File size: 7,405 Bytes
58a71ea
664df97
 
 
 
aa82a83
 
664df97
 
 
cfc8af7
 
 
 
 
664df97
cfc8af7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
664df97
cfc8af7
664df97
 
 
 
 
 
 
 
 
 
 
 
 
 
cfc8af7
664df97
 
 
cfc8af7
 
664df97
 
aa82a83
 
 
664df97
 
 
 
 
 
 
 
 
cfc8af7
 
 
 
 
664df97
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
cfc8af7
664df97
 
 
cfc8af7
664df97
 
 
58a71ea
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
664df97
 
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
"""import gradio as gr
import torch
import time
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
from threading import Thread
import pytz
from datetime import datetime
print("Loading model and tokenizer...")
model_name = "large-traversaal/Phi-4-Hindi"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)
print("Model and tokenizer loaded successfully!")
option_mapping = {
    "translation": "### TRANSLATION ###",
    "mcq": "### MCQ ###",
    "nli": "### NLI ###",
    "summarization": "### SUMMARIZATION ###",
    "long response": "### LONG RESPONSE ###",
    "short response": "### SHORT RESPONSE ###",
    "direct response": "### DIRECT RESPONSE ###",
    "paraphrase": "### PARAPHRASE ###",
    "code": "### CODE ###"
}
def generate_response(message, temperature, max_new_tokens, top_p, task):
    append_text = option_mapping.get(task, "")
    prompt = f"INPUT : {message} {append_text} RESPONSE : "
    print(f"Prompt: {prompt}")
    start_time = time.time()
    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
    streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=True)
    gen_kwargs = {
        "input_ids": inputs["input_ids"],
        "streamer": streamer,
        "temperature": temperature,
        "max_new_tokens": max_new_tokens,
        "top_p": top_p,
        "do_sample": True if temperature > 0 else False,
    }
    thread = Thread(target=model.generate, kwargs=gen_kwargs)
    thread.start()
    result = []
    for text in streamer:
        result.append(text)
        yield "".join(result)
    end_time = time.time()
    time_taken = end_time - start_time
    output_text = "".join(result)
    if "RESPONSE : " in output_text:
        output_text = output_text.split("RESPONSE : ", 1)[1].strip()
    print(f"Output: {output_text}")
    print(f"Time taken: {time_taken:.2f} seconds")
    pst_timezone = pytz.timezone('America/Los_Angeles')
    current_time_pst = datetime.now(pst_timezone).strftime("%Y-%m-%d %H:%M:%S %Z%z")
    print(f"Current timestamp (PST): {current_time_pst}")
with gr.Blocks() as demo:
    gr.Markdown("# Phi-4-Hindi Demo")
    with gr.Row():
        with gr.Column():
            input_text = gr.Textbox(
                label="Input",
                placeholder="Enter your text here...",
                lines=5
            )
            task_dropdown = gr.Dropdown(
                choices=["translation", "mcq", "nli", "summarization", "long response", "short response", "direct response", "paraphrase", "code"],
                value="long response",
                label="Task"
            )
            with gr.Row():
                with gr.Column():
                    temperature = gr.Slider(
                        minimum=0.0,
                        maximum=1.0,
                        value=0.1,
                        step=0.01,
                        label="Temperature"
                    )
                with gr.Column():
                    max_new_tokens = gr.Slider(
                        minimum=50,
                        maximum=1000,
                        value=400,
                        step=10,
                        label="Max New Tokens"
                    )
                with gr.Column():
                    top_p = gr.Slider(
                        minimum=0.0,
                        maximum=1.0,
                        value=0.1,
                        step=0.01,
                        label="Top P"
                    )
            with gr.Row():
                clear_btn = gr.Button("Clear")
                send_btn = gr.Button("Send", variant="primary")
        with gr.Column():
            output_text = gr.Textbox(
                label="Output",
                lines=15
            )
    send_btn.click(
        fn=generate_response,
        inputs=[input_text, temperature, max_new_tokens, top_p, task_dropdown],
        outputs=output_text
    )
    clear_btn.click(
        fn=lambda: ("", ""),
        inputs=None,
        outputs=[input_text, output_text]
    )
if __name__ == "__main__":
    demo.queue().launch()
"""
import gradio as gr
import torch
import time
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
from threading import Thread
import pytz
from datetime import datetime
print("Loading model and tokenizer...")
model_name = "large-traversaal/Phi-4-Hindi"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")
print("Model and tokenizer loaded successfully!")
option_mapping = {"translation": "### TRANSLATION ###", "mcq": "### MCQ ###", "nli": "### NLI ###", "summarization": "### SUMMARIZATION ###",
    "long response": "### LONG RESPONSE ###", "direct response": "### DIRECT RESPONSE ###", "paraphrase": "### PARAPHRASE ###", "code": "### CODE ###"}
def generate_response(message, temperature, max_new_tokens, top_p, task):
    append_text = option_mapping.get(task, "")
    prompt = f"INPUT : {message} {append_text} ### RESPONSE : "
    print(f"Prompt: {prompt}")
    start_time = time.time()
    inputs = tokenizer.encode(prompt, return_tensors="pt").to(model.device)
    outputs = model.generate(input_ids=inputs, max_new_tokens=max_new_tokens, use_cache=True, temperature=temperature, min_p=top_p, pad_token_id=tokenizer.eos_token_id)
    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
    processed_response = response.split("### RESPONSE :assistant")[-1].strip()
    end_time = time.time()
    time_taken = end_time - start_time
    print(f"Output: {processed_response}")
    print(f"Time taken: {time_taken:.2f} seconds")
    pst_timezone = pytz.timezone('America/Los_Angeles')
    current_time_pst = datetime.now(pst_timezone).strftime("%Y-%m-%d %H:%M:%S %Z%z")
    print(f"Current timestamp (PST): {current_time_pst}")
    return processed_response
with gr.Blocks() as demo:
    gr.Markdown("# Phi-4-Hindi Demo")
    with gr.Row():
        with gr.Column():
            input_text = gr.Textbox(label="Input", placeholder="Enter your text here...", lines=5)
            task_dropdown = gr.Dropdown(choices=["translation", "mcq", "nli", "summarization", "long response", "direct response", "paraphrase", "code"], value="long response", label="Task")
            with gr.Row():
                with gr.Column():
                    temperature = gr.Slider(minimum=0.0, maximum=1.0, value=0.1, step=0.01, label="Temperature")
                with gr.Column():
                    max_new_tokens = gr.Slider(minimum=10, maximum=1000, value=10, step=10, label="Max New Tokens")
                with gr.Column():
                    top_p = gr.Slider(minimum=0.0, maximum=1.0, value=0.1, step=0.01, label="Top P")
            with gr.Row():
                clear_btn = gr.Button("Clear")
                send_btn = gr.Button("Send", variant="primary")
        with gr.Column():
            output_text = gr.Textbox(label="Output", lines=15)
    send_btn.click(fn=generate_response, inputs=[input_text, temperature, max_new_tokens, top_p, task_dropdown], outputs=output_text)
    clear_btn.click(fn=lambda: ("", ""), inputs=None, outputs=[input_text, output_text])
if __name__ == "__main__":
    demo.queue().launch()