7 |
super().__init__()
|
8 |
|
9 |
self.context_length = context_length
|
10 |
|
11 |
+
self.multi_head_attention = nn.MultiheadAttention(embedding_dim, num_heads, dropout=dropout_rate, device=device)
|
12 |
+
self.projection = nn.Linear(embedding_dim, output_dim, device=device)
|
13 |
|
14 |
+
self.register_buffer("mask", torch.triu(torch.ones(context_length, context_length), diagonal=1).bool().to(device))
|
15 |
|
16 |
def forward(self, x):
|
17 |
number_of_tokens = x.shape[0]
|
{v1 β v2}/usta_multi_head_attention_old.py
RENAMED
@@ -22,5 +22,4 @@ class UstaMultiHeadAttention(nn.Module):
|
|
22 |
|
23 |
attention_out = torch.cat(attention_outs, dim=1)
|
24 |
|
25 |
-
return self.projection(attention_out)
|
26 |
-
|
|
|
22 |
|
23 |
attention_out = torch.cat(attention_outs, dim=1)
|
24 |
|
25 |
+
return self.projection(attention_out)
|
|
{v1 β v2}/usta_self_attention.py
RENAMED
File without changes
|
{v1 β v2}/usta_tokenizer.py
RENAMED
@@ -9,6 +9,19 @@ class UstaTokenizer:
|
|
9 |
self.vocab = json.load(f)
|
10 |
self.reverse_vocab = {v: k for k, v in self.vocab.items()}
|
11 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
def encode(self, text):
|
13 |
tokens = []
|
14 |
|
@@ -31,7 +44,9 @@ class UstaTokenizer:
|
|
31 |
i += 1
|
32 |
tokens.append(self.vocab[" "])
|
33 |
|
34 |
-
|
|
|
|
|
35 |
return torch.tensor(tokens)
|
36 |
|
37 |
def tokenize(self, text):
|
|
|
9 |
self.vocab = json.load(f)
|
10 |
self.reverse_vocab = {v: k for k, v in self.vocab.items()}
|
11 |
|
12 |
+
def encode_batch(self, texts, context_length):
|
13 |
+
sentences_tokens = []
|
14 |
+
for text in texts:
|
15 |
+
tokens = self.encode(text).tolist()
|
16 |
+
if len(tokens) > context_length:
|
17 |
+
tokens = tokens[:context_length]
|
18 |
+
else:
|
19 |
+
tokens = tokens + [self.vocab["<pad>"]] * (context_length - len(tokens))
|
20 |
+
|
21 |
+
sentences_tokens.append(tokens)
|
22 |
+
|
23 |
+
return torch.tensor(sentences_tokens)
|
24 |
+
|
25 |
def encode(self, text):
|
26 |
tokens = []
|
27 |
|
|
|
44 |
i += 1
|
45 |
tokens.append(self.vocab[" "])
|
46 |
|
47 |
+
# check if text is not ends with a space
|
48 |
+
if not text.endswith(" "):
|
49 |
+
tokens.pop()
|
50 |
return torch.tensor(tokens)
|
51 |
|
52 |
def tokenize(self, text):
|