cocktailpeanut / fluxgym

Dead simple FLUX LoRA training UI with LOW VRAM support
1.25k stars 104 forks source link

UnicodeEncodeError: 'charmap' codec can't encode character '\xfc' in position 91: character maps to <undefined> #24

Open mikheys opened 2 months ago

mikheys commented 2 months ago

I get two errors when automatically signing images. If I delete the signatures, the train starts. `PowerShell 7.4.5 PS E:\AI\fluxgym> env\Scripts\activate (env) PS E:\AI\fluxgym> python app.py Running on local URL: http://127.0.0.1:7860

To create a public link, set share=True in launch(). launched max_train_epochs=16 num_images=10, num_repeats=10, total_steps=1600 max_train_epochs=16 num_images=9, num_repeats=10, total_steps=1440 run_captioning concept sentence tposter captions ('tposter', 'tposter', 'tposter', 'tposter', 'tposter', 'tposter', 'tposter', 'tposter', 'tposter', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '', '') device=cuda E:\AI\fluxgym\env\Lib\site-packages\transformers\tokenization_utils_base.py:1601: FutureWarning: clean_up_tokenization_spaces was not set. It will be set to True by default. This behavior will be depracted in transformers v4.45, and will be then set to False by default. For more details check this issue: https://github.com/huggingface/transformers/issues/31884 warnings.warn( tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[-2.1016, -2.1016, -2.1016, ..., 1.6836, 1.6836, 1.6836], [-2.1016, -2.1016, -2.1016, ..., 1.6836, 1.6836, 1.6836], [-2.0840, -2.0840, -2.0840, ..., 1.6836, 1.6836, 1.6836], ..., [ 1.6836, 1.6836, 1.6836, ..., 1.7012, 1.7012, 1.7012], [ 1.6836, 1.6836, 1.6836, ..., 1.7012, 1.7012, 1.7012], [ 1.6836, 1.6836, 1.6836, ..., 1.7012, 1.7012, 1.7012]],

     [[-2.0176, -2.0176, -2.0176,  ...,  1.8506,  1.8506,  1.8506],
      [-2.0176, -2.0176, -2.0176,  ...,  1.8506,  1.8506,  1.8506],
      [-2.0000, -2.0000, -2.0000,  ...,  1.8506,  1.8506,  1.8506],
      ...,
      [ 1.8506,  1.8506,  1.8506,  ...,  1.8682,  1.8682,  1.8682],
      [ 1.8506,  1.8506,  1.8506,  ...,  1.8682,  1.8682,  1.8682],
      [ 1.8506,  1.8506,  1.8506,  ...,  1.8682,  1.8682,  1.8682]],

     [[-1.7871, -1.7871, -1.7871,  ...,  2.0645,  2.0645,  2.0645],
      [-1.7871, -1.7871, -1.7871,  ...,  2.0645,  2.0645,  2.0645],
      [-1.7695, -1.7695, -1.7695,  ...,  2.0645,  2.0645,  2.0645],
      ...,
      [ 2.0645,  2.0645,  2.0645,  ...,  2.0820,  2.0820,  2.0820],
      [ 2.0645,  2.0645,  2.0645,  ...,  2.0820,  2.0820,  2.0820],
      [ 2.0645,  2.0645,  2.0645,  ...,  2.0820,  2.0820,  2.0820]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 11566, 13, 5, 8976, 35890, 2751, 2009, 954, 6, 4246, 10, 11577, 8, 14128, 1521, 19, 7457, 2788, 8, 1530, 4, 20, 11566, 16, 3820, 19, 10, 3143, 9, 8089, 6, 217, 15629, 6, 16543, 6, 28930, 2176, 6, 8, 37027, 6, 8, 5, 2788, 16, 1982, 11, 10, 2297, 28716, 4, 20, 1521, 16, 2295, 12, 24882, 8, 13869, 6, 442, 24, 41, 3571, 8, 3571, 515, 4, 2]], device='cuda:0') generated_text: The image shows a poster for the Yeah Indie Rock Club 2019, featuring a vibrant and colorful design with bold text and numbers. The poster is filled with a variety of colors, including blues, greens, yellows, and oranges, and the text is written in a modern font. The design is eye-catching and inviting, making it an exciting and exciting event. parsed_answer = {'': 'The image shows a poster for the Yeah Indie Rock Club 2019, featuring a vibrantand colorful design with bold text and numbers. The poster is filled with a variety of colors, including blues, greens,yellows, and oranges, and the text is written in a modern font. The design is eye-catching and inviting, making it an exciting and exciting event.'} caption_text = a poster for the Yeah Indie Rock Club 2019, featuring a vibrant and colorful design with bold text and numbers. The poster is filled with a variety of colors, including blues, greens, yellows, and oranges, and the text is written in a modern font. The design is eye-catching and inviting, making it an exciting and exciting event., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], ..., [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480]],

     [[2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      ...,
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277]],

     [[2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      ...,
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 11566, 19, 10, 1104, 3618, 8, 10, 1275, 516, 878, 149, 24, 4, 20, 2788, 15, 5, 11566, 7005, 22, 7605, 6508, 19, 840, 8807, 845, 2]], device='cuda:0') generated_text: The image shows a poster with a white background and a red line running through it. The text on the poster reads "From Poland with Shorts". parsed_answer = {'': 'The image shows a poster with a white background and a red line running throughit. The text on the poster reads "From Poland with Shorts".'} caption_text = a poster with a white background and a red line running through it. The text on the poster reads "From Poland with Shorts"., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], ..., [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480], [2.2480, 2.2480, 2.2480, ..., 2.2480, 2.2480, 2.2480]],

     [[2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      ...,
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277],
      [2.4277, 2.4277, 2.4277,  ..., 2.4277, 2.4277, 2.4277]],

     [[2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      ...,
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406],
      [2.6406, 2.6406, 2.6406,  ..., 2.6406, 2.6406, 2.6406]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 909, 8, 1104, 11566, 19, 10, 3428, 1521, 15, 24, 4, 20, 11566, 34, 2788, 1982, 15, 24, 6, 533, 9072, 5, 1521, 9, 5, 3428, 4, 20, 3428, 1521, 16, 2007, 648, 2295, 12, 24882, 6, 19, 10, 26183, 124, 7110, 8, 3124, 7110, 29, 6, 8, 237, 5856, 4, 20, 2788, 16, 1982, 11, 10, 7457, 28716, 6, 442, 24, 1413, 66, 136, 5, 1104, 3618, 4, 2]], device='cuda:0') generated_text: The image shows a black and white poster with a chair design on it. The poster has text written on it, likely describing the design of the chair. The chair design is simple yet eye-catching, with a curved backrest and armrests, and four legs. The text is written in a bold font, making it stand out against the white background. parsed_answer = {'': 'The image shows a black and white poster with a chair design on it. The poster has text written on it, likely describing the design of the chair. The chair design is simple yet eye-catching, with a curved backrest and armrests, and four legs. The text is written in a bold font, making it stand out against the white background.'} caption_text = a black and white poster with a chair design on it. The poster has text written on it, likely describingthe design of the chair. The chair design is simple yet eye-catching, with a curved backrest and armrests, and four legs. The text is written in a bold font, making it stand out against the white background., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[ 1.3242, 1.3066, 1.2725, ..., 1.0674, 1.1699, 1.2900], [ 0.6562, 0.6562, 0.6733, ..., 0.3652, 0.4680, 0.5879], [ 0.6221, 0.6392, 0.6562, ..., 0.4851, 0.6050, 0.7246], ..., [ 1.0332, 1.0156, 1.0156, ..., -1.7070, -1.4668, -1.0908], [ 1.0332, 1.0156, 1.0156, ..., -1.6553, -1.4502, -1.1074], [ 1.0156, 1.0156, 1.0156, ..., -1.6211, -1.4326, -1.1074]],

     [[ 1.4834,  1.4658,  1.4307,  ...,  1.2207,  1.3252,  1.4482],
      [ 0.8003,  0.8003,  0.8179,  ...,  0.5029,  0.6079,  0.7305],
      [ 0.7656,  0.7827,  0.8003,  ...,  0.6255,  0.7480,  0.8706],
      ...,
      [ 1.1855,  1.1680,  1.1680,  ..., -1.6152, -1.3701, -0.9854],
      [ 1.1855,  1.1680,  1.1680,  ..., -1.5635, -1.3525, -1.0029],
      [ 1.1680,  1.1680,  1.1680,  ..., -1.5283, -1.3350, -1.0029]],

     [[ 1.6992,  1.6816,  1.6465,  ...,  1.4375,  1.5420,  1.6641],
      [ 1.0195,  1.0195,  1.0361,  ...,  0.7227,  0.8271,  0.9492],
      [ 0.9844,  1.0020,  1.0195,  ...,  0.8447,  0.9668,  1.0889],
      ...,
      [ 1.4023,  1.3848,  1.3848,  ..., -1.3857, -1.1426, -0.7588],
      [ 1.4023,  1.3848,  1.3848,  ..., -1.3340, -1.1250, -0.7759],
      [ 1.3848,  1.3848,  1.3848,  ..., -1.2988, -1.1074, -0.7759]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 909, 8, 1104, 1345, 9, 10, 313, 11, 10, 3235, 8, 3318, 6, 1826, 10, 8669, 4, 20, 2788, 15, 5, 2274, 7005, 22, 39249, 7871, 111, 1308, 25455, 10039, 1464, 520, 38, 6871, 155, 845, 2]], device='cuda:0') generated_text: The image shows a black and white photo of a man in a suit and tie, holding a guitar. The text on the image reads "Johnny Cash - My Daddy Left Me When I Was 3". parsed_answer = {'': 'The image shows a black and white photo of a man in a suit and tie, holding a guitar. The text on the image reads "Johnny Cash - My Daddy Left Me When I Was 3".'} caption_text = a black and white photo of a man in a suit and tie, holding a guitar. The text on the image reads "Johnny Cash - My Daddy Left Me When I Was 3"., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[1.8203, 1.7861, 1.7354, ..., 1.7520, 1.7520, 1.7520], [1.8203, 1.7861, 1.7354, ..., 1.7520, 1.7520, 1.7520], [1.8203, 1.7861, 1.7354, ..., 1.7520, 1.7520, 1.7520], ..., [1.7861, 1.7861, 1.7520, ..., 1.7861, 1.7861, 1.7861], [1.7520, 1.7178, 1.6836, ..., 1.7178, 1.7178, 1.7178], [1.9238, 1.8896, 1.8555, ..., 1.8896, 1.8896, 1.8896]],

     [[1.6758, 1.6406, 1.5879,  ..., 1.6055, 1.6055, 1.6055],
      [1.6758, 1.6406, 1.5879,  ..., 1.6055, 1.6055, 1.6055],
      [1.6758, 1.6406, 1.5879,  ..., 1.6055, 1.6055, 1.6055],
      ...,
      [1.6406, 1.6406, 1.6055,  ..., 1.6406, 1.6406, 1.6406],
      [1.6055, 1.5703, 1.5361,  ..., 1.5703, 1.5703, 1.5703],
      [1.7812, 1.7461, 1.7109,  ..., 1.7461, 1.7461, 1.7461]],

     [[1.4727, 1.4375, 1.3848,  ..., 1.4023, 1.4023, 1.4023],
      [1.4727, 1.4375, 1.3848,  ..., 1.4023, 1.4023, 1.4023],
      [1.4727, 1.4375, 1.3848,  ..., 1.4023, 1.4023, 1.4023],
      ...,
      [1.4375, 1.4375, 1.4023,  ..., 1.4375, 1.4375, 1.4375],
      [1.4023, 1.3672, 1.3330,  ..., 1.3672, 1.3672, 1.3672],
      [1.5771, 1.5420, 1.5068,  ..., 1.5420, 1.5420, 1.5420]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 28, 7876, 11566, 19, 909, 2301, 8, 5, 1617, 22, 13562, 36645, 113, 1982, 15, 24, 6, 2351, 10, 5690, 5709, 227, 5, 80, 8089, 4, 2]], device='cuda:0') generated_text: The image shows a beige poster with black lines and the words "horizons" written on it, creatinga striking contrast between the two colors. parsed_answer = {'': 'The image shows a beige poster with black lines and the words "horizons" written on it, creating a striking contrast between the two colors.'} caption_text = a beige poster with black lines and the words "horizons" written on it, creating a striking contrast between the two colors., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[-1.2959, -1.2959, -1.2959, ..., -1.2783, -1.2783, -1.2783], [-1.2959, -1.2959, -1.2959, ..., -1.2783, -1.2783, -1.2783], [-1.2959, -1.2959, -1.2959, ..., -1.2783, -1.2783, -1.2783], ..., [-1.2783, -1.2783, -1.2783, ..., -1.2783, -1.2783, -1.2783], [-1.2783, -1.2783, -1.2783, ..., -1.2783, -1.2783, -1.2783], [-1.2783, -1.2783, -1.2783, ..., -1.2783, -1.2783, -1.2783]],

     [[-1.1250, -1.1250, -1.1250,  ..., -1.1074, -1.1074, -1.1074],
      [-1.1250, -1.1250, -1.1250,  ..., -1.1074, -1.1074, -1.1074],
      [-1.1250, -1.1250, -1.1250,  ..., -1.1074, -1.1074, -1.1074],
      ...,
      [-1.1074, -1.1074, -1.1074,  ..., -1.1074, -1.1074, -1.1074],
      [-1.1074, -1.1074, -1.1074,  ..., -1.1074, -1.1074, -1.1074],
      [-1.1074, -1.1074, -1.1074,  ..., -1.1074, -1.1074, -1.1074]],

     [[ 0.3916,  0.3916,  0.3916,  ...,  0.4092,  0.4092,  0.4092],
      [ 0.3916,  0.3916,  0.3916,  ...,  0.4092,  0.4092,  0.4092],
      [ 0.3916,  0.3916,  0.3916,  ...,  0.4092,  0.4092,  0.4092],
      ...,
      [ 0.4092,  0.4092,  0.4092,  ...,  0.4092,  0.4092,  0.4092],
      [ 0.4092,  0.4092,  0.4092,  ...,  0.4092,  0.4092,  0.4092],
      [ 0.4092,  0.4092,  0.4092,  ...,  0.4092,  0.4092,  0.4092]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 2440, 11566, 19, 5718, 5430, 24684, 66, 5, 2136, 22, 90, 1242, 113, 15, 24, 6, 1412, 30, 221, 4, 104, 4, 2954, 27980, 4, 20, 11566, 16, 14092, 9, 10, 2440, 3618, 19, 1104, 2788, 6, 2351, 10, 5690, 5709, 4, 2]], device='cuda:0') generated_text: The image shows a blue poster with yellow letters spelling out the word "tot" on it, created by P.S. Erickson. The poster is composed of a blue background with white text, creating a striking contrast. parsed_answer = {'': 'The image shows a blue poster with yellow letters spelling out the word "tot" on it, created by P.S. Erickson. The poster is composed of a blue background with white text, creating a striking contrast.'} caption_text = a blue poster with yellow letters spelling out the word "tot" on it, created by P.S. Erickson. The poster is composed of a blue background with white text, creating a striking contrast., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[1.6152, 1.6152, 1.6152, ..., 1.5811, 1.5811, 1.5811], [1.6152, 1.6152, 1.6152, ..., 1.6152, 1.6152, 1.6152], [1.6152, 1.6152, 1.6152, ..., 1.6494, 1.6494, 1.6494], ..., [1.6152, 1.6328, 1.6328, ..., 1.6494, 1.6494, 1.6494], [1.6152, 1.6328, 1.6328, ..., 1.6494, 1.6494, 1.6494], [1.6152, 1.6328, 1.6328, ..., 1.6494, 1.6494, 1.6494]],

     [[1.7637, 1.7637, 1.7637,  ..., 1.7285, 1.7285, 1.7285],
      [1.7637, 1.7637, 1.7637,  ..., 1.7637, 1.7637, 1.7637],
      [1.7637, 1.7637, 1.7637,  ..., 1.7979, 1.7979, 1.7979],
      ...,
      [1.7637, 1.7812, 1.7812,  ..., 1.7979, 1.7979, 1.7979],
      [1.7637, 1.7812, 1.7812,  ..., 1.7979, 1.7979, 1.7979],
      [1.7637, 1.7812, 1.7812,  ..., 1.7979, 1.7979, 1.7979]],

     [[1.8906, 1.8906, 1.8906,  ..., 1.8555, 1.8555, 1.8555],
      [1.8906, 1.8906, 1.8906,  ..., 1.8906, 1.8906, 1.8906],
      [1.8906, 1.8906, 1.8906,  ..., 1.9258, 1.9258, 1.9258],
      ...,
      [1.8906, 1.9082, 1.9082,  ..., 1.9258, 1.9258, 1.9258],
      [1.8906, 1.9082, 1.9082,  ..., 1.9258, 1.9258, 1.9258],
      [1.8906, 1.9082, 1.9082,  ..., 1.9258, 1.9258, 1.9258]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 11566, 19, 5718, 8, 909, 2788, 15, 10, 1104, 3618, 4, 20, 2788, 7005, 22, 34091, 856, 5172, 7438, 113, 61, 19303, 7, 22, 34091, 13, 7438, 113, 11, 2370, 4, 2]], device='cuda:0') generated_text: The image shows a poster with yellow and black text on a white background. The text reads "Design für Design" which translates to "Design for Design" in English. parsed_answer = {'': 'The image shows a poster with yellow and black text on a white background. The text reads "Design für Design" which translates to "Design for Design" in English.'} caption_text = a poster with yellow and black text on a white background. The text reads "Design für Design" which translates to "Design for Design" in English., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[ 2.2148, 2.2148, 2.2148, ..., 2.2148, 2.2148, 2.2148], [ 2.2148, 2.2148, 2.2148, ..., 2.2148, 2.2148, 2.2148], [ 2.2148, 2.2148, 2.2148, ..., 2.2148, 2.2148, 2.2148], ..., [ 2.2148, 2.2148, 2.2148, ..., 2.2148, 2.2148, 2.2148], [ 2.1973, 2.1973, 2.1973, ..., 2.1973, 2.1973, 2.1973], [ 2.1797, 2.1797, 2.1797, ..., 2.1797, 2.1797, 2.1797]],

     [[ 1.6758,  1.6758,  1.6758,  ...,  1.6758,  1.6758,  1.6758],
      [ 1.6758,  1.6758,  1.6758,  ...,  1.6758,  1.6758,  1.6758],
      [ 1.6758,  1.6758,  1.6758,  ...,  1.6758,  1.6758,  1.6758],
      ...,
      [ 1.6406,  1.6406,  1.6406,  ...,  1.6406,  1.6406,  1.6406],
      [ 1.7109,  1.7109,  1.7109,  ...,  1.7109,  1.7109,  1.7109],
      [ 1.8154,  1.8154,  1.8154,  ...,  1.8154,  1.8154,  1.8154]],

     [[-1.8047, -1.8047, -1.8047,  ..., -1.8047, -1.8047, -1.8047],
      [-1.8047, -1.8047, -1.8047,  ..., -1.8047, -1.8047, -1.8047],
      [-1.8047, -1.8047, -1.8047,  ..., -1.8047, -1.8047, -1.8047],
      ...,
      [-1.7520, -1.7520, -1.7520,  ..., -1.7520, -1.7520, -1.7520],
      [-1.7871, -1.7871, -1.7871,  ..., -1.7871, -1.7871, -1.7871],
      [-1.1074, -1.1074, -1.1074,  ..., -1.1074, -1.1074, -1.1074]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 5718, 11566, 19, 5, 2136, 22, 20983, 113, 1982, 11, 7457, 909, 2301, 6, 2351, 10, 5690, 5709, 136, 5, 4520, 5718, 3618, 4, 2]], device='cuda:0') generated_text: The image shows a yellow poster with the word "Manchester" written in bold black lines, creatinga striking contrast against the bright yellow background. parsed_answer = {'': 'The image shows a yellow poster with the word "Manchester" written in bold black lines, creating a striking contrast against the bright yellow background.'} caption_text = a yellow poster with the word "Manchester" written in bold black lines, creating a striking contrast against the bright yellow background., concept_sentence=tposter tposter inputs {'input_ids': tensor([[ 0, 47066, 21700, 11, 4617, 99, 16, 2343, 11, 5, 2274, 4, 2]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0'), 'pixel_values': tensor([[[[-2.0840, -2.0840, -2.0840, ..., -2.0840, -2.0840, -2.0840], [-2.0840, -2.0840, -2.0840, ..., -2.0840, -2.0840, -2.0840], [-2.0840, -2.0840, -2.0840, ..., -2.0840, -2.0840, -2.0840], ..., [-2.1172, -2.1172, -2.1172, ..., -2.1172, -2.1172, -2.1172], [-2.1172, -2.1172, -2.1172, ..., -2.1172, -2.1172, -2.1172], [-2.1172, -2.1172, -2.1172, ..., -2.1172, -2.1172, -2.1172]],

     [[-2.0000, -2.0000, -2.0000,  ..., -2.0000, -2.0000, -2.0000],
      [-2.0000, -2.0000, -2.0000,  ..., -2.0000, -2.0000, -2.0000],
      [-2.0000, -2.0000, -2.0000,  ..., -2.0000, -2.0000, -2.0000],
      ...,
      [-2.0352, -2.0352, -2.0352,  ..., -2.0352, -2.0352, -2.0352],
      [-2.0352, -2.0352, -2.0352,  ..., -2.0352, -2.0352, -2.0352],
      [-2.0352, -2.0352, -2.0352,  ..., -2.0352, -2.0352, -2.0352]],

     [[-1.7695, -1.7695, -1.7695,  ..., -1.7695, -1.7695, -1.7695],
      [-1.7695, -1.7695, -1.7695,  ..., -1.7695, -1.7695, -1.7695],
      [-1.7695, -1.7695, -1.7695,  ..., -1.7695, -1.7695, -1.7695],
      ...,
      [-1.8047, -1.8047, -1.8047,  ..., -1.8047, -1.8047, -1.8047],
      [-1.8047, -1.8047, -1.8047,  ..., -1.8047, -1.8047, -1.8047],
      [-1.8047, -1.8047, -1.8047,  ..., -1.8047, -1.8047, -1.8047]]]],
   device='cuda:0', dtype=torch.float16)}

generated_ids tensor([[ 2, 0, 133, 2274, 924, 10, 664, 1816, 2498, 10, 909, 8443, 19, 69, 1420, 15, 69, 471, 6, 2934, 11, 760, 9, 10, 909, 3618, 4, 20, 2788, 15, 5, 2274, 7005, 22, 510, 5037, 1776, 20796, 113, 8, 89, 16, 10, 514, 6920, 11, 5, 2576, 235, 2797, 4, 2]], device='cuda:0') generated_text: The image shows a young girl wearing a black jacket with her hands on her head, standing in front of a black background. The text on the image reads "Puma Big Shot" and there is a watermark in the bottom right corner. parsed_answer = {'': 'The image shows a young girl wearing a black jacket with her hands on her head,standing in front of a black background. The text on the image reads "Puma Big Shot" and there is a watermark in the bottom right corner.'} caption_text = a young girl wearing a black jacket with her hands on her head, standing in front of a black background.The text on the image reads "Puma Big Shot" and there is a watermark in the bottom right corner., concept_sentence=tposter Creating dataset resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\4d99ab745b2c9d8f3ee2e46714f08e7a.jpg : 512x728 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\4d99ab745b2c9d8f3ee2e46714f08e7a.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\4d99ab745b2c9d8f3ee2e46714f08e7a.txt, original_caption=tposter a poster for the Yeah Indie Rock Club 2019, featuring a vibrant and colorful design with bold text and numbers. The poster is filled with a variety of colors, including blues, greens, yellows, and oranges, and the text is written in a modern font. The design is eye-catching and inviting, making it an exciting and exciting event. resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\29ddc6f7e495f9f9ede91061409ca262.jpg : 512x731 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\29ddc6f7e495f9f9ede91061409ca262.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\29ddc6f7e495f9f9ede91061409ca262.txt, original_caption=tposter a poster with a white background and a red line running through it. The text on the poster reads "From Poland with Shorts". resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\1518f5d6af8cec2bdacbaf1539abba2d.jpg : 512x716 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\1518f5d6af8cec2bdacbaf1539abba2d.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\1518f5d6af8cec2bdacbaf1539abba2d.txt, original_caption=tposter a black and white poster with a chair design on it. The poster has text written on it, likely describing the design of the chair. The chair design is simple yet eye-catching, with a curved backrest and armrests, and four legs. The text is written in a bold font, making it stand out against the white background. resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\2017f76a1468b23309e0bc92b26ec9c0.jpg : 512x723 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\2017f76a1468b23309e0bc92b26ec9c0.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\2017f76a1468b23309e0bc92b26ec9c0.txt, original_caption=tposter a black and white photo of a man in a suit and tie, holding a guitar. The text on the image reads "Johnny Cash - My Daddy Left Me When I Was3". resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\b6fccdd666fbb328b482213b8c78e1d8.jpg : 512x728 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\b6fccdd666fbb328b482213b8c78e1d8.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\b6fccdd666fbb328b482213b8c78e1d8.txt, original_caption=tposter a beige poster with black lines and the words "horizons" written on it, creating a striking contrast between the two colors. resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\c903733af46cbff048c542d351d71002.jpg : 512x724 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\c903733af46cbff048c542d351d71002.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\c903733af46cbff048c542d351d71002.txt, original_caption=tposter a blue poster with yellow letters spelling out the word "tot" on it, created by P.S. Erickson. The poster is composed of a blue background with white text, creating a striking contrast. resize datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\d6ebf60034e5d022f86c83ad8136f8a3.jpg : 512x657 image_path=datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\d6ebf60034e5d022f86c83ad8136f8a3.jpg, caption_path = datasets/4cc4f6ba-78bc-49f8-9d17-ef34ba9040fd\d6ebf60034e5d022f86c83ad8136f8a3.txt, original_caption=tposter a poster with yellowand black text on a white background. The text reads "Design für Design" which translates to "Design for Design" in English. Traceback (most recent call last): File "E:\AI\fluxgym\env\Lib\site-packages\gradio\queueing.py", line 536, in process_events response = await route_utils.call_process_api( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\route_utils.py", line 321, in call_process_api output = await app.get_blocks().process_api( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\blocks.py", line 1935, in process_api result = await self.call_function( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\blocks.py", line 1520, in call_function prediction = await anyio.to_thread.run_sync( # type: ignore ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\anyio\to_thread.py", line 56, in run_sync return await get_async_backend().run_sync_in_worker_thread( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\anyio_backends_asyncio.py", line 2177, in run_sync_in_worker_thread return await future ^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\anyio_backends_asyncio.py", line 859, in run result = context.run(func, args) ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\utils.py", line 826, in wrapper response = f(args, *kwargs) ^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\app.py", line 105, in create_dataset file.write(original_caption) File "C:\Python311\Lib\encodings\cp1251.py", line 19, in encode return codecs.charmap_encode(input,self.errors,encoding_table)[0] ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ UnicodeEncodeError: 'charmap' codec can't encode character '\xfc' in position 91: character maps to gen_sh: network_dim:4, max_train_epochs=16, save_every_n_epochs=4, timestep_sampling=shift, guidance_scale=1, vram=20G, Traceback (most recent call last): File "E:\AI\fluxgym\env\Lib\site-packages\gradio\queueing.py", line 536, in process_events response = await route_utils.call_process_api( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\route_utils.py", line 321, in call_process_api output = await app.get_blocks().process_api( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\blocks.py", line 1935, in process_api result = await self.call_function( ^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\blocks.py", line 1532, in call_function prediction = await utils.async_iteration(iterator) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\utils.py", line 671, in async_iteration return await iterator.anext() ^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\utils.py", line 664, in anext return await anyio.to_thread.run_sync( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\anyio\to_thread.py", line 56, in run_sync return await get_async_backend().run_sync_in_worker_thread( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\anyio_backends_asyncio.py", line 2177, in run_sync_in_worker_thread return await future ^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\anyio_backends_asyncio.py", line 859, in run result = context.run(func, args) ^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\utils.py", line 647, in run_sync_iterator_async return next(iterator) ^^^^^^^^^^^^^^ File "E:\AI\fluxgym\env\Lib\site-packages\gradio\utils.py", line 809, in gen_wrapper response = next(iterator) ^^^^^^^^^^^^^^ File "E:\AI\fluxgym\app.py", line 341, in start_training gen_toml(dataset_folder, resolution, class_tokens, num_repeats) File "E:\AI\fluxgym\app.py", line 284, in gen_toml image_dir = '{resolve_path_without_quotes(dataset_folder)}' ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "E:\AI\fluxgym\app.py", line 173, in resolve_path_without_quotes norm_path = os.path.normpath(os.path.join(current_dir, p)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "", line 147, in join File "", line 152, in _check_arg_types TypeError: join() argument must be str, bytes, or os.PathLike object, not 'NoneType' `

mikheys commented 2 months ago

Figured it out. It was all about a character that is not supported)

BrokenEnigma commented 1 month ago

Figured it out. It was all about a character that is not supported)

What was the characted as I am getting the same error?