Hugging Face's logo

MichaelAI23
/
zephyr-7b-beta

English
Generated from Trainer
Inference Endpoints
Model card Files Community
Michael Brunzel commited on
Commit
8c88b7a
•
1 Parent(s): 170382f

Add files for the custom inference of the zephyr beta model

Browse files
Files changed (4) hide show
  1. README.md +241 -0
  2. createEndpoint.PNG +0 -0
  3. handler.py +58 -0
  4. requirements.txt +2 -0
README.md CHANGED
@@ -1,3 +1,244 @@
1
  ---
 
 
 
 
 
2
  license: mit
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ tags:
3
+ - generated_from_trainer
4
+ model-index:
5
+ - name: zephyr-7b-beta
6
+ results: []
7
  license: mit
8
+ datasets:
9
+ - HuggingFaceH4/ultrachat_200k
10
+ - HuggingFaceH4/ultrafeedback_binarized
11
+ language:
12
+ - en
13
+ base_model: mistralai/Mistral-7B-v0.1
14
  ---
15
+
16
+ # FORK of HuggingFaceH4/zephyr-7b-beta
17
+
18
+ > This is a fork of HuggingFaceH4/zephyr-7b-beta implementing a custom `handler.py` as an example for how to pass a system message to zephyr-7b-beta when deployed with inference-endpoints.
19
+
20
+ You can deploy the HuggingFaceH4/zephyr-7b-beta with a [1-click](https://ui.endpoints.huggingface.co/new?repository=MichaelAI23/zephyr-7b-beta).
21
+ We can use instance type to **"GPU [medium] · 1x Nvidia A10G"**.
22
+
23
+ ![createEndpoint](createEndpoint.png)
24
+
25
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
26
+ should probably proofread and complete it, then remove this comment. -->
27
+
28
+ <img src="https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha/resolve/main/thumbnail.png" alt="Zephyr Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
29
+
30
+
31
+ # Model Card for Zephyr 7B β
32
+
33
+ Zephyr is a series of language models that are trained to act as helpful assistants. Zephyr-7B-β is the second model in the series, and is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) that was trained on on a mix of publicly available, synthetic datasets using [Direct Preference Optimization (DPO)](https://arxiv.org/abs/2305.18290). We found that removing the in-built alignment of these datasets boosted performance on [MT Bench](https://huggingface.co/spaces/lmsys/mt-bench) and made the model more helpful. However, this means that model is likely to generate problematic text when prompted to do so and should only be used for educational and research purposes. You can find more details in the [technical report](https://arxiv.org/abs/2310.16944).
34
+
35
+
36
+ ## Model description
37
+
38
+ - **Model type:** A 7B parameter GPT-like model fine-tuned on a mix of publicly available, synthetic datasets.
39
+ - **Language(s) (NLP):** Primarily English
40
+ - **License:** MIT
41
+ - **Finetuned from model:** [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1)
42
+
43
+ ### Model Sources
44
+
45
+ <!-- Provide the basic links for the model. -->
46
+
47
+ - **Repository:** https://github.com/huggingface/alignment-handbook
48
+ - **Demo:** https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat
49
+ - **Chatbot Arena:** Evaluate Zephyr 7B against 10+ LLMs in the LMSYS arena: http://arena.lmsys.org
50
+
51
+ ## Performance
52
+
53
+ At the time of release, Zephyr-7B-β is the highest ranked 7B chat model on the [MT-Bench](https://huggingface.co/spaces/lmsys/mt-bench) and [AlpacaEval](https://tatsu-lab.github.io/alpaca_eval/) benchmarks:
54
+
55
+ | Model | Size | Alignment | MT-Bench (score) | AlpacaEval (win rate %) |
56
+ |-------------|-----|----|---------------|--------------|
57
+ | StableLM-Tuned-α | 7B| dSFT |2.75| -|
58
+ | MPT-Chat | 7B |dSFT |5.42| -|
59
+ | Xwin-LMv0.1 | 7B| dPPO| 6.19| 87.83|
60
+ | Mistral-Instructv0.1 | 7B| - | 6.84 |-|
61
+ | Zephyr-7b-α |7B| dDPO| 6.88| -|
62
+ | **Zephyr-7b-β** 🪁 | **7B** | **dDPO** | **7.34** | **90.60** |
63
+ | Falcon-Instruct | 40B |dSFT |5.17 |45.71|
64
+ | Guanaco | 65B | SFT |6.41| 71.80|
65
+ | Llama2-Chat | 70B |RLHF |6.86| 92.66|
66
+ | Vicuna v1.3 | 33B |dSFT |7.12 |88.99|
67
+ | WizardLM v1.0 | 70B |dSFT |7.71 |-|
68
+ | Xwin-LM v0.1 | 70B |dPPO |- |95.57|
69
+ | GPT-3.5-turbo | - |RLHF |7.94 |89.37|
70
+ | Claude 2 | - |RLHF |8.06| 91.36|
71
+ | GPT-4 | -| RLHF |8.99| 95.28|
72
+
73
+ In particular, on several categories of MT-Bench, Zephyr-7B-β has strong performance compared to larger open models like Llama2-Chat-70B:
74
+
75
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6200d0a443eb0913fa2df7cc/raxvt5ma16d7T23my34WC.png)
76
+
77
+ However, on more complex tasks like coding and mathematics, Zephyr-7B-β lags behind proprietary models and more research is needed to close the gap.
78
+
79
+
80
+ ## Intended uses & limitations
81
+
82
+ The model was initially fine-tuned on a filtered and preprocessed of the [`UltraChat`](https://huggingface.co/datasets/stingning/ultrachat) dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT.
83
+ We then further aligned the model with [🤗 TRL's](https://github.com/huggingface/trl) `DPOTrainer` on the [openbmb/UltraFeedback](https://huggingface.co/datasets/openbmb/UltraFeedback) dataset, which contains 64k prompts and model completions that are ranked by GPT-4. As a result, the model can be used for chat and you can check out our [demo](https://huggingface.co/spaces/HuggingFaceH4/zephyr-chat) to test its capabilities.
84
+
85
+ You can find the datasets used for training Zephyr-7B-β [here](https://huggingface.co/collections/HuggingFaceH4/zephyr-7b-6538c6d6d5ddd1cbb1744a66)
86
+
87
+ Here's how you can run the model using the `pipeline()` function from 🤗 Transformers:
88
+
89
+ ```python
90
+ # Install transformers from source - only needed for versions <= v4.34
91
+ # pip install git+https://github.com/huggingface/transformers.git
92
+ # pip install accelerate
93
+
94
+ import torch
95
+ from transformers import pipeline
96
+
97
+ pipe = pipeline("text-generation", model="HuggingFaceH4/zephyr-7b-beta", torch_dtype=torch.bfloat16, device_map="auto")
98
+
99
+ # We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
100
+ messages = [
101
+ {
102
+ "role": "system",
103
+ "content": "You are a friendly chatbot who always responds in the style of a pirate",
104
+ },
105
+ {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
106
+ ]
107
+ prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
108
+ outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
109
+ print(outputs[0]["generated_text"])
110
+ # <|system|>
111
+ # You are a friendly chatbot who always responds in the style of a pirate.</s>
112
+ # <|user|>
113
+ # How many helicopters can a human eat in one sitting?</s>
114
+ # <|assistant|>
115
+ # Ah, me hearty matey! But yer question be a puzzler! A human cannot eat a helicopter in one sitting, as helicopters are not edible. They be made of metal, plastic, and other materials, not food!
116
+ ```
117
+
118
+ ## Bias, Risks, and Limitations
119
+
120
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
121
+
122
+ Zephyr-7B-β has not been aligned to human preferences with techniques like RLHF or deployed with in-the-loop filtering of responses like ChatGPT, so the model can produce problematic outputs (especially when prompted to do so).
123
+ It is also unknown what the size and composition of the corpus was used to train the base model (`mistralai/Mistral-7B-v0.1`), however it is likely to have included a mix of Web data and technical sources like books and code. See the [Falcon 180B model card](https://huggingface.co/tiiuae/falcon-180B#training-data) for an example of this.
124
+
125
+
126
+ ## Training and evaluation data
127
+
128
+ During DPO training, this model achieves the following results on the evaluation set:
129
+
130
+ - Loss: 0.7496
131
+ - Rewards/chosen: -4.5221
132
+ - Rewards/rejected: -8.3184
133
+ - Rewards/accuracies: 0.7812
134
+ - Rewards/margins: 3.7963
135
+ - Logps/rejected: -340.1541
136
+ - Logps/chosen: -299.4561
137
+ - Logits/rejected: -2.3081
138
+ - Logits/chosen: -2.3531
139
+
140
+
141
+ ### Training hyperparameters
142
+
143
+ The following hyperparameters were used during training:
144
+ - learning_rate: 5e-07
145
+ - train_batch_size: 2
146
+ - eval_batch_size: 4
147
+ - seed: 42
148
+ - distributed_type: multi-GPU
149
+ - num_devices: 16
150
+ - total_train_batch_size: 32
151
+ - total_eval_batch_size: 64
152
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
153
+ - lr_scheduler_type: linear
154
+ - lr_scheduler_warmup_ratio: 0.1
155
+ - num_epochs: 3.0
156
+
157
+ ### Training results
158
+
159
+ The table below shows the full set of DPO training metrics:
160
+
161
+
162
+ | Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
163
+ |:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
164
+ | 0.6284 | 0.05 | 100 | 0.6098 | 0.0425 | -0.1872 | 0.7344 | 0.2297 | -258.8416 | -253.8099 | -2.7976 | -2.8234 |
165
+ | 0.4908 | 0.1 | 200 | 0.5426 | -0.0279 | -0.6842 | 0.75 | 0.6563 | -263.8124 | -254.5145 | -2.7719 | -2.7960 |
166
+ | 0.5264 | 0.15 | 300 | 0.5324 | 0.0414 | -0.9793 | 0.7656 | 1.0207 | -266.7627 | -253.8209 | -2.7892 | -2.8122 |
167
+ | 0.5536 | 0.21 | 400 | 0.4957 | -0.0185 | -1.5276 | 0.7969 | 1.5091 | -272.2460 | -254.4203 | -2.8542 | -2.8764 |
168
+ | 0.5362 | 0.26 | 500 | 0.5031 | -0.2630 | -1.5917 | 0.7812 | 1.3287 | -272.8869 | -256.8653 | -2.8702 | -2.8958 |
169
+ | 0.5966 | 0.31 | 600 | 0.5963 | -0.2993 | -1.6491 | 0.7812 | 1.3499 | -273.4614 | -257.2279 | -2.8778 | -2.8986 |
170
+ | 0.5014 | 0.36 | 700 | 0.5382 | -0.2859 | -1.4750 | 0.75 | 1.1891 | -271.7204 | -257.0942 | -2.7659 | -2.7869 |
171
+ | 0.5334 | 0.41 | 800 | 0.5677 | -0.4289 | -1.8968 | 0.7969 | 1.4679 | -275.9378 | -258.5242 | -2.7053 | -2.7265 |
172
+ | 0.5251 | 0.46 | 900 | 0.5772 | -0.2116 | -1.3107 | 0.7344 | 1.0991 | -270.0768 | -256.3507 | -2.8463 | -2.8662 |
173
+ | 0.5205 | 0.52 | 1000 | 0.5262 | -0.3792 | -1.8585 | 0.7188 | 1.4793 | -275.5552 | -258.0276 | -2.7893 | -2.7979 |
174
+ | 0.5094 | 0.57 | 1100 | 0.5433 | -0.6279 | -1.9368 | 0.7969 | 1.3089 | -276.3377 | -260.5136 | -2.7453 | -2.7536 |
175
+ | 0.5837 | 0.62 | 1200 | 0.5349 | -0.3780 | -1.9584 | 0.7656 | 1.5804 | -276.5542 | -258.0154 | -2.7643 | -2.7756 |
176
+ | 0.5214 | 0.67 | 1300 | 0.5732 | -1.0055 | -2.2306 | 0.7656 | 1.2251 | -279.2761 | -264.2903 | -2.6986 | -2.7113 |
177
+ | 0.6914 | 0.72 | 1400 | 0.5137 | -0.6912 | -2.1775 | 0.7969 | 1.4863 | -278.7448 | -261.1467 | -2.7166 | -2.7275 |
178
+ | 0.4655 | 0.77 | 1500 | 0.5090 | -0.7987 | -2.2930 | 0.7031 | 1.4943 | -279.8999 | -262.2220 | -2.6651 | -2.6838 |
179
+ | 0.5731 | 0.83 | 1600 | 0.5312 | -0.8253 | -2.3520 | 0.7812 | 1.5268 | -280.4902 | -262.4876 | -2.6543 | -2.6728 |
180
+ | 0.5233 | 0.88 | 1700 | 0.5206 | -0.4573 | -2.0951 | 0.7812 | 1.6377 | -277.9205 | -258.8084 | -2.6870 | -2.7097 |
181
+ | 0.5593 | 0.93 | 1800 | 0.5231 | -0.5508 | -2.2000 | 0.7969 | 1.6492 | -278.9703 | -259.7433 | -2.6221 | -2.6519 |
182
+ | 0.4967 | 0.98 | 1900 | 0.5290 | -0.5340 | -1.9570 | 0.8281 | 1.4230 | -276.5395 | -259.5749 | -2.6564 | -2.6878 |
183
+ | 0.0921 | 1.03 | 2000 | 0.5368 | -1.1376 | -3.1615 | 0.7812 | 2.0239 | -288.5854 | -265.6111 | -2.6040 | -2.6345 |
184
+ | 0.0733 | 1.08 | 2100 | 0.5453 | -1.1045 | -3.4451 | 0.7656 | 2.3406 | -291.4208 | -265.2799 | -2.6289 | -2.6595 |
185
+ | 0.0972 | 1.14 | 2200 | 0.5571 | -1.6915 | -3.9823 | 0.8125 | 2.2908 | -296.7934 | -271.1505 | -2.6471 | -2.6709 |
186
+ | 0.1058 | 1.19 | 2300 | 0.5789 | -1.0621 | -3.8941 | 0.7969 | 2.8319 | -295.9106 | -264.8563 | -2.5527 | -2.5798 |
187
+ | 0.2423 | 1.24 | 2400 | 0.5455 | -1.1963 | -3.5590 | 0.7812 | 2.3627 | -292.5599 | -266.1981 | -2.5414 | -2.5784 |
188
+ | 0.1177 | 1.29 | 2500 | 0.5889 | -1.8141 | -4.3942 | 0.7969 | 2.5801 | -300.9120 | -272.3761 | -2.4802 | -2.5189 |
189
+ | 0.1213 | 1.34 | 2600 | 0.5683 | -1.4608 | -3.8420 | 0.8125 | 2.3812 | -295.3901 | -268.8436 | -2.4774 | -2.5207 |
190
+ | 0.0889 | 1.39 | 2700 | 0.5890 | -1.6007 | -3.7337 | 0.7812 | 2.1330 | -294.3068 | -270.2423 | -2.4123 | -2.4522 |
191
+ | 0.0995 | 1.45 | 2800 | 0.6073 | -1.5519 | -3.8362 | 0.8281 | 2.2843 | -295.3315 | -269.7538 | -2.4685 | -2.5050 |
192
+ | 0.1145 | 1.5 | 2900 | 0.5790 | -1.7939 | -4.2876 | 0.8438 | 2.4937 | -299.8461 | -272.1744 | -2.4272 | -2.4674 |
193
+ | 0.0644 | 1.55 | 3000 | 0.5735 | -1.7285 | -4.2051 | 0.8125 | 2.4766 | -299.0209 | -271.5201 | -2.4193 | -2.4574 |
194
+ | 0.0798 | 1.6 | 3100 | 0.5537 | -1.7226 | -4.2850 | 0.8438 | 2.5624 | -299.8200 | -271.4610 | -2.5367 | -2.5696 |
195
+ | 0.1013 | 1.65 | 3200 | 0.5575 | -1.5715 | -3.9813 | 0.875 | 2.4098 | -296.7825 | -269.9498 | -2.4926 | -2.5267 |
196
+ | 0.1254 | 1.7 | 3300 | 0.5905 | -1.6412 | -4.4703 | 0.8594 | 2.8291 | -301.6730 | -270.6473 | -2.5017 | -2.5340 |
197
+ | 0.085 | 1.76 | 3400 | 0.6133 | -1.9159 | -4.6760 | 0.8438 | 2.7601 | -303.7296 | -273.3941 | -2.4614 | -2.4960 |
198
+ | 0.065 | 1.81 | 3500 | 0.6074 | -1.8237 | -4.3525 | 0.8594 | 2.5288 | -300.4951 | -272.4724 | -2.4597 | -2.5004 |
199
+ | 0.0755 | 1.86 | 3600 | 0.5836 | -1.9252 | -4.4005 | 0.8125 | 2.4753 | -300.9748 | -273.4872 | -2.4327 | -2.4716 |
200
+ | 0.0746 | 1.91 | 3700 | 0.5789 | -1.9280 | -4.4906 | 0.8125 | 2.5626 | -301.8762 | -273.5149 | -2.4686 | -2.5115 |
201
+ | 0.1348 | 1.96 | 3800 | 0.6015 | -1.8658 | -4.2428 | 0.8281 | 2.3769 | -299.3976 | -272.8936 | -2.4943 | -2.5393 |
202
+ | 0.0217 | 2.01 | 3900 | 0.6122 | -2.3335 | -4.9229 | 0.8281 | 2.5894 | -306.1988 | -277.5699 | -2.4841 | -2.5272 |
203
+ | 0.0219 | 2.07 | 4000 | 0.6522 | -2.9890 | -6.0164 | 0.8281 | 3.0274 | -317.1334 | -284.1248 | -2.4105 | -2.4545 |
204
+ | 0.0119 | 2.12 | 4100 | 0.6922 | -3.4777 | -6.6749 | 0.7969 | 3.1972 | -323.7187 | -289.0121 | -2.4272 | -2.4699 |
205
+ | 0.0153 | 2.17 | 4200 | 0.6993 | -3.2406 | -6.6775 | 0.7969 | 3.4369 | -323.7453 | -286.6413 | -2.4047 | -2.4465 |
206
+ | 0.011 | 2.22 | 4300 | 0.7178 | -3.7991 | -7.4397 | 0.7656 | 3.6406 | -331.3667 | -292.2260 | -2.3843 | -2.4290 |
207
+ | 0.0072 | 2.27 | 4400 | 0.6840 | -3.3269 | -6.8021 | 0.8125 | 3.4752 | -324.9908 | -287.5042 | -2.4095 | -2.4536 |
208
+ | 0.0197 | 2.32 | 4500 | 0.7013 | -3.6890 | -7.3014 | 0.8125 | 3.6124 | -329.9841 | -291.1250 | -2.4118 | -2.4543 |
209
+ | 0.0182 | 2.37 | 4600 | 0.7476 | -3.8994 | -7.5366 | 0.8281 | 3.6372 | -332.3356 | -293.2291 | -2.4163 | -2.4565 |
210
+ | 0.0125 | 2.43 | 4700 | 0.7199 | -4.0560 | -7.5765 | 0.8438 | 3.5204 | -332.7345 | -294.7952 | -2.3699 | -2.4100 |
211
+ | 0.0082 | 2.48 | 4800 | 0.7048 | -3.6613 | -7.1356 | 0.875 | 3.4743 | -328.3255 | -290.8477 | -2.3925 | -2.4303 |
212
+ | 0.0118 | 2.53 | 4900 | 0.6976 | -3.7908 | -7.3152 | 0.8125 | 3.5244 | -330.1224 | -292.1431 | -2.3633 | -2.4047 |
213
+ | 0.0118 | 2.58 | 5000 | 0.7198 | -3.9049 | -7.5557 | 0.8281 | 3.6508 | -332.5271 | -293.2844 | -2.3764 | -2.4194 |
214
+ | 0.006 | 2.63 | 5100 | 0.7506 | -4.2118 | -7.9149 | 0.8125 | 3.7032 | -336.1194 | -296.3530 | -2.3407 | -2.3860 |
215
+ | 0.0143 | 2.68 | 5200 | 0.7408 | -4.2433 | -7.9802 | 0.8125 | 3.7369 | -336.7721 | -296.6682 | -2.3509 | -2.3946 |
216
+ | 0.0057 | 2.74 | 5300 | 0.7552 | -4.3392 | -8.0831 | 0.7969 | 3.7439 | -337.8013 | -297.6275 | -2.3388 | -2.3842 |
217
+ | 0.0138 | 2.79 | 5400 | 0.7404 | -4.2395 | -7.9762 | 0.8125 | 3.7367 | -336.7322 | -296.6304 | -2.3286 | -2.3737 |
218
+ | 0.0079 | 2.84 | 5500 | 0.7525 | -4.4466 | -8.2196 | 0.7812 | 3.7731 | -339.1662 | -298.7007 | -2.3200 | -2.3641 |
219
+ | 0.0077 | 2.89 | 5600 | 0.7520 | -4.5586 | -8.3485 | 0.7969 | 3.7899 | -340.4545 | -299.8206 | -2.3078 | -2.3517 |
220
+ | 0.0094 | 2.94 | 5700 | 0.7527 | -4.5542 | -8.3509 | 0.7812 | 3.7967 | -340.4790 | -299.7773 | -2.3062 | -2.3510 |
221
+ | 0.0054 | 2.99 | 5800 | 0.7520 | -4.5169 | -8.3079 | 0.7812 | 3.7911 | -340.0493 | -299.4038 | -2.3081 | -2.3530 |
222
+
223
+
224
+ ### Framework versions
225
+
226
+ - Transformers 4.35.0.dev0
227
+ - Pytorch 2.0.1+cu118
228
+ - Datasets 2.12.0
229
+ - Tokenizers 0.14.0
230
+
231
+ ## Citation
232
+
233
+ If you find Zephyr-7B-β is useful in your work, please cite it with:
234
+
235
+ ```
236
+ @misc{tunstall2023zephyr,
237
+ title={Zephyr: Direct Distillation of LM Alignment},
238
+ author={Lewis Tunstall and Edward Beeching and Nathan Lambert and Nazneen Rajani and Kashif Rasul and Younes Belkada and Shengyi Huang and Leandro von Werra and Clémentine Fourrier and Nathan Habib and Nathan Sarrazin and Omar Sanseviero and Alexander M. Rush and Thomas Wolf},
239
+ year={2023},
240
+ eprint={2310.16944},
241
+ archivePrefix={arXiv},
242
+ primaryClass={cs.LG}
243
+ }
244
+ ```
createEndpoint.PNG ADDED
handler.py ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from typing import Dict, Any, List
2
+ from transformers import pipeline
3
+ import torch
4
+
5
+
6
+ class EndpointHandler:
7
+ def __init__(self, path=""):
8
+ # load model and processor from path
9
+ self.pipe = pipeline("text-generation", model="HuggingFaceH4/zephyr-7b-beta", torch_dtype=torch.bfloat16, device_map="auto")
10
+
11
+ def __call__(self, data: Dict[str, Any]) -> List[List[Dict[str, str]]]:
12
+ """
13
+ Args:
14
+ data (:dict:):
15
+ The payload with the text prompt and generation parameters.
16
+ """
17
+ # process input
18
+ inputs = data.pop("inputs", data)
19
+ parameters = data.pop("parameters", None)
20
+
21
+ if isinstance(inputs, list) and isinstance(inputs[0], list) or isinstance(inputs[0], dict):
22
+ if isinstance(inputs[0], dict):
23
+ inputs = [inputs]
24
+ messages = inputs
25
+
26
+ else:
27
+ if isinstance(inputs, str):
28
+ messages = [[
29
+ {
30
+ "role": "system",
31
+ "content": "You are a helpful AI assistant",
32
+ },
33
+ {"role": "user", "content": inputs},
34
+ ]]
35
+ else:
36
+ messages = [[
37
+ {
38
+ "role": "system",
39
+ "content": "You are a helpful AI assistant",
40
+ },
41
+ {"role": "user", "content": input},
42
+ ] for input in inputs]
43
+
44
+ prompts = []
45
+ for message in messages:
46
+ prompts += [self.pipe.tokenizer.apply_chat_template(message, tokenize=False, add_generation_prompt=True)]
47
+
48
+ # pass inputs with all kwargs in data
49
+ if parameters is not None:
50
+ outputs = self.pipe(
51
+ prompts,
52
+ **parameters)
53
+ else:
54
+ outputs = self.pipe(
55
+ prompts, max_new_tokens=32,
56
+ )
57
+
58
+ return [{"generated_text": outputs}]
requirements.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ git+https://github.com/huggingface/transformers.git@main
2
+ Jinja2