Hugging Face's logo

Spaces:
rynmurdock
/
Babel
Runtime error

App Files Community
Babel
File size: 7,452 Bytes 
c5ca37a
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
 
export PYTHONPATH="${PYTHONPATH}:/workspace/code"

# export TRAIN_FILE=../data/datasets/penn/train.txt
# export TEST_FILE=../data/datasets/penn/test.txt


# export TRAIN_FILE=../data/datasets/wikitext-2/train.txt
# export TEST_FILE=../data/datasets/wikitext-2/valid.txt
# export GPU_ID=0,1

# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_encoding_generation.py \
#     --checkpoint_dir=../output/philly_clm_wiki2_0.0 \
#     --output_dir=../output/philly_clm_wiki2_0.0 \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-uncased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1



# export TRAIN_FILE=../data/datasets/debug_data/train.txt
# export TEST_FILE=../data/datasets/debug_data/test.txt
# export GPU_ID=0,1

# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_encoding_generation.py \
#     --checkpoint_dir=../output/local_lm_vae_debug_bert_gpt \
#     --output_dir=../output/local_lm_vae_debug_bert_gpt \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-uncased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 400


export TRAIN_FILE=../data/datasets/debug_data/train.txt
export TEST_FILE=../data/datasets/debug_data/test.txt
export GPU_ID=1


# # interpolation from pre-trained model on wiki
# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_latent_generation.py \
#     --dataset Debug \
#     --checkpoint_dir=../output/pretrain/philly_rr3_vc4_g8_base_vae_wikipedia_pretraining_beta_schedule_beta1.0_d1.0_ro0.5_ra0.25_768_v2 \
#     --output_dir=../output/pretrain/philly_rr3_vc4_g8_base_vae_wikipedia_pretraining_beta_schedule_beta1.0_d1.0_ro0.5_ra0.25_768_v2 \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-cased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --train_data_file=$TRAIN_FILE \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 508523 \
#     --block_size 100 \
#     --max_seq_length 100 \
#     --latent_size 768 \
#     --play_mode interpolation \
#     --num_interpolation_steps 10


# # reconstruction from pre-trained model on wiki
# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_latent_generation.py \
#     --dataset Debug \
#     --checkpoint_dir=../output/pretrain/philly_rr3_vc4_g8_base_vae_wikipedia_pretraining_beta_schedule_beta0.0_d1.0_ro0.5_ra0.25_32_v2 \
#     --output_dir=../output/pretrain/philly_rr3_vc4_g8_base_vae_wikipedia_pretraining_beta_schedule_beta0.0_d1.0_ro0.5_ra0.25_32_v2 \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-cased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --train_data_file=$TRAIN_FILE \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 400000 \
#     --block_size 100 \
#     --max_seq_length 100 \
#     --latent_size 32 \
#     --play_mode reconstrction



# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_latent_generation.py \
#     --dataset Debug \
#     --checkpoint_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
#     --output_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-cased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --train_data_file=$TRAIN_FILE \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 31250 \
#     --block_size 100 \
#     --max_seq_length 100 \
#     --latent_size 768 \
#     --play_mode interpolation \
#     --num_interpolation_steps 10

# reconstrction
# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_latent_generation.py \
#     --dataset Debug \
#     --checkpoint_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
#     --output_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-cased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --train_data_file=$TRAIN_FILE \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 31250 \
#     --block_size 100 \
#     --max_seq_length 100 \
#     --latent_size 768 \
#     --play_mode reconstrction


# interact_with_user_input
CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_latent_generation.py \
    --dataset Debug \
    --checkpoint_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
    --output_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
    --encoder_model_type=bert \
    --encoder_model_name_or_path=bert-base-cased \
    --decoder_model_type=gpt2 \
    --decoder_model_name_or_path=gpt2 \
    --train_data_file=$TRAIN_FILE \
    --eval_data_file=$TEST_FILE \
    --per_gpu_eval_batch_size=1 \
    --gloabl_step_eval 31250 \
    --block_size 100 \
    --max_seq_length 100 \
    --latent_size 768 \
    --interact_with_user_input \
    --play_mode analogy \
    --sent_source="a yellow cat likes to chase a long string ." \
    --sent_target="a yellow cat likes to chase a short string ." \
    --sent_input="a brown dog likes to eat long pasta ." \
    --degree_to_target=1.0
        


# interact_with_user_input
# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_latent_generation.py \
#     --dataset Debug \
#     --checkpoint_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
#     --output_dir=../output/LM/Snli/768/philly_vae_snli_b1.0_d5_r00.5_ra0.25_length_weighted/checkpoint-31250 \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-cased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --train_data_file=$TRAIN_FILE \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 31250 \
#     --block_size 100 \
#     --max_seq_length 100 \
#     --latent_size 768 \
#     --interact_with_user_input \
#     --play_mode interpolation \
#     --sent_source="a yellow cat likes to chase a short string ." \
#     --sent_target="a brown dog likes to eat his food very slowly ." \
#     --num_interpolation_steps=10


# export TRAIN_FILE=../data/datasets/debug_data/train.txt
# export TEST_FILE=../data/datasets/debug_data/test.txt
# export GPU_ID=1

# CUDA_VISIBLE_DEVICES=$GPU_ID python examples/big_ae/run_encoding_generation.py \
#     --dataset Debug \
#     --checkpoint_dir=../output/local_lm_vae_debug_bert_gpt \
#     --output_dir=../output/local_lm_vae_debug_bert_gpt \
#     --encoder_model_type=bert \
#     --encoder_model_name_or_path=bert-base-uncased \
#     --decoder_model_type=gpt2 \
#     --decoder_model_name_or_path=gpt2 \
#     --train_data_file=$TRAIN_FILE \
#     --eval_data_file=$TEST_FILE \
#     --per_gpu_eval_batch_size=1 \
#     --gloabl_step_eval 800 \
#     --total_sents 10