removed unnecesary files

placeholder.py

@@ -1,175 +0,0 @@
-import os
-import sys
-import json
-import time
-import openai
-import pickle
-import argparse
-import requests
-from tqdm import tqdm
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaForCausalLM, LlamaTokenizer
-
-from fastchat.model import load_model, get_conversation_template, add_model_args
-
-from nltk.tag.mapping import _UNIVERSAL_TAGS
-
-import gradio as gr
-from transformers import pipeline
-
-demo = gr.Blocks()
-
-uni_tags = list(_UNIVERSAL_TAGS)
-uni_tags[-1] = 'PUNC'
-
-bio_tags = ['B', 'I', 'O']
-chunk_tags = ['ADJP', 'ADVP', 'CONJP', 'INTJ', 'LST', 'NP', 'O', 'PP', 'PRT', 'SBAR', 'UCP', 'VP']
-
-syntags = ['NP', 'S', 'VP', 'ADJP', 'ADVP', 'SBAR', 'TOP', 'PP', 'POS', 'NAC', "''", 'SINV', 'PRN', 'QP', 'WHNP', 'RB', 'FRAG',
- 'WHADVP', 'NX', 'PRT', 'VBZ', 'VBP', 'MD', 'NN', 'WHPP', 'SQ', 'SBARQ', 'LST', 'INTJ', 'X', 'UCP', 'CONJP', 'NNP', 'CD', 'JJ',
- 'VBD', 'WHADJP', 'PRP', 'RRC', 'NNS', 'SYM', 'CC']
-
-openai.api_key = " "
-
-# determinant vs. determiner
-# https://wikidiff.com/determiner/determinant
-ents_prompt = ['Noun','Verb','Adjective','Adverb','Preposition/Subord','Coordinating Conjunction',# 'Cardinal Number',
-    'Determiner',
-    'Noun Phrase','Verb Phrase','Adjective Phrase','Adverb Phrase','Preposition Phrase','Conjunction Phrase','Coordinate Phrase','Quantitave Phrase','Complex Nominal',
-    'Clause','Dependent Clause','Fragment Clause','T-unit','Complex T-unit',# 'Fragment T-unit',
-][7:]
-ents = ['NN', 'VB', 'JJ', 'RB', 'IN', 'CC', 'DT', 'NP', 'VP', 'ADJP', 'ADVP', 'PP', 'CONJP', 'CP', 'QP', 'CN', 'C', 'DC', 'FC', 'T', 'CT'][7:]
-
-
-ents_prompt_uni_tags = ['Verb', 'Noun', 'Pronoun', 'Adjective', 'Adverb', 'Preposition and Postposition', 'Coordinating Conjunction', 
-                        'Determiner', 'Cardinal Number', 'Particles or other function words', 
-                        'Words that cannot be assigned a POS tag', 'Punctuation']
-
-ents = uni_tags + ents
-ents_prompt = ents_prompt_uni_tags + ents_prompt
-
-for i, j in zip(ents, ents_prompt):
-    print(i, j)
-
-model_mapping = {
-    'gpt3.5': 'gpt2',
-    #'vicuna-7b': 'lmsys/vicuna-7b-v1.3',
-    #'llama-7b': './llama/hf/7B',
-}
-
-with open('sample_uniform_1k_2.txt', 'r') as f:
-    selected_idx = f.readlines()
-selected_idx = [int(i.strip()) for i in selected_idx]#[s:e]
-
-ptb = []
-with open('ptb.jsonl', 'r') as f:
-    for l in f:
-        ptb.append(json.loads(l))
-
-
-## Prompt 1
-template_all = '''Please output the <Noun, Verb, Adjective, Adverb, Preposition/Subord, Coordinating Conjunction, Cardinal Number, Determiner, Noun Phrase, Verb Phrase, Adjective Phrase, Adverb Phrase, Preposition Phrase, Conjunction Phrase, Coordinate Phrase, Quantitave Phrase, Complex Nominal, Clause, Dependent Clause, Fragment Clause, T-unit, Complex T-unit, Fragment T-unit> in the following sentence without any additional text in json format: "{}"'''
-template_single = '''Please output any <{}> in the following sentence one per line without any additional text: "{}"'''
-
-## Prompt 2
-prompt2_pos = '''Please pos tag the following sentence using Universal POS tag set without generating any additional text: {}'''
-prompt2_chunk = '''Please do sentence chunking for the following sentence as in CoNLL 2000 shared task without generating any addtional text: {}'''
-prompt2_parse = '''Generate textual representation of the constituency parse tree of the following sentence using Penn TreeBank tag set without outputing any additional text: {}'''
-
-prompt2_chunk = '''Please chunk the following sentence in CoNLL 2000 format with BIO tags without outputing any additional text: {}'''
-
-## Prompt 3
-with open('demonstration_3_42_pos.txt', 'r') as f:
-    demon_pos = f.read()
-with open('demonstration_3_42_chunk.txt', 'r') as f:
-    demon_chunk = f.read()
-with open('demonstration_3_42_parse.txt', 'r') as f:
-    demon_parse = f.read()
-
-# Your existing code
-theme = gr.themes.Soft()
-
-# issue get request for gpt 3.5
-gpt_pipeline = pipeline(task="text2text-generation", model="gpt2")
-#vicuna7b_pipeline = pipeline(task="text2text-generation", model="lmsys/vicuna-7b-v1.3")
-#llama7b_pipeline = pipeline(task="text2text-generation", model="./llama/hf/7B")
-
-# Dropdown options for model and task
-model_options = list(model_mapping.keys())
-task_options = ['POS', 'Chunking'] # remove parsing
-
-
-# Function to process text based on model and task
-def process_text(tab, text):
-    if tab == 'POS Tab':
-        strategy1_format = template_all.format(text)
-        strategy2_format = prompt2_pos.format(text)
-        strategy3_format = demon_pos
-
-        vicuna_result1 = gpt_pipeline(strategy1_format)[0]['generated_text']
-        vicuna_result2 = gpt_pipeline(strategy2_format)[0]['generated_text']
-        vicuna_result3 = gpt_pipeline(strategy3_format)[0]['generated_text']
-        
-        return (vicuna_result1, vicuna_result2, vicuna_result3)
-    elif tab == 'Chunk Tab':
-        strategy1_format = template_all.format(text)
-        strategy2_format = prompt2_chunk.format(text)
-        strategy3_format = demon_chunk
-
-        result1 = gpt_pipeline(strategy1_format)[0]['generated_text']
-        result2 = gpt_pipeline(strategy2_format)[0]['generated_text']
-        result3 = gpt_pipeline(strategy3_format)[0]['generated_text']
-        return (result1, result2, result3)
-
-# Gradio interface
-with demo:
-    gr.Markdown("# LLM Evaluator With Linguistic Scrutiny")
-    
-    with gr.Tabs():
-        with gr.TabItem("POS", id="POS Tab"):
-            with gr.Row():
-              gr.Markdown("<center>Vicuna 7b</center>")
-              gr.Markdown("<center> LLaMA-7b </center>")
-              gr.Markdown("<center> GPT 3.5 </center>")
-            with gr.Row():
-                model1_S1_output = gr.Textbox(label="Strategy 1 QA")
-                model2_S1_output = gr.Textbox(label=".")
-                model3_S1_output = gr.Textbox(label=".")
-            with gr.Row():
-                model1_S2_output = gr.Textbox(label="Strategy 2 Instruction")
-                model2_S2_output = gr.Textbox(label=".")
-                model3_S2_output = gr.Textbox(label=".")
-            with gr.Row():
-                model1_S3_output = gr.Textbox(label="Strategy 3 Structured Prompting")
-                model2_S3_output = gr.Textbox(label=".")
-                model3_S3_output = gr.Textbox(label=".")
-            with gr.Row():
-                prompt_POS = gr.Textbox(show_label=False, placeholder="Enter prompt")
-                send_button_POS = gr.Button("Send", scale=0)
-              
-        with gr.TabItem("Chunking", id="Chunk Tab"):
-            with gr.Row():
-                gr.Markdown("<center>Vicuna 7b</center>")
-                gr.Markdown("<center> LLaMA-7b </center>")
-                gr.Markdown("<center> GPT 3.5 </center>")
-            with gr.Row():
-                model1_S1_output = gr.Textbox(label="Strategy 1 QA")
-                model2_S1_output = gr.Textbox(label=".")
-                model3_S1_output = gr.Textbox(label=".")
-            with gr.Row():
-                model1_S2_output = gr.Textbox(label="Strategy 2 Instruction")
-                model2_S2_output = gr.Textbox(label=".")
-                model3_S2_output = gr.Textbox(label=".")
-            with gr.Row():
-                model1_S3_output = gr.Textbox(label="Strategy 3 Structured Prompting")
-                model2_S3_output = gr.Textbox(label=".")
-                model3_S3_output = gr.Textbox(label=".")
-            with gr.Row():
-                prompt_Chunk = gr.Textbox(id="prompt_Chunk", show_label=False, placeholder="Enter prompt")
-                send_button_Chunk = gr.Button("Send", scale=0)
-    
-    send_button_POS.click(process_text, inputs=["POS Tab", prompt_Chunk], outputs=[model1_S1_output, model1_S1_output, model1_S1_output])
-    send_button_Chunk.click(process_text, inputs=["Chunk Tab", prompt_POS], outputs=[model1_S1_output, model1_S1_output, model1_S1_output])
-
-demo.launch()
-

run_llm2.py

@@ -1,147 +0,0 @@
-import os
-import sys
-import json
-import time
-import openai
-import pickle
-import argparse
-import requests
-from tqdm import tqdm
-import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM, LlamaForCausalLM, LlamaTokenizer
-
-from fastchat.model import load_model, get_conversation_template, add_model_args
-
-from nltk.tag.mapping import _UNIVERSAL_TAGS
-
-import gradio as gr
-from transformers import pipeline
-
-uni_tags = list(_UNIVERSAL_TAGS)
-uni_tags[-1] = 'PUNC'
-
-bio_tags = ['B', 'I', 'O']
-chunk_tags = ['ADJP', 'ADVP', 'CONJP', 'INTJ', 'LST', 'NP', 'O', 'PP', 'PRT', 'SBAR', 'UCP', 'VP']
-
-syntags = ['NP', 'S', 'VP', 'ADJP', 'ADVP', 'SBAR', 'TOP', 'PP', 'POS', 'NAC', "''", 'SINV', 'PRN', 'QP', 'WHNP', 'RB', 'FRAG',
- 'WHADVP', 'NX', 'PRT', 'VBZ', 'VBP', 'MD', 'NN', 'WHPP', 'SQ', 'SBARQ', 'LST', 'INTJ', 'X', 'UCP', 'CONJP', 'NNP', 'CD', 'JJ',
- 'VBD', 'WHADJP', 'PRP', 'RRC', 'NNS', 'SYM', 'CC']
-
-openai.api_key = " "
-
-# determinant vs. determiner
-# https://wikidiff.com/determiner/determinant
-ents_prompt = ['Noun','Verb','Adjective','Adverb','Preposition/Subord','Coordinating Conjunction',# 'Cardinal Number',
-    'Determiner',
-    'Noun Phrase','Verb Phrase','Adjective Phrase','Adverb Phrase','Preposition Phrase','Conjunction Phrase','Coordinate Phrase','Quantitave Phrase','Complex Nominal',
-    'Clause','Dependent Clause','Fragment Clause','T-unit','Complex T-unit',# 'Fragment T-unit',
-][7:]
-ents = ['NN', 'VB', 'JJ', 'RB', 'IN', 'CC', 'DT', 'NP', 'VP', 'ADJP', 'ADVP', 'PP', 'CONJP', 'CP', 'QP', 'CN', 'C', 'DC', 'FC', 'T', 'CT'][7:]
-
-
-ents_prompt_uni_tags = ['Verb', 'Noun', 'Pronoun', 'Adjective', 'Adverb', 'Preposition and Postposition', 'Coordinating Conjunction', 
-                        'Determiner', 'Cardinal Number', 'Particles or other function words', 
-                        'Words that cannot be assigned a POS tag', 'Punctuation']
-
-ents = uni_tags + ents
-ents_prompt = ents_prompt_uni_tags + ents_prompt
-
-for i, j in zip(ents, ents_prompt):
-    print(i, j)
-
-model_mapping = {
-    'gpt3.5': 'gpt2',
-    #'vicuna-7b': 'lmsys/vicuna-7b-v1.3',
-    #'llama-7b': './llama/hf/7B',
-}
-
-with open('sample_uniform_1k_2.txt', 'r') as f:
-    selected_idx = f.readlines()
-selected_idx = [int(i.strip()) for i in selected_idx]#[s:e]
-
-ptb = []
-with open('ptb.jsonl', 'r') as f:
-    for l in f:
-        ptb.append(json.loads(l))
-
-
-## Prompt 1
-template_all = '''Please output the <Noun, Verb, Adjective, Adverb, Preposition/Subord, Coordinating Conjunction, Cardinal Number, Determiner, Noun Phrase, Verb Phrase, Adjective Phrase, Adverb Phrase, Preposition Phrase, Conjunction Phrase, Coordinate Phrase, Quantitave Phrase, Complex Nominal, Clause, Dependent Clause, Fragment Clause, T-unit, Complex T-unit, Fragment T-unit> in the following sentence without any additional text in json format: "{}"'''
-template_single = '''Please output any <{}> in the following sentence one per line without any additional text: "{}"'''
-
-## Prompt 2
-prompt2_pos = '''Please pos tag the following sentence using Universal POS tag set without generating any additional text: {}'''
-prompt2_chunk = '''Please do sentence chunking for the following sentence as in CoNLL 2000 shared task without generating any addtional text: {}'''
-prompt2_parse = '''Generate textual representation of the constituency parse tree of the following sentence using Penn TreeBank tag set without outputing any additional text: {}'''
-
-prompt2_chunk = '''Please chunk the following sentence in CoNLL 2000 format with BIO tags without outputing any additional text: {}'''
-
-## Prompt 3
-with open('demonstration_3_42_pos.txt', 'r') as f:
-    demon_pos = f.read()
-with open('demonstration_3_42_chunk.txt', 'r') as f:
-    demon_chunk = f.read()
-with open('demonstration_3_42_parse.txt', 'r') as f:
-    demon_parse = f.read()
-
-# Your existing code
-theme = gr.themes.Soft()
-
-# issue get request for gpt 3.5
-gpt_pipeline = pipeline(task="text2text-generation", model="gpt2")
-#vicuna7b_pipeline = pipeline(task="text2text-generation", model="lmsys/vicuna-7b-v1.3")
-#llama7b_pipeline = pipeline(task="text2text-generation", model="./llama/hf/7B")
-
-# Dropdown options for model and task
-model_options = list(model_mapping.keys())
-task_options = ['POS', 'Chunking'] # remove parsing
-
-
-# Function to process text based on model and task
-def process_text(model_name, task, text):
-    gid_list = selected_idx[0:20]
-
-    for gid in tqdm(gid_list, desc='Query'):
-            text = ptb[gid]['text']
-    
-    if model_name == 'vicuna-7b': 
-        if task == 'POS':
-            strategy1_format = template_all.format(text)
-            strategy2_format = prompt2_pos.format(text)
-            strategy3_format = demon_pos
-
-            result1 = gpt_pipeline(strategy1_format)[0]['generated_text']
-            result2 = gpt_pipeline(strategy2_format)[0]['generated_text']
-            result3 = gpt_pipeline(strategy3_format)[0]['generated_text']
-            return (result1, result2, result3)
-        elif task == 'Chunking':
-            strategy1_format = template_all.format(text)
-            strategy2_format = prompt2_chunk.format(text)
-            strategy3_format = demon_chunk
-
-            result1 = gpt_pipeline(strategy1_format)[0]['generated_text']
-            result2 = gpt_pipeline(strategy2_format)[0]['generated_text']
-            result3 = gpt_pipeline(strategy3_format)[0]['generated_text']
-            return (result1, result2, result3)
-
-# Gradio interface
-iface = gr.Interface(
-    fn=process_text,
-    inputs=[
-        gr.Dropdown(model_options, label="Select Model"),
-        gr.Dropdown(task_options, label="Select Task"),
-        gr.Textbox(label="Input Text", placeholder="Enter the text to process..."),
-    ],
-    outputs=[
-        gr.Textbox(label="Strategy 1 QA Result"),
-        gr.Textbox(label="Strategy 2 Instruction Result"),
-        gr.Textbox(label="Strategy 3 Structured Prompting Result"),
-    ],
-    title = "LLM Evaluator For Linguistic Scrutiny",
-    theme = theme,
-    live=False,
-)
-
-iface.launch()
-
-

sample_uniform_1k_2.txt

@@ -1,1000 +0,0 @@
-22
-40
-199
-444
-457
-669
-679
-1215
-1412
-1423
-1429
-1460
-1491
-1647
-1671
-1740
-1755
-1779
-1823
-1878
-2036
-2145
-2161
-2185
-2219
-2243
-2328
-2352
-2615
-2657
-2666
-2713
-2870
-2884
-2909
-2915
-2936
-3005
-3065
-3095
-3194
-3251
-3375
-3404
-3679
-3864
-3904
-4016
-4041
-4045
-4064
-4068
-4070
-4072
-4079
-4090
-4134
-4425
-4450
-4455
-4457
-4464
-4523
-4524
-4541
-4550
-4575
-4614
-4634
-4715
-5013
-5015
-5021
-5029
-5040
-5057
-5070
-5121
-5250
-5257
-5328
-5338
-5351
-5415
-5497
-5501
-5536
-5576
-5599
-5617
-5641
-5683
-5702
-5709
-5711
-5716
-5771
-5877
-5888
-5899
-5921
-5945
-5991
-6059
-6065
-6066
-6087
-6096
-6114
-6256
-6265
-6330
-6520
-6633
-6943
-7051
-7108
-7526
-7681
-7748
-7763
-7768
-7778
-7878
-7906
-7936
-7975
-8035
-8185
-8431
-8942
-9070
-9212
-9584
-9615
-9626
-9631
-9709
-9711
-9722
-9870
-9874
-9878
-9942
-9946
-9947
-9958
-9973
-9987
-10017
-10030
-10244
-10250
-10284
-10333
-10438
-10645
-10647
-10888
-10995
-11083
-11220
-11349
-11416
-11491
-11498
-11630
-11874
-11889
-11972
-12296
-12369
-12382
-12504
-12580
-12606
-12611
-12623
-12823
-12884
-13128
-13145
-13339
-13358
-13368
-13689
-13709
-13839
-14041
-14042
-14130
-14475
-14946
-15614
-15706
-15863
-16111
-16219
-16658
-17367
-17382
-17388
-17584
-17705
-17858
-17996
-18440
-18839
-19022
-19324
-19378
-19459
-19467
-19475
-19531
-19543
-19612
-19633
-19634
-19640
-19666
-19701
-19722
-19780
-19807
-19836
-19982
-20425
-20430
-20433
-20447
-20448
-20451
-20577
-20624
-20661
-20878
-20930
-21463
-21487
-21520
-21533
-21575
-21593
-21692
-21811
-21864
-22191
-22195
-22209
-22216
-22447
-22617
-22700
-22718
-22720
-22758
-22948
-23059
-23357
-23415
-23460
-23471
-23479
-23483
-23498
-23501
-23506
-23517
-23524
-23615
-24121
-24178
-24241
-25189
-25444
-25691
-25970
-26117
-26379
-26440
-26531
-27037
-27873
-28995
-30674
-30682
-32767
-33573
-34496
-35035
-35128
-35214
-36123
-37616
-37885
-37886
-38014
-38348
-38972
-39029
-40054
-42861
-43518
-43727
-44175
-44388
-45668
-46366
-46700
-47791
-49203
-49378
-50205
-50859
-51993
-52118
-52276
-52443
-52461
-52539
-52571
-53484
-54161
-55357
-55376
-55585
-58690
-58714
-59806
-60070
-60217
-61200
-61839
-63596
-64306
-64719
-65841
-66221
-66620
-66669
-66865
-67195
-68639
-68808
-68959
-69099
-69215
-69337
-72196
-72238
-72802
-73555
-74922
-75248
-76084
-76487
-76817
-76893
-77930
-79266
-79520
-80210
-81184
-81294
-81644
-82214
-82518
-82537
-82604
-83578
-83794
-83804
-83883
-84330
-85286
-85639
-86281
-86453
-87624
-87732
-87804
-87822
-87933
-88638
-90076
-90481
-90665
-90683
-90945
-91665
-92617
-93512
-94615
-94806
-94813
-96676
-97060
-97291
-97582
-98335
-98557
-98825
-100249
-100411
-100434
-100922
-101125
-101527
-103116
-104391
-104793
-105808
-105975
-106681
-107229
-107334
-107552
-107934
-108060
-108098
-108106
-110030
-110065
-110563
-111079
-111667
-111671
-111957
-112351
-112544
-114246
-114249
-115279
-115933
-115938
-116006
-116122
-117380
-118649
-118675
-118864
-119478
-119561
-119682
-119808
-120075
-120176
-120750
-121706
-122975
-123612
-124509
-124814
-125632
-125889
-125949
-126100
-126676
-127339
-127504
-127655
-127728
-127753
-127857
-128152
-128476
-128738
-129016
-129211
-129244
-129469
-130186
-130221
-130852
-130991
-131039
-131466
-131763
-133157
-133548
-133661
-134168
-134206
-134752
-135454
-137001
-138509
-138980
-139058
-139922
-139967
-140086
-140103
-140186
-141180
-141629
-141664
-142655
-143450
-143592
-143709
-144382
-146718
-148280
-148723
-151351
-151759
-151778
-152285
-152453
-152585
-152990
-153834
-153966
-154283
-154448
-154788
-155021
-155296
-156949
-156954
-157620
-158390
-158963
-160769
-161380
-161660
-162673
-163005
-163016
-163596
-164120
-164152
-164330
-164925
-166595
-166660
-166683
-166768
-168438
-169072
-169133
-170444
-171462
-172071
-172473
-173397
-173564
-174306
-174802
-175695
-175912
-176612
-176740
-177250
-177265
-177696
-177885
-177903
-178320
-179224
-179337
-179437
-179444
-179481
-179994
-180430
-180452
-180709
-181259
-181270
-181292
-181421
-181904
-181910
-181915
-181925
-182069
-182260
-182474
-182518
-183209
-183462
-184125
-184260
-184413
-185253
-186629
-186898
-188504
-190084
-190127
-190328
-190481
-191207
-191444
-192882
-193144
-193519
-193641
-193671
-193683
-193772
-194206
-194249
-194554
-195315
-195553
-195858
-196421
-196853
-197746
-198738
-198744
-199692
-199929
-199976
-200003
-200223
-200250
-200828
-201263
-201276
-201645
-202167
-202214
-202741
-203324
-203419
-203819
-203844
-203901
-203975
-204085
-204116
-204119
-204429
-204701
-204797
-205005
-205831
-205922
-205974
-206001
-206545
-206805
-206842
-206882
-206922
-207164
-207174
-207302
-207424
-207444
-207829
-207833
-207847
-208021
-208149
-208815
-209067
-209093
-209274
-209310
-209508
-209808
-209899
-209906
-210013
-210131
-210146
-210152
-210341
-210348
-210351
-210361
-210629
-210896
-211021
-211271
-211363
-211501
-211517
-211718
-212002
-212097
-212120
-212369
-212458
-212482
-212503
-212541
-212596
-212810
-212881
-213122
-213338
-213601
-213738
-213743
-213861
-213999
-214063
-214189
-214252
-214423
-214623
-214682
-214720
-214945
-215080
-215163
-215306
-215351
-215370
-215425
-215516
-215821
-216035
-216072
-216340
-216397
-216485
-216620
-216864
-217136
-217183
-217196
-217240
-217493
-217687
-217697
-217718
-218026
-218191
-218271
-218341
-218363
-218563
-218654
-218830
-218862
-218879
-219076
-219241
-219339
-219369
-219578
-219724
-219733
-219770
-219788
-219870
-219893
-220572
-220576
-220596
-220728
-220956
-221143
-221203
-222033
-222043
-222257
-222320
-222448
-222906
-223072
-223109
-223219
-223526
-223544
-223605
-223700
-223755
-223796
-224043
-224265
-224268
-224280
-224304
-224352
-224454
-224666
-224709
-224720
-224741
-224820
-224962
-225127
-225265
-225311
-225363
-225492
-225536
-225724
-225750
-225818
-226026
-226065
-226124
-226359
-226570
-226618
-226930
-227005
-227108
-227555
-227779
-227881
-227892
-228031
-228267
-228841
-228981
-229028
-229114
-229186
-229393
-229443
-229478
-229633
-229722
-229777
-229812
-230022
-230214
-230345
-230351
-230869
-230888
-230937
-230987
-231147
-231352
-231636
-231820
-231850
-231852
-232242
-232543
-232765
-232921
-232931
-232977
-232989
-232990
-233200
-233478
-233545
-233762
-233822
-233881
-233921
-234026
-234041
-234061
-234161
-234333
-234483
-234640
-234787
-234826
-235204
-235214
-235543
-235785
-235823
-236025
-236061
-236070
-236081
-236102
-236168
-236252
-236282
-236406
-236681
-236810
-236855
-236919
-237247
-237476
-237737
-237850
-237896
-237939
-237960
-237961
-237969
-238043
-238306
-238393
-238476
-238790
-238840
-238878
-239022
-239061
-239062
-239217
-239277
-239282
-239473
-239666
-240313
-240505
-240607
-240617
-240649
-240703
-240769
-240791
-240864
-240911
-241137
-241288
-241462
-241610
-241661
-241825
-242047
-242066
-242459
-242499
-242882
-243138
-243145
-243168
-243343
-243382
-243408
-243574
-243579
-243587
-243613
-243810
-243910
-243974
-244036
-244047
-244267
-244308
-244530
-244666
-244741
-244793
-244861
-244865
-245024
-245067
-245148
-245157
-245351
-245458
-245565
-245926
-246001
-246390
-246703
-246732
-246793
-246805
-247038
-247132
-247365
-247431
-247808
-247911
-247953
-248261
-248347
-248351
-248362
-248508
-248607
-248734
-248797
-249480
-249588
-249750
-249781
-249807
-249964
-249973
-249982
-250122
-250256
-250373
-250449
-250466
-250597
-250664
-250715
-250885
-250981
-251018
-251230
-251240

test3.py

@@ -1,30 +0,0 @@
-import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-vicuna_model = AutoModelForCausalLM.from_pretrained("lmsys/vicuna-7b-v1.3")
-vicuna_tokenizer = AutoTokenizer.from_pretrained("lmsys/vicuna-7b-v1.3")
-
-# llama_model = AutoModelForCausalLM.from_pretrained("luodian/llama-7b-hf")
-# llama_tokenizer = AutoTokenizer.from_pretrained("luodian/llama-7b-hf")
-
-# Define the function for generating responses
-def generate_response(model, tokenizer, prompt):
-    inputs = tokenizer(prompt, return_tensors="pt")
-    outputs = model.generate(**inputs, max_length=500, pad_token_id=tokenizer.eos_token_id)
-    response = tokenizer.decode(outputs[0])
-    return response
-
-# Define the Gradio interface
-def chatbot_interface(prompt):
-    vicuna_response = generate_response(vicuna_model, vicuna_tokenizer, prompt)
-    # llama_response = generate_response(llama_model, llama_tokenizer, prompt)
-
-    return {"Vicuna-7B": vicuna_response}
-
-iface = gr.Interface(fn=chatbot_interface, 
-                     inputs="text", 
-                     outputs="text",
-                     interpretation="default",
-                     title="Chatbot with Three Models")
-
-iface.launch()