SillyTavern/SillyTavern-extras
1
0
Fork 0
mirror of https://github.com/SillyTavern/SillyTavern-Extras.git synced 2026-04-30 19:31:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

Removed textcap model option (didn't work in all modes)

Browse Source
This commit is contained in:
SillyLossy
2023-03-13 02:04:17 +02:00
parent e839112d02
commit 1756d869a2
3 changed files with 3 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
README.md
View File

@@ -239,7 +239,7 @@ File content
| `--cpu` | Run the models on the CPU instead of CUDA |
| `--summarization-model` | Load a custom summarization model.<br>Expects a HuggingFace model ID.<br>Default: [Qiliang/bart-large-cnn-samsum-ChatGPT_v3](https://huggingface.co/Qiliang/bart-large-cnn-samsum-ChatGPT_v3) |
| `--classification-model` | Load a custom sentiment classification model.<br>Expects a HuggingFace model ID.<br>Default (6 emotions): [bhadresh-savani/distilbert-base-uncased-emotion](https://huggingface.co/bhadresh-savani/distilbert-base-uncased-emotion)<br>Other solid option is (28 emotions): [joeddav/distilbert-base-uncased-go-emotions-student](https://huggingface.co/joeddav/distilbert-base-uncased-go-emotions-student) |
| `--captioning-model` | Load a custom captioning model.<br>Expects a HuggingFace model ID.<br>Default: [Salesforce/blip-image-captioning-large](https://huggingface.co/Salesforce/blip-image-captioning-large)<br>Other option (better for recognizing images with text on them, but worse for anime-style pictures): [microsoft/git-large-r-textcaps](https://huggingface.co/microsoft/git-large-r-textcaps) |
| `--captioning-model` | Load a custom captioning model.<br>Expects a HuggingFace model ID.<br>Default: [Salesforce/blip-image-captioning-large](https://huggingface.co/Salesforce/blip-image-captioning-large) |
| `--keyphrase-model` | Load a custom key phrase extraction model.<br>Expects a HuggingFace model ID.<br>Default: [ml6team/keyphrase-extraction-distilbert-inspec](https://huggingface.co/ml6team/keyphrase-extraction-distilbert-inspec) |
| `--prompt-model` | Load a custom prompt generation model.<br>Expects a HuggingFace model ID.<br>Default: [FredZhang7/anime-anything-promptgen-v2](https://huggingface.co/FredZhang7/anime-anything-promptgen-v2) |
| `--sd-model` | Load a custom Stable Diffusion image generation model.<br>Expects a HuggingFace model ID.<br>Default: [ckpt/anything-v4.5-vae-swapped](https://huggingface.co/ckpt/anything-v4.5-vae-swapped)<br>*Must have VAE pre-baked in PyTorch format or the output will look drab!* |

3
colab/GPU.ipynb
View File

@@ -70,10 +70,9 @@
"KoboldAI_Provider = \"Cloudflare\" #@param [\"Localtunnel\", \"Cloudflare\"]\n",
"use_google_drive = False #@param {type:\"boolean\"}\n",
"extras_enable_captioning = True #@param {type:\"boolean\"}\n",
"Captions_Model = \"Salesforce/blip-image-captioning-large\" #@param [ \"Salesforce/blip-image-captioning-large\", \"Salesforce/blip-image-captioning-base\", \"microsoft/git-large-r-textcaps\" ]\n",
"Captions_Model = \"Salesforce/blip-image-captioning-large\" #@param [ \"Salesforce/blip-image-captioning-large\", \"Salesforce/blip-image-captioning-base\" ]\n",
"#@markdown * Salesforce/blip-image-captioning-large - good base model\n",
"#@markdown * Salesforce/blip-image-captioning-base - slightly faster but less accurate\n",
"#@markdown * microsoft/git-large-r-textcaps - works best for pictures with text (memes), but slightly worse for cartoon/anime drawings\n",
"extras_enable_emotions = True #@param {type:\"boolean\"}\n",
"Emotions_Model = \"bhadresh-savani/distilbert-base-uncased-emotion\" #@param [\"bhadresh-savani/distilbert-base-uncased-emotion\", \"joeddav/distilbert-base-uncased-go-emotions-student\"]\n",
"#@markdown * bhadresh-savani/distilbert-base-uncased-emotion = 6 supported emotions<br>\n",

2
server.py
View File

@@ -26,7 +26,7 @@ colorama_init()
DEFAULT_SUMMARIZATION_MODEL = 'Qiliang/bart-large-cnn-samsum-ChatGPT_v3'
# Also try: 'joeddav/distilbert-base-uncased-go-emotions-student'
DEFAULT_CLASSIFICATION_MODEL = 'bhadresh-savani/distilbert-base-uncased-emotion'
# Also try: 'Salesforce/blip-image-captioning-base' or 'microsoft/git-large-r-textcaps'
# Also try: 'Salesforce/blip-image-captioning-base'
DEFAULT_CAPTIONING_MODEL = 'Salesforce/blip-image-captioning-large'
DEFAULT_KEYPHRASE_MODEL = 'ml6team/keyphrase-extraction-distilbert-inspec'
DEFAULT_PROMPT_MODEL = 'FredZhang7/anime-anything-promptgen-v2'