llsourcell / ai_startup_prototype Goto Github PK

View Code? Open in Web Editor NEW

368.0 368.0 264.0 1.4 MB

This is the code for "Watch Me Build an AI Startup" By Siraj Raval on Youtube

Jupyter Notebook 26.60% Python 43.67% CSS 1.19% HTML 27.39% Dockerfile 0.78% Makefile 0.37%

ai_startup_prototype's People

Contributors

Stargazers

Watchers

Forkers

crosstuck dannyfiker deven96 nightier unoqualsiasi lxrd-aj yattishr asirem16 avanderreest machinelearningrifche theafricanquant lp-dataninja mllog avinregmi daytradecentral stjordanis navanil018 jvkassi bsirmacek iamvazu znaixian justinmonsi jshuadvd ankurkillerboy elephantsofneptune sakthik26 ogwang rsingh2083 palatinuse binilj04 flamato asr-engineering-consulting randerson112358 anishbatra7 bastinrobin arvi-ap vikasgupta-github armandds naushad-rahman someshjaishwal docstar1 cyuancheng spiritualism cr7yash shashwatwork jhird dnzengou michaelbroox adityakhannaadk hemanthtech serviolimareina iamfredrickokereke johntekdek roycezjq chandangeek nsrinivasapps jcsolis96 maciulaitis gary7 daggieblanqx subratasarkar32 roy010 ericohenrique muzilli fastlater ensindon2005 rodolfoohl chrimerss adolfoeliazat ithinkmfallin nanaakwasiabayieboateng sujanay lourdesolmos momor666 mrunalthakur guitarplayer82 numantics techtree tpadi goldbellco scaiado madasepandri mohdamanzaidansari goyalyash1224 fsuplicy ricardo-0x07 ryanbrwr darpan9o1 shido6 sgallen mrthundergod ia1na09 saisriteja parthnagarkar875 damonclifford bhardwajrahul kmlebedev lawanfalalu terrywalker866 federicosan

ai_startup_prototype's Issues

ValueError: You are trying to load a weight file containing 58 layers into a model with 107 layers.

Requirement.txt Needed

Please Freeze the project - (pipdeptree)

importError: cannot import name 'quote'

sing TensorFlow backend.
Traceback (most recent call last):
File "manage.py", line 4, in
from app import app, db, models
File "/home/harish/Desktop/my startup ideas and stuff/AI_Startup_Prototype-master (1)/AI_Startup_Prototype-master/flaskSaaS-master/app/init.py", line 46, in
from app import admin
File "/home/harish/Desktop/my startup ideas and stuff/AI_Startup_Prototype-master (1)/AI_Startup_Prototype-master/flaskSaaS-master/app/admin.py", line 7, in
from flask.ext.admin.contrib.fileadmin import FileAdmin
File "/home/harish/.local/lib/python3.6/site-packages/flask_admin/contrib/fileadmin/init.py", line 15, in
from flask_admin._compat import urljoin, as_unicode, quote
ImportError: cannot import name 'quote'

tenserflow has no attribute

BERT_MODEL not defined

Can anyone explain me this?

Tensorflow has no module

I am getting this error.
graph = tf.get_default_graph()
AttributeError: module 'tensorflow' has no attribute 'get_default_graph'

wrongly classifiying?

The result is hard coded?

Webapp using imagenet weights for prediction instead of the new trained weights

The main.py is using the imagenet weights instead if the new trained model weights. In the function, upload file is not importing the model with the updated weights.

Look at the model being used -
model= ResNet50(weights='imagenet')

@app.route('/uploaded', methods = ['GET', 'POST']) def upload_file(): if request.method == 'POST': f = request.files['file'] path = os.path.join(app.config['UPLOAD_FOLDER'], f.filename) model= ResNet50(weights='imagenet') img = image.load_img(path, target_size=(224,224)) x = image.img_to_array(img) x = np.expand_dims(x, axis=0) x = preprocess_input(x) preds = model.predict(x) preds_decoded = decode_predictions(preds, top=3)[0] print(decode_predictions(preds, top=3)[0]) f.save(path) return render_template('uploaded.html', title='Success', predictions=preds_decoded, user_image=f.filename)

Getting 404 Issue when trying to upload img

When trying to load an image up on the website that has been created i get a 404 not found error anyone know what i may have done wrong?

Website: Stripe won't redirect

After I 'pay', stripe accepts, but won't redirect me to the upload page.

NameError: name 'BERT_MODEL' is not defined

In the jupyter notebook BERT_MODEL is not defined. Can someone help me defining BERT_MODEL

ValueError: You are trying to load a weight file containing 58 layers into a model with 107 layers. [from keras.applications.resnet50 import ResNet50 model = ResNet50(weights= 'Puenomonia.h5')]

Trained Model is Classification But result shown is more of segmentation.

Hi,

Thank you for your efforts, but I want to point out that your model output is classification i.e. 3 outcomes but the image you showed is more of a segmentation task that requires something like mask-rcnn sort of thing and similarly the data according to that which can be found here:

https://www.kaggle.com/c/rsna-pneumonia-detection-challenge/data

The Result is hardcorded!

return render_template('uploaded.html', title='Success', predictions=preds_decoded, user_image=f.filename), What if we can return response in json? that can be much better.